 I'm going to talk about corporate governance and I'm tempted to start continuing Eleanor Hearn's line, but it's even worse than you thought, but I'm not. What I'm going to do is just run you through some of the corporate governance aspects of the CRD4 package, but also I want to give you a snapshot of some of the current academic thinking and research in some of these areas. Two general points to make about corporate governance regulation before I start. When I visualise corporate governance regulation, this is how I visualise it, a leaky bucket. I'd really like to be able to show a picture of a really strong, enduring dam, capable of holding back all the waters, but I can't. This is how I think of it. No matter how many holes we fix, more holes will reappear. Each time they do, we seek to learn from our past mistakes. We review the existing regulation. We identify the problems. We seek to respond to them with regulation, sometimes hard law, sometimes soft law, each time an increasing level of legislation or regulation. For example, the Cadbury Code was introduced 21 years ago. It was only two pages. Think of that, those of you who have looked through CRD4 or even Dodd-Frank, two pages only to set out the key parts. It was in response to criticism of lack of board accountability in BCCI and Polypec. Ten years later, it was another crisis, and Ron led the introduction of Sabane's Oxley in the States and the Higgs Review in the UK. So it goes on. Since seven years later, Lehman collapsed, and Joseph Stieglitz in the opening lines of his book Freefall said, the only surprise about the economic crisis of 2008 was that it actually came to a surprise to anyone at all. A related point on this is that many of the fixes we introduce are capable of introducing problems of their own. An age-old problem, which was identified by Adam Smith in the 18th century, was that managers of other people's money rarely watch over them with the same anxious vigilance with which we watch over our own. And partly to deal with that conundrum, stock options and share awards were introduced in the 1990s, and the whole idea of that was to align the interests of shareholders with the interests of directors. But even before the crisis, problems were arising in this and were being identified. In 2003, John Coffey from Columbia produced an excellent research paper where he identified a consequent and proportionate number of restatements in American listed company accounts, where directors' short-term interests had ensured that they adopted an overly optimistic view of accounts. Another example of potentially perverse consequences is the emphasis played on independence for non-executive directors. Now, it seems at least arguable that the focus on independence and the prioritizing of independence led to a lack of focus on the importance of expertise and experience in bank boards. And the market for corporate control is another example of perverse consequences we talked about M&As earlier. But just as Winston Churchill said that democracy was the worst form of government, apart from all the others that had been tried, so too corporate governance regulation may be the worst form of corporate governance, apart from all the others that have been tried. So it's important to keep this image in our mind when we talk about corporate governance regulation. We have to avoid complacency. The CRD changes that I'm going to talk about this morning hopefully will improve the situation. Hopefully they will lead to better board behavior. Hopefully they will preempt some of the governance problems which are coming down the track. Hopefully they won't engender problems of their own, but they will not constitute a panacea to all the difficulties. There have been a slew of reports, and you're familiar with them, on the financial crisis. Some of them such as the UