Auditors: Market concentration and their role - Economic Affairs Committee Contents

Examination of Witnesses (Questions 83-113)

Mr Jonathan Hayward, Mr Stephen Kingsley, Dr Gunnar Niels and Mr Timothy Bush

26 OCTOBER 2010

  Q83The Chairman: Gentlemen, good afternoon and welcome to the Committee. Before we start hearing evidence, as I was unable to attend our first meeting I just wanted to make a declaration of interest, which I think probably applies to most people, as a director of a number of companies, all of which are listed in the Register of Members' Interests. All of those companies are audited by members of the accountancy profession. I think that is a blanket cover.

  This week's witnesses are consultants. Three of those who are with us this afternoon have already contributed written evidence and the fourth, Dr Niels, is from Oxera, which has published influential reports such as Competition and Choice in the UK Audit Market. They may be a bit livelier than last week's spokesmen for the professional associations. We shall wait to see.

  I will take the first question and then pass to my right. I am taking the chair today in the absence of Lord MacGregor, Chairman of the Committee. Copies of any other Members' interests have already been declared and are shown in the Register of Interests. Two Members of this Committee, Lord Currie and Baroness Kingsmill, are not taking part in this inquiry.

  You are all very welcome and thanks to those of you who have already supplied written evidence. I would be grateful if you could kindly speak loudly and clearly as this session is being webcast, and for the benefit of the shorthand writer. When we reach the questions we will be happy, in order to save time, to take silence from the other three of you as agreement with the first to speak, but if you disagree please do not hesitate to say so.

  Would any of the four of you like to make an opening statement? If not we will go straight to questions.

  Dr Niels: Just very briefly, if I may. It's a pleasure to be here. As you have mentioned already, I directed Oxera's work for the FRC and the DTI and also the study we did later for the European Commission. I'm hoping to help the Committee this afternoon with insights on the topics covered by those studies, competition and choice, in particular the microdynamics of how competition and choice in the audit market work. I'm probably less well placed to comment on some of the other topics that you cover in the inquiry.

  Q84  The Chairman: Thank you. I will take the first question and then I will pass to my right and we'll go round the table. The question I would like to ask you is—and I see this sitting on the board of a number of companies—there seems to be a view that only the mega audit firms with huge background knowledge and deep competencies are really able to audit companies, say in the FTSE 350. Do you consider that is overblown? I think it would be very helpful to the Committee if you could let us have your views on that.

  Mr Hayward: I think it is overblown, certainly, and I can answer it in two parts. Within the FTSE 350 there are some relatively straightforward companies. Property companies are a case in point. Even FTSE 100 property companies are, as organisations, relatively small. They may have huge balance sheets, but in terms of the number of people they employ and the amount of activity that goes on they are not big organisations. They tend to be largely domestic. Medium-sized audit firms have plenty of experience of auditing property companies; they would be able to audit those companies. However, no audit committee chairman of a FTSE property company that I know of is particularly interested in changing to a medium-sized audit firm because there is lots of downside and no upside for an audit committee in doing that. That example is the easy one. If any companies were going to change it could be those, but they don't.

  Less obvious are the multinational companies with more complex businesses, where again, I think, the dependence on a large international network is overstated. It arises because of the presumption, which is now embedded in the system of auditing, that a multinational audit will be done by auditing all the separate statutory entities around the world and adding up the results. That was a very good way of doing an audit when accounts were written down and then sent by steamship to head office in London, but in days when companies run an SAP system that manages data all over the world it's obsolete. Companies need to manage the way they operate globally in a different sort of way and it is perfectly possible to audit a company by working through the way in which it manages and evaluating whether or not it actually has a decent grip on the information within the business. That would be perfectly possible and it would mean that a different resource model could be employed. You would require fewer but much more expert auditors. It's only the fact of doing statutory audits all over the world which means that you require lots of statutory auditors in all jurisdictions. You then end up with a situation where having those auditors available to management is very useful to management because it's a resource pool that is available for them to call on when required.

  So what we have now is a system that works quite well. It suits the audit firms, it suits the management and it usually gets the right answer, but it is not the only way in which you could audit a large multinational company and other ways do not require those resources.

  Mr Bush: I am slightly concerned that the regulatory regime itself might overly identify with the larger firms and therefore by default squeeze out some of the competencies that are in smaller firms. Certainly when I qualified there was effectively a Big 10 and some firms that were quite small specialised in certain aspects of insurance; my firm I trained with, Ernst & Whinney, audited very large banks. I think there is a regulatory sense that big is best and I have an example that I think is quite interesting of a small firm in Ireland that identified that the accounting standard FRS 26 did not comply with the Credit Union Act and concluded it was imprudent and did not apply it. The Irish credit union regulator then flagged the fact that the standard didn't agree with the law and I'm not aware of any Irish credit union collapsing due to making insufficient bad debt provisions. Now, that is a freedom that seems to have come from a small firm in Belfast that I don't think was subject to a very strict regulatory regime that is effectively forcing auditors to follow standards even if they disagree with them. So I think that there are some issues between the best way of delivering what the law requires and what some of the standards are giving. That is a bit of problem.

  Dr Niels: On that question of the mega audit firms, 90% of the audit committee chairs that we surveyed thought that the mid-tier firms were perfectly technically capable, but if you then look at the three main products that in essence audit firms offer to the companies that they audit there are three aspects and on each of these three the Big Four have perceived or real advantages over the rest of the market. The first one is the technical audit itself where it is often perceived that, especially for the larger companies, you do need the international coverage, the depth and the sector knowledge, so they are better placed than the smaller firms. The second aspect is the value added services over and above the audit itself. That is a service highly valued by the audit committee chair and by senior management, and again the perception is that the Big Four are better placed, capable of providing those services. And then the third one is, of course, the reputational advantages of having a Big Four as your auditor, which you have discussed already at previous meetings. Again, that is where the Big Four have this advantage over the smaller firms.

  The Chairman: Mr Kingsley, did you want to add anything else?

  Mr Kingsley: No. I shall adopt your silence protocol, on this one anyway.

  Q85  Lord Lawson of Blaby: My greatest concern in this area is over the auditing of banks, to which I know that Mr Bush in particular has given a lot of thought. Reflecting on what we've been through with the banking meltdown, it has been discovered that banks were either taking excessive risks or they were taking risks they didn't understand or they were taking risks that they thought they had laid off, either through securitisation or through insurance and in fact they hadn't. It was a complete mess and it has been a major cause of the difficulties that the world economy and the British economy are in at the present time—not the only cause, but a major one. So I have some questions and I think your answers will be very helpful to us.

  The auditors were the dog, or one of the dogs, that didn't bark. Was this because they were unaware of the problems that I've just mentioned? If so, what have you to say about that? Were they aware of them but they were not required by law to make anything of it? Did they nevertheless privately warn the management and the boards of these banks of their concerns, even though they were not required to do anything on paper or publicly? If these auditors who were disquieted warned the board and the management and they didn't feel that the board and the management were taking sufficient notice of what they were saying, did they say to the regulatory body, for example, "I'm worried about this practice that's going on. You might like to look at it"? If not, why not? Did confidentiality, as it were, trump the wider public interest? I'm sorry to ask so many questions but they're all interconnected. Finally, what changes, if any, would you think would be sensible in the public interest to enable auditors in future to be a dog that does bark?

  Mr Bush: To answer the question on whether they spoke to regulators, all I'm aware of is that I think in the Banking Act there was a requirement for auditors to have a conversation with the regulators. I think that was the 1987 Banking Act.

  Lord Lawson of Blaby: It was. It's the Act I introduced. That's right. That's why I'm particularly interested in it. I hoped something would come of it but it doesn't seem that it did.

  Mr Bush: I think it disappeared from the Financial Services and Markets Act 2000. That said, I think that the biggest factor has been the loss of prudence in accounting standards. The principle of prudence was put in the law to protect the creditors so that when a company makes a distribution, or basically operates, the creditors aren't left with a bag of poor assets and therefore every asset should be valued on a reduce-to-cash basis. There are several problems with IFRS; I'll give just two in respect of prudence. First of all it has a bad debt provisioning model that is backward looking. It only requires provisioning for loans that have already shown an "impairment event", which is a bit of jargon in the standards. It doesn't allow you to take a forward assessment of risk and say on an historical basis, "Customers like that won't pay X amount and therefore we reduce our loans by that much". Therefore there is an inbuilt imprudence in the bad debt provisioning model. It went into the standard in 2003 at the last minute and then became operative within the UK and Ireland from 2005 onwards.

  The same standard has also excluded the general presumption to assess and disclose contingent liabilities. So you have a double problem within that standard of assets that may be over valued and then revert to the mean. Also, not disclosing contingent liabilities is a particular problem if you have had securitisations where, say in the case of some banks, you had make good clauses between the bank and the securitised vehicle, therefore if the loans in the securitised vehicle went bad the bank had to put good assets in. That's effectively a margin call. My understanding is that one bank that failed, initially with a liquidity problem, then had very large margin calls drawn against its securitised vehicles and I cannot find a contingent liability note in those accounts. If you look at the website you can find the word "contingent" in the directors' remuneration report but it does not appear in the note to the accounts. So I think there is a deficiency in that standard.

  That is pretty serious in itself because you're dealing with over-valued risk on one hand and possible cash outflows on the other. I believe that accounts that miss these things out or overstate assets are falsely giving boards the green light, because it's not just the process of getting the accounts signed off, it's the due diligence and the work that goes into producing a set of accounts and then the board understanding their risks from reading those accounts as they sign them off. I believe that these deficiencies are a major factor in the banks that failed within the UK and Ireland. I say the UK and Ireland because the UK and Ireland is a common standards area.

  By a bank failure, I categorise that as a bank that had lost 100% or more of what it had last claimed to have had as share capital and distributable reserves. There are at least eight banks that lost more than 100% of their capital, which is almost unheard of in modern times. In each case the bad debt provisioning is a major factor. I can add a ninth bank, which is an Icelandic bank operating in the UK, and for all of them the bad debt provisioning level sinks; for some of them it absolutely plummets. To reduce bad debt provisions by just 1% of the balance sheet basically increases the capital base by 20%. Certainly within Ireland you can see the lending capacity of the Irish banks taking off as the bad debt provisions are going down. The problem then gives a circular macroeconomic effect: there is more money in the economy, the collateral looks good because house prices are going up, but as we all know in the long run asset prices tend to mean revert. So I believe that that was a major factor.

  IFRS, International Financial Reporting Standards, came via the EU but the EU only required them to be applied for group accounts. The UK and Ireland were relatively unique in allowing them to be used in banking companies themselves, therefore there is the possibility of absolutely no risk assessment whatsoever if companies are using that standard within banking companies. The French are using French GAAP for banking companies and banking holding companies and you can see from looking at their accounts that they are making prudential provisions. Sorry, it is a rather long answer, Lord Lawson.

  Q86  Lord Lawson of Blaby: It is an extremely helpful answer. It is a very serious problem and you have obviously very thoroughly thought about it and made a number of interesting suggestions. I think, Chairman, it might be helpful if we could have a short note—I know you have written about them in various places—summarising the changes that you propose should be put in place. That would be extremely helpful to us.

  May I ask one last quick question? You mentioned a lot of things; all of them seemed to me to make very good sense. One thing you didn't mention was mark-to-market accounting. Do you feel that that needs to be re-examined?

  Mr Bush: Mark-to-market is quite difficult. We always had mark-to-market accounting for the most liquid positions on dealing books of stockbrokers. It's the best way of accounting for what they do. Very liquid positions were accounted for on a mark-to-market basis I think from the early 1980s, possibly the late 1970s. What happened with IFRS was that the general presumption that you shouldn't mark anything other than the most liquid positions to market was again relaxed to create a general presumption that you could almost value anything, not only on a market basis, which creates a problem if you have a very large volume because if you start cornering the market you can make the price of the asset go up. The EU Intervention Board used to do that and the Bank of England have done it with quantitative easing. So the problem is if you allow that as a method of recognising profit, you can make your own profit by just cornering the market. UK GAAP required a block discount if you had large holdings.

  Secondly, there are a group of assets where you are allowed to mark to model, so not even marking to a market but effectively trying to guess what a market does, which I think is quite difficult because it's very difficult for any one player to estimate what the total demand in a market is. You actually need to know how much everybody else is holding to even begin to plan. I believe that that aspect of IFRS was fairly well regulated outside of the territory of the United Kingdom because I think it was seen as a serious problem. I think it was more of a problem in the United States and it was certainly a problem in certain parts of Europe but, as we know, one major UK bank went and bought a Dutch bank where that balance sheet effectively collapsed shortly after the acquisition. So I think mark-to-market is a bit of a mixed bag. I've been looking at how the standards have been applied in different territories and how that has panned out.

  If I may add, in my short note I think what I'm going to say is we need UK GAAP back. London is a financial services centre because it has competitive advantage in a legal framework and a trustworthy information framework of accounting standards. For some reason we decided to go for an international model that is potentially a race to the bottom and I do not see the sense in the United Kingdom giving up sovereignty of accounting standards. It's as simple as that.

  Q87  Lord Forsyth of Drumlean: My question is for Mr Bush as well and I think he may have touched on the answer at the end there. First of all, your paper is absolutely devastating. We had an earlier inquiry where we looked into the financial crisis and one of the questions that remained hovering in the air was: why did the auditors not pick this up? Why did these very clever non-executive directors not see it coming? The first part of the question was should we go back to GAAP, which you have already answered. The second part of the question is, given your analysis, which really explains why particular banks got into particular difficulties and gives a reason and points to the failures in the accounting system, why is this not a bigger issue? Why is this not a great raging debate and why is nothing being done about it? What are the forces that are preventing this being addressed, given that we are now some way out from the crisis and there has been endless analysis about it?

  Mr Bush: It's an extremely arcane area and it's very difficult to start a conversation on the subject. Very often people will just freeze if you even start to mention it. Dare I say it, it's a bit like Murder on the Orient Express. If you start to find who was involved there are an awful lot of people that had their go in it. It seems to have crossed various Departments of Government; various regulators seem to have signed up to it. Dare I say it, IFRS was really an idea from round about the year 2000 and I think it was a bit of a gimmick. I think it sounded good to have international standards. It sounds good, like having one model of international law it sounds good, but I think it has failed at the practical level.

  I think it has also failed politically, because in my view the difficult things that have been left out of the accounting standards have been left out because they didn't suit the American accounting profession. Things like assessing risk and assessing contingent liabilities are things that the American accounting profession didn't want to address and that seems to have been the most malign factor in some of the standards in particular being poor.

  What I found interesting in looking at the law over the years is that company law, through being connected to the DTI investigations department, had worked out all the ruses that people had got up to over 100 years and safeguards were put into the legislation. What happened in 2005 was all of those safeguards were taken out in one go. The fact this is a crisis that has been largely caused by accounting is good news because I think we can regain our faith in human beings. Most people were not stupid or greedy, they were just doing what they were told to do. I think it would be good if the narrative can move on from greedy bankers and whatever to putting the system right and accepting that people have failings but basically if you allow a banker to produce a paper profit, they will.

  Lord Forsyth of Drumlean: So, go back to Old Kent Road, start again?

  Mr Bush: Absolutely, yes.

  Mr Kingsley: I just wanted to maybe present a slightly different view to Mr Bush's. I have been a bank audit practitioner for many years; I now sit on a board of a financial institution as a non-executive. The thing that strikes me about what has happened to accounting standards over the last 35 years is that they have become extremely complicated but they seem to be set and framed in a way that is divorced from reality. Taking fair value accounting as a very good example of that, the standard setters, almost using economic principles rather than principles of prudence and reasonable reporting, have come up with a set of standards that companies, banking companies in particular, are required to adopt. The profession has almost slavishly had to follow the guidance that has been given through these accounting standards, almost right or wrong. And so, when I was a kid there was a great store set on the truth and fairness of accounts, which could be roughly translated as, "Do these accounts make sense?" That seems to have gone out of the window and been replaced by something that says, more or less, "These accounts comply with accounting standards", regardless of whether they make sense. That is clear from the comments of the Lehman examiner, Mr Valukas. That was over in the United States, but I think similar concerns can be levelled here too.

  As to your dog barking, I found that intriguing. I think if we get to the stage where we have to rely on the external auditors to flag fundamental issues in banking institutions, I think we're in a bad place. You may be shocked by that, but I think the problem is much deeper than that and you are almost asking the wrong question, as it were, because they reflect an expectation gap as to what external auditors are really capable of doing.

  Q88  Lord Lawson of Blaby: Let me make clear, I did not say—

  Mr Kingsley: No, I am not criticising.

  Lord Lawson of Blaby: No, I am just clarifying, and you're perfectly entitled to criticise anyway. I was not suggesting, for a moment that the only failure—if there was a failure—was on the part of the auditor. I don't think there is any way in which auditors should bear the whole burden of reducing systemic risk, or whatever you like to call it. The question is whether auditors have a role to play. It seems, from what Mr Bush said, that he accepts that auditors could play—with the sort of changes that he was advocating—a useful role, even though, of course, there is no question that they can bear the whole burden of improving the system.

  Mr Kingsley: I think you're absolutely right, which sounds strange given what I have just said. I think the issue is that auditors are spending far too much time on trying to deal with accounting standards, disclosure issues, and all that kind of detailed stuff, and are not stepping back and seeing the wood for the trees. By which I mean that if the auditors of some of the more unbalanced banks in this country, in terms of management or balance sheet or risk, had stood back and looked at the business, looked at the people who were running at it, looked at the degree of oversight provided by the finance function, by the compliance function, by the risk function, by the NED, and so on, they might have thought, "Well, wait a minute, this thing is going to go wrong".

  Now you can say exactly the same about the supervisors, you could say the same about the analysts, but the auditors are pretty close, they're in there and they can see it—not every day but pretty regularly—and they can see, because they're there over a long period of time, how it is that risks and balance sheets and businesses are evolving. The truth I think is that for some of our banking institutions, the size and complexity got away from the capability of management process and systems to cope and identifying that is something that I think the auditors could and should have contributed to.

  Q89  Lord Tugendhat: May I ask two questions? The first is that it has been said, by a number of people, that non-executives of banks did not always understand very clearly the complexity and the underlying principles of what the business was engaged in. To what extent do you think that accusation could be levied at auditors? Of course these are highly trained people; of course they have a considerable technical knowledge of the technical subjects they are dealing with, but to what extent do you think that sometimes they were quite simply out of their depth in understanding the implications of some of the activities that they were looking at?

  Mr Kingsley: Sorry, I don't mean to monopolise the conversation. There is no doubt, I think, that you're right. Because if management was out of their depths then either the auditors had to second guess them or they were just as, if you like, blind as their audit client.

  Q90  Lord Tugendhat: Another thing that has arisen, when one looks at some of the banks that got into trouble—the Royal Bank of Scotland, Northern Rock, Lehman Brothers, and I suspect Anglo Irish would come into this category—you have the phenomenon of a very strong chief executive; a very dominant character. We have seen, in some of the banks I have mentioned, these people who have dominated their colleagues—both executive and non-executive. To what extent do you think we've seen the phenomenon of the very dominant chief executive dominating the auditors as well?

  Mr Kingsley: If you like I can monopolise the conversation. I don't think you can run a financial institution effectively without a lot of balance around the place, particularly in the boardroom and particularly in the executive group. By which I mean that there are people in the executive group who are charged with running the business—delivering profit and value to shareholders—and there needs to be a group that provides some challenge and oversight internally, and that is the finance function, the risk function, the compliance function and the internal audit function. If those functions are either not present in the executive committee room, or they're not listened to, or they aren't up to the challenges of the particular institution, then the management of that institution will become unbalanced. The reason that is important, and why that is different in banking from most other industries, is that banking is a very agile industry. It's a virtual industry with virtual products and it can expand, both in terms of size and in terms of complexity, very quickly. If you compare it to a supermarket chain, if I want to double the size of a Tesco or a Sainsbury's, it's going to take me a long time to do that. Even if I could get the planning permission it would take me a long time. If I want to double the size of a bank balance sheet it's not going to take me very long, provided I can acquire the capital to allow me to do it.

  That, combined with the virtuality of the products that they sell—most banking products are accounting entries wrapped up in a legal contract, which is why they can be so complex—means that a bank is capable of great complexity and capable of great agility, which makes it all the more important that there is challenge in the boardroom; not just between the non-executives and the executives, but within the executive group. If the challenge isn't there within the executive group then the effectiveness most of the external governance will likely be lost.

  Dr Niels: A brief related point. In our study there were basically two sectors that stood out as particularly complex to audit, which were insurance and banking. It was, I think, common knowledge that in banking there were only three of the four audit firms that do audit banks, so in the banking sector that reduced choice by one. In addition you then have the special rules on conflict. So you have a banking relationship, so that firm cannot also be your auditor or you can't audit that bank. So in banks the whole problem of choice has always been even worse, if you like, than in many other sectors.

  Mr Hayward: If I may just come back. I think Lord Tugendhat has identified one of the fundamental weaknesses of corporate governance, in relation to the dominant chief executive. It's not fashionable to say so but my experience of corporate governance, which is now the field in which I work—10 years ago I was a bank auditor but now I work in corporate governance—is that it is to a very large extent dependent on the extent to which the management wish to make it work. If the chief executive does not want corporate governance to work, it's very difficult for non-executives to do anything other than acquiesce or fire him. A chief executive is able to be dominant because he is successful, almost always. They're not always bullies; sometimes they are very subtly dominant and, therefore, being successful, can be extremely difficult for non-executives to deal with.

  I don't think there is a very easy way around that. It is a fundamental weakness in corporate governance. I raise it now because sometimes you hear corporate governance being talked about by regulators, for example, as if it were self-regulation. The expectations of corporate governance for preventing corporate disaster are very high. I don't think that corporate governance is always up to meeting the expectations that are placed upon it. Sometimes they are, and corporate governance can make a good situation better, but one of the ingredients necessary to make corporate governance as good as it can be is a good dose of humility on the part of the chief executive. Not very many chief executives get to that position by practising humility, so there are some real human problems in here and we should be careful how much we expect from it.

  Q91  Lord Best: What about corporate governance being proved, and would the power of a dominant chief executive be moderated if we had a different system of appointment of external auditors? Could we strengthen the independence of those auditors from either the management or the board? I'm thinking particularly about the change that I know exists in America, where it's a statutory requirement for major companies for the audit committee to appoint the auditor. Would that help in the UK and, indeed, in the EU?

  Mr Hayward: I doubt it. In practice in this country, the audit committee would have a veto on the appointment of auditors. In practice, in America, the audit committee requires the management to do the leg work on managing the process of selection of auditors. In either case, both management and audit committee are involved. Where the formal responsibility lies I don't see as making an immense amount of difference.

  Dr Niels: Just to add to that, my impression is that the audit committee clearly has a role and is playing a good role, and management also has a role in inputting into that. The other side of the choice could be to give more of a role to the investors themselves. So, for example, the appointment of the auditor could be made a more prominent item at the AGM, or there could be more ongoing discussion or questions from the investors to the audit committees about a choice of auditor.

  Q92  The Chairman: Mr Hayward, were you suggesting that, in the US, the non-executives would expect the management to do most of the leg work, as you put it, to recommend the choice of auditor but that isn't the case in the UK?

  Mr Hayward: No, that is the case here. The difference is that in the US it is formally the audit committee that makes the appointment.

  The Chairman: But that's a formality?

  Mr Hayward: It's a formality, yes.

  Mr Bush: Right, can I just—

  Q93  Lord Lipsey: Mr Bush, I hope this question will be such as to enable you to come in with whatever you were going to say. Another way of skinning the cat might be to insist that the firm changes auditors every five or seven years, or if you weren't going to be that draconian you could insist that it at least puts audit out to tender every five or seven years. Do you think that would help the situation? And are the costs involved so high that we should, nevertheless, rule it out?

  Mr Hayward: The problem you have to deal with here is that you have, on the one hand, a presumption that longevity reduces objectivity, which might or might not be a valid presumption. On the other hand, you have the distinct probability that newness will bring ignorance. So you have to choose between the possibility of impaired objectivity versus the great likelihood of reduced competence through ignorance. The only academic research I'm aware of on this comes from Italy, about 10 years ago, where they did a study of audit rotation; Italy being one country where they do require mandatory audit rotation. That concluded that the risks of audit failure in the early years, after a change, were greater than the risks of audit failure in the later years.

  Mr Bush: Can I add, I think this audit independence issue coming from the US is slightly addressing the issue in the wrong way. In America, the role of the finance director doesn't exist. They have the chief financial officer and they're very much lower down the food chain than the chief executive. Certainly, when I was in practice the finance director was the key, and the test of auditor independence was whether the finance director had captured the auditors, rather than the chief executive had captured the auditors. Therefore, this is a debate about auditor independence from a regime where the chief executive is all. There is nuance, of who's captured who, which I don't think is necessarily picked up. I think the risk to the public, and the shareholders, is if the chief executive has captured the finance director and then the finance director has also captured the auditors, and it seems to me you need that level of failure. But generally, within the UK, I don't think that audit independence is a significant issue.

  Dr Niels: On the question of rotation, there are clearly switching costs and there is investment required, in time and money, for a new audit firm to get to know the company. That can take several years in some cases and, indeed, in the early years things may go wrong. Therefore, perhaps one question to ask the firms themselves is: you already have rules on audit partner rotation and practices on mandatory audit partner rotation. You could ask them how that is working. Has that already produced systems where you can, internally within the firm, overcome switching costs? Are there systems in place that things can be handed over to the new auditor quite easily? So it may well be, if that is the case, that the transition to mandatory audit firm rotation is not that big a step change.

  Q94  Lord Lipsey: I take the point that research suggests there may be audit failure in the early years. However, you also need to ask whether there could not also be audit success in the early years. That is to say, a new firm comes in and it spots something that has been going on for years, which is dangerous to the company. You could have both an increase in failure and an increase in success, and then you'd have a difficult judgement: whether it was worth paying the price of the former for the latter.

  Mr Hayward: That is a possibility. But there are very few instances of accounts being restated following a change of auditors, which suggests that there are not really material discoveries of the sort that you suggest.

  Q95  Lord Hollick: We have seen suggestion that there is very little to choose between one audit firm and another. They have to comply with the IFRS, so therefore they have little discretion. They are technically competent—we hope—so price is possibly the only thing upon which they compete. Is there any evidence that if you change auditor there is in fact a saving for the company?

  Mr Hayward: If I may, I think they compete on two things at an audit tender—and remember audit tenders happen very infrequently, so you only get market competition between audit firms at very infrequent intervals. When they do they compete, it is on two things: one is price and the other is the quality of service, which is service to management.

  Q96  Lord Hollick: Which, I think we've seen from some of the papers presented today, could lead to a position of conflict, where the auditor has already provided advice and is then auditing the advice it has provided.

  Mr Kingsley: With respect, I think it is a bit worse than that. Auditors find it, I suspect, very difficult to maintain what is almost a schizophrenic position, in that on the one side they are responsible to the users of the accounts—primarily the shareholders—but on the other side the people that they relate to every day is the management. Therein lies, I think, a difficulty. I think what is curious to me is that there is a big variation. We talked about the differences of different firms. There is one difference, which is that some firms can take anything up to £1.10 in non-audit fees for every £1 in audit fees, and others are taking 25 pence. Given the size of the firms, relative to each other, which means they are relatively similar, this isn't a statistical aberration. There is something going on here, in terms of internal rules and guidelines or possibly some firms not being very good at cross-selling their services. I don't know which it is but it is curious, nevertheless.

  Dr Niels: There is evidence that the greater the concentration in the sector and the less the switching the higher the audit fees are. I think it is worth distinguishing between two types of competition concern, so one is a narrower concern about: this is a concentrated market, is there price competition, quality competition, between the Big Four? It's the kind of concern that you would expect the OFT or the Competition Commission to look at. Of course the two are not unrelated but there is a wider concern about choice, as such, and the systemic implications of this lack of choice. That is perhaps less to do with the audit fees which, at the end of the day, are a small percentage of a company's costs.

  Q97  Lord Forsyth of Drumlean: I wonder if we could pursue this idea of the alignment of the auditors with the board or the top management team, and I wonder what you think about the ideas that have been put forward by Joshua Ronen, the professor of accounting at the New York Stern School of Business, of having some kind of financial statements insurance approach—which I must say was a completely new idea to me—which does have the advantage of at least aligning the shareholders' interests with the auditors'. I don't know what your view is on that.

  Mr Kingsley: Can I take that? I have always been fascinated by it. I came across this concept about 15 years ago. It's been around for a while, believe it or not, although being arcane it has probably been buried under a lot of cobwebs. I think it is interesting, in the sense that when something goes wrong in an audit the audit firm becomes an insurer. Yet it doesn't act like an insurer in the work that it does particularly. It doesn't set its fee levels in relation to risk by way of example, which is what an insurance company would do.

  If you track back and think of what the accounts are and who is responsible for them, they are management's accounts. You read the audit report, that's what the audit report says, "These are management's accounts". The rules and regulations, in particular the law, are very explicit in laying responsibility on management for producing the right numbers, with requisite punishment if they don't. That I think that would start a chain of thinking which might lead you down to say, "Well, maybe we don't need mandatory audits of companies where there is external capital participation, where there are shareholders, et cetera. In those circumstances, you would go down a number of steps, one of which would be, "Okay, if management doesn't necessarily need external audit they do need internal audit". I think that's almost a sine qua non. You then have to establish very clear events of claim and responsibility, so that you know what kind of insurance policy you're writing, and obviously you can still permit management to have external audit because they can elect to do so. Obviously they would then have to justify those costs to their shareholders, but they could still do it, and then you have an insurance policy of some kind. I suspect the pool of money, in terms of audit fees, would fund quite a decent policy. I've never looked at it and I'm not equipped to look at it, but it certainly is something that is worth exploring.

  Mr Bush: That is very interesting because, about six years ago there was a debate about auditor liability. Some investors looked into this and in 1929 there was a proposal from Scottish Widows that they were going to insure financial accounts, rather than the accountants doing it. Then in fact the accountants decided to offer unlimited liability as their trade-off. I think they worked out that if the insurers did it, the insurers would end up employing the auditors; therefore, the auditors decided that it was a better proposition to go for a model that tried to avoid problems developing, as a fire alarm rather than fire assurance, I think, is the way of putting it. I think it is an interesting proposition. My understanding is that the core statutory proposition is that, because a company has unlimited liability, there is the risk of inherent corruption of whoever is running it defrauding the creditors, and that is right at the root of the law, which is why, other than some of the very small companies that have been exempted, most companies in Europe now do need statutory audits to protect the creditors, irrespective of whether they are listed or not. I'm not sure whether the insurance companies could provide that, but they probably could.

  Q98  Lord Smith of Clifton: Do audit firms still "low ball"—that is to say, cross subsidise—on their charges for audit work, in anticipation of additional and perhaps higher fees they expect to receive from their audit clients for non-audit work; or is this a thing of the past? You did say, just now, that the ratio between the audit income as opposed to the non-audit income varies. Could we have your comments on that, please?

  Mr Bush: I believe it is a thing of the past. In fact, I think that the change in the regulatory and the accounting standards framework has been very profitable for the firms. I think they probably don't need to cross-subsidise or low ball in the way that they were, and I think that the accounting regulatory regime has basically worked that problem out of the system. That's my view.

  Mr Hayward: That has been largely my experience also.

  Q99  Lord Lawson of Blaby: Leaving aside the cross-subsidisation problem there is also a kind of conflict of interest, which I was referred to by one of our witnesses, in terms of the management versus shareholders' interests conflicting to some extent. Do any of you feel that there is a conflict of interest problem to some extent, when an accounting firm will both be the external auditor and do advisory work for the same company or, alternatively, doing external and internal audit? Do any of you feel that there might be some sense in saying, "You can't do that"; saying, "You can either be the auditor or the advisor" to a particular firm? They can obviously do advisory work for other firms but not for the same one.

  Mr Hayward: I have a rather unfashionable view on this: I think there is certainly an issue of perception but I think the perception is not particularly grounded in the reality. I have seen some excellent audits done by audit partners, who provide lots of extra services to their clients, but they were intent on helping that client do the right thing and do it well. I have seen very second-rate audits done by audit partners who had no advisory work and weren't very interested in the client. The attitude of the auditors is much more important than any number, or ratio, that you put on the non-audit fees. This is a human issue not a simple mechanical one of, "If I get 75% of your audit fees and non-audit fees, I'm okay, and if I get 125% I'm corrupted". It doesn't work like that.

  Lord Lawson of Blaby: Could we hear whether any of the witnesses take a different view, or whether you all agree with what has just been said?

  Mr Kingsley: I take a different view despite the fact I've known Jonathan for a long time. I think it is an issue of culture and it depends—there's almost a question that underlies the question you've posed which is: what are audits for? What do we expect from these processes? If you take the view that public auditing is a public good, it's a public service, then I find it difficult to see how you can have a culture of public service sitting side by side with a culture of strong commercialism, as it were. Not that commercialism and the public service are necessarily incompatible, but I think that the culture of auditing, if it was seen as a service to the public and not as a service to companies and management, would lead you quite clearly down a path of saying, "Firms that do auditing shouldn't do other stuff". So the question in my mind comes back to saying, "What do we expect of audits and auditors?

  Dr Niels: Yes, in a way that would be not a demand side role, that if you audit a firm, you can't provide that company with consultancy services. You can also look at it from the supply side, which is you have to be an audit-only firm and I think that is a question that hasn't been debated as much as the demand side restrictions. I think it is worth thinking about.

  Mr Hayward: But it is a very difficult question, because what lies behind that is the question of what you want an audit to accomplish. If we make it completely independent, then you're saying, management do their own thing and the auditor comes along afterwards and passes judgement on whether it's acceptable or not. That is an entirely separate, distinct and potentially negative role. Alternatively, you have the role that auditors currently occupy at its best, which is a "right first time" role, to ensure that companies get their accounts into a condition where they can deserve a clean audit opinion. The majority of companies do get clean audit opinions and that doesn't mean that the auditors are blasé. A lot of those will have had a great deal of negotiation behind the scenes to get them to that stage, and that is a useful audit output that I wouldn't want to dismiss. I think if we are going to have that sort of audit, the right first time sort, then doing advisory work can improve the auditor's ability to do it, because he has a better understanding of the business and of the motivations and pressures of management. It places a great deal of importance on the attitude of the auditors and the culture of the firm. If you go to the other extreme and have it as an entirely independent thing, you're now into a different situation. You are going to have a more confrontational relationship with management, which will not necessarily produce a better audit, in fact. So I think there's a very basic distinction here between two fundamentally different approaches to audit.

  Q100  Lord Best: Lord Lawson was making the point or asking the question whether or not internal audit as well as external audit could or should be in the hands of the same auditors, or whether one should prohibit the same firm doing the internal audit operation. Would you go so far as to say that you wouldn't worry about that either?

  Mr Hayward: I would argue differently on theoretical and practical grounds. On a theoretical basis, I could make a good case for saying they could be done by one firm. If you had no internal audit function, then the external auditor would do more work. It would look very similar to internal audit work. In practice, if I'm a non-executive director, I would like more than one source of assurance, and I think it would be very foolish of me as a non-executive director to put all my eggs in one assurance basket.

  The Chairman: I think Mr Bush wanted to say something on this issue.

  Mr Bush: Could I just add one quick comment on this? I think one area where I would have concern is tax and tax compliance, because I would be concerned where there's a skeleton in the cupboard that the auditor isn't incentivised to uncover, and I think some tax planning can tie companies in knots for years, and if that is audited by the same firm that advised on the tax planning, then you're going to have a real problem, because you can be pretty sure the chief executive won't necessarily understand it and the problem will linger for quite a long time. So I think I would distinguish tax as a particular area.

  Q101  Lord Lipsey: I want to just follow in, if I might, on the point I was trying to raise, because I think you used a good phrase in saying there are human issues involved, but the trouble is that doesn't necessarily cut the way of your argument. I mean, the human issue involved is if you are the auditor partner and you know that your firm has twice as much consultancy work to take—in extreme cases, they have audit work coming out of the firm—is that not going to change the nature of your relationship with the finance director and his staff? Now, you may say that might lead to—I think you just did—a better audit, because you are so keen to get on with him well that it leads to a better audit, but going back to the perception issue, I wouldn't think it would be perceived as a better audit if it was done between two people who are big chums, because the human perception was that they were in each other's pockets.

  Mr Hayward: Indeed, and I think the reality of this is very complex, and I think as far as perception is concerned, it is probably a lost cause. The public will look at the simple appearances. If the purpose of an audit is to increase the public confidence in the financial statements, then there's not much point in trying to fight this battle.

  Mr Kingsley: Can I just go back to Lord Lawson's point on internal audit, because I think a company that outsources its internal audit hook, line and sinker is making a mistake, because like any outsourcer, you need to keep control of the relationship and direct the work that is being done. If the company gets itself into a situation where it doesn't have anybody who is doing that, then the value of internal audit has diminished significantly. There is a very big difference, at least in my view, between internal and external auditing. Internal auditing, and particularly the leadership of it, ought to be almost part of the management team and provide the kind of challenge to decisions that are being taken on a recurrent basis that an external auditor simply can't do.

  Q102  The Chairman: While we're on the subject of internal audit, do you think it should be mandatory for listed companies?

  Mr Kingsley: I think if you want the insurance route, for certain. For me, it's a management decision, because it's essential for the management group to have a function that is overseeing the quality of processing, what the CFO is doing and what his group is doing. By and large, I would say the answer is probably yes, but it would depend a little bit on complexity.

  Q103  The Chairman: But then presumably on that basis, internal audit could be just that, it could be internal, without bringing in any external advisor to look at it?

  Mr Kingsley: Absolutely, and in the largest places it should be, probably.

  Q104  Lord Hollick: Can I just ask the following? It's become clear, certainly in the financial services, that both the external audit—which has been described in various not particularly flattering ways by you gentlemen—and the internal audit failed to look at and address the real issue, which was risk and judgement around risk. Do you think the recommendations made by Sir David Walker around the risk committee are that the right way to address the risks in the financial services, and is that another role for accounting firms to play a part in? Should those be made mandatory? Should they be externally advised? Coming back to your question about the quality and diversity of assurance, it seems to me that's a most important part of the framework going forward for ensuring we don't get into the same mess again.

  Mr Kingsley: I think that's right, Lord Hollick, but whilst the auditors can play a role here, the power of the Walker recommendations goes back to something that I said earlier, which is around the balance in the board and the amount of challenge and oversight that's provided internally and externally. Walker was about balance, challenge and competence for the most part, and I think many of his recommendations are absolutely spot on, and if they are implemented in full, we hopefully will see fewer recurrences of what we have seen in the last three or four years. It will happen again, but hopefully it won't happen with the same force. The auditor's role in those circumstances I think is to provide oversight as to whether what has been done to implement the recommendations has been done to effect rather than to help in the implementation, if you see what I mean.

  Mr Hayward: I think you happen to have touched on a very important point here with internal audit, and I find it very pleasing that you have put it on your agenda, because it's unusual to find internal audit mentioned very much. It's hardly mentioned. Walker mentions internal audit about four times and always to its disadvantage compared to risk management. Internal audit is only mentioned in the UK Corporate Governance Code as one of the things that the audit committee has to look after. There is no real regulatory emphasis on the importance of internal audit, and I think that is something that could usefully be corrected.

  Q105  Lord Tugendhat: The FRC's Market Participants Group was established in 2006, I think, and it has published several reports. Do you think it has had any effect at all, or how would you assess its impact?

  Mr Bush: Can I come in? I think it's in an invidious position. I think it's very difficult for any regulator to effectively change the shape of the market. I'm not aware of a market being able to unpick an oligopoly, so in terms of competition, if there's a perceived oligopoly, I think it's one of the reserve powers of Government and politicians to resolve. I was chairman of a working party for about six months on this issue, and it's a very good way of losing friends.

  Q106  Lord Tugendhat: Can I approach it from another point of view? The Market Participants Group has a preference for market-led rather than regulatory action. I do too, lots of people have a preference for market-led, but I think we've established both in this discussion and in the other ones we've had that the market doesn't lead in this area, because for all the reasons that you have gone over and others have, the way that auditors are appointed is not proving very susceptible to market forces. So market-led doesn't get us anywhere. Is there a case therefore for regulatory action in order to widen the market, and if so, how would you see that working? Secondly, it may not be a very probable thing, but suppose one of the auditors was to decide to jack it in, if one of the auditors was to decide that they were going to withdraw from a particular area of business, wouldn't that require perhaps regulatory intervention to try to maintain even the level of competition that we have?

  Mr Bush: There is a structural problem with auditing that doesn't seem to exist in other professions. If I take architecture, the law, in all of those, a group of people can pull away and set up as a boutique. The independence rules of auditing mean that you can't, I think, have more than one client who is more than 10% of fee revenues. It's very difficult to get seed corn client base to be able to break away. Advertising agencies can do, almost every other industry can have breakaway people, but auditing seems to be one area where your staff get locked in and the client base gets locked in. I think that's why market solutions don't work, and I think there's nothing to stop the whole profession just becoming one firm, if it wanted to.

  Dr Niels: To be fair to the Market Participants Group, I think yes, market-led, but what it focused on was trying to remove certain regulatory obstacles as well to liberalise the market and facilitate market movement, such as the growth of the mid-tier firms, which hasn't happened. It's very difficult. One factor is indeed the factor that Tim mentioned, but it requires a lot of investment for mid-tier firm and gradual growth to become a significant player. That's very difficult. It will take a long time. On the other hand, we had a Big Eight 21 years ago and Big Six 13 years ago. There is no reason why in principle you could not have a market with six players or eight players who have that required minimum scale. But we arrived here through some idiosyncratic processes, including the approval of mergers by the competition authorities in the past, which with hindsight were perhaps not such a good idea, in particular the six to five merger, the Pricewaterhouse / Coopers & Lybrand merger. With hindsight, the European Commission said, "Okay, we allow the six to five merger, but we wouldn't allow a five to four merger". Well, five to four happened of course through market forces and we are where we are, but there is not a particular reason to try to get back to a situation where you have six or eight firms, because then that's probably the main way of avoiding the problem currently of lack of choice, which leads to systemic risk, if indeed one of the Big Four were to go out of the market, as you mentioned.

  Q107  Lord Lawson: Very briefly, please, on the audit dimension to the banking disaster, which I readily admit is not the only dimension, but it's the only one that is relevant to our inquiry, do any of you have any doubts and concerns about VAR and how it operated in practice, and if so, what should be done about it?

  Mr Kingsley: The trouble with VAR is it's economically extremely elegant and very persuasive in the way that it attempts to model what happened in the past, produces a curve of outcomes and therefore creates an expectation of what might happen in the future, and that is where I think the problem is. There are two issues. One is that in order to understand what happened in the past, we need to go back a long way, and the models that were used to drive VAR generally did not do that. So they didn't capture some of the worst things that happened in the markets in the past. Secondly it caused management to err in two bits of thinking: one is they could almost predict what was going to happen in the future, but secondly, there's something very beguiling about numbers. If you can reduce something to numbers, like risk, you think you have it licked. You think that you have it under control.

  Lord Lawson: It's the illusion of certainty.

  Mr Kingsley: Yes, and I think that is the biggest problem, so management would be presented with VAR numbers. They'd see, "Tomorrow we're at risk of £25 million. Tick" and the fact that it might be £125 million because of some black swan or whatever it happens to be wasn't today's problem, it became tomorrow's problem. So it's the illusion of certainty, as you absolutely put it.

  Q108  Lord Lawson: So what's the solution?

  Mr Kingsley: The solution is to use VAR as part of a more balanced approach to the management of risk, of which judgement and common sense ought to play a large part, and I suspect didn't in some institutions.

  Q109  Lord Lawson: Do auditors have a role here?

  Mr Kingsley: I think they have a role in observing the quality of decision-making that is being made. It's back to standing back and looking at the quality of the institution. One of the problems about being an auditor is that you're only as good as your client, at the end of the day.

  Q110  Lord Forsyth of Drumlean: Sorry to interrupt, but isn't the issue that if there's a shedload of money to be made, then common sense and this balanced approach goes out the window and the auditors are thinking, "Well, if we stand up against this, we're going to get fired"?

  Mr Kingsley: That applies to a number of issues that have appeared as a result of the credit crisis, including the idea of macro-prudential supervision, which has exactly the same problem attached.

  Lord Forsyth of Drumlean: Your answer to Lord Lawson's point is not much help.

  Mr Kingsley: No, not that in that sense. I had forgotten the human factor, but that's not fair. I shall double back. If in the boardroom there were voices of challenge through the CRO, the chief risk officer, through the internal audit function, through even the finance function, that said, "Look, these risks are oversized compared to our capacity to control them, and the balance between risk reward and capital that we're now facing isn't a safe one" then if that discussion takes place in a balanced way, you might not get, if you like, the greed overtaking the rationality. So your question runs to the core of some of the things that went wrong three and four years ago, and the need to introduce more balance into financial institution decision-making.

  Mr Bush: My former chairman, the late Alistair Ross Goobey, had a very simple test as to whether there was an economic problem. It was the number of cranes he could see from his office in the City. Whatever the value at risk formulas are doing, I used to say, if there were two free newspapers outside the tube station in the morning, and when one is called The Daily Deal it is definitely indicating a bubble. Just talking to estate agents who are telling you that people are going through a sealed bid process is definitely a sign that there's going to be a bubble. So I think there are definitely more empirical factors out there that you can use to apply in a commonsense way, rather than some quite complicated mathematical theory that might work in biological systems, but I don't think it necessarily works in financial markets.

  Q111  Lord Best: Assuming that it would be a good thing to have wider choice in the audit market, what would be the one thing that each of you would suggest to achieve that?

  Dr Niels: I think the option of structural separation is worth exploring more than has been so far. Structural separation can be done in two ways. One is to have audit-only firms and the rest. The other one is more a break up of the Big Four audit firms into smaller, but still large, audit firms. I'm not advocating that at all as a solution, but I'm saying that that is worth considering.

  Mr Hayward: In the paper that I submitted as written evidence, I outlined how this is a very systemic problem, there are all sorts of things that work together to give you a system that works quite well on its own terms. Unfortunately, those terms don't meet the expectations of those of us who are not auditors. But the problem with a systemic failure like that is that you don't fix it very easily by adjusting one single thing. So I think you can either do something outside it, the Florence Nightingale approach, ignore the War Office's approach to nursing and enlist The Times to create an entirely independent system alongside it. That's one approach when you have a systemic failure. The other approach is to do something dramatic that is such a shock to the system. A major audit firm failing would be a shock of that sort, but I wouldn't recommend it. I think the single shock that could be voluntarily imposed, that might have the most advantageous effects, would be to prevent group auditors from being the auditors of statutory subsidiaries around the world, because that would force them to adopt a different business model for the audit of large multi-national groups, and that different business model would increase the opportunities for smaller audit firms.

  Mr Bush: I agree with Jonathan on that absolutely. There's a distinction between being a director of a holding company and being a director of each company group, and I think one interesting example of a bank that was quite successful with that model was HSBC, which operates a pure holding company model. What Jonathan says I think might focus people's minds on which bit of the group has the risk, because I believe—I perhaps shouldn't say in this Committee—you can combine retail banking and investment banking under the same brand, provided you have proper firewalls between the retail bit of the bank and the investment bit of the bank.

  Lord Lawson: No, I am just observing, sorry. Sotto voce, firewalls are fine in theory, but they don't work in practice.

  The Chairman: Lord Tugendhat.

  Q112  Lord Tugendhat: In an earlier meeting, we were talking about the French system, where they have joint audits. As a result of that, the Big Four have a significantly lower proportion of the total French market than is the case in many other countries. Now, obviously that would be a regulatory intervention and I imagine that in the first instance it would add to audit costs, but if it is believed—and I say if it is believed—that it would be desirable to have a widening of choice, would this be an appropriate way to bring on new talent, as it were?

  Dr Niels: I had the impression that when you discussed it at the previous hearing there was a slightly negative view on that option.

  Lord Tugendhat: Yes.

  Dr Niels: I think I have been more positive about that option. Certainly, yes, it adds to the cost, but it is a good way for the mid-tier audit firms to get to know the companies, the bigger companies, to build a corporate CV, if you like, and then gradually gain more business. That is I think precisely the reason why in France the issue of concentration is slightly less marked than in the rest of the world.

  The Chairman: Mr Kingsley, did you want to add anything else?

  Mr Kingsley: No, I don't.

  Q113  Lord Best: There was one suggestion to the House of Lords in Question Time on 14 October that the Audit Commission's future might be as the fifth major audit firm entering the corporate sector and giving others a run for their money. Did that idea appeal to anybody?

  Mr Hayward: Personally, I would prefer the Audit Commission to carry on doing what it has been doing rather well up until now.1

  The Chairman: Gentlemen, thank you all very much for your contributions this afternoon. You have given us some very clear advice and views, and we're very grateful to you all.

1  Note by witness: If corporate clients are currently unwilling to choose, for example, Grant Thornton as their external auditor, it is hard to see why they should be more keen to appoint the Audit Commission. The fact that the Audit Commission exists is not in itself sufficient to overcome the barriers to entry in this market.

previous page contents next page

© Parliamentary copyright 2011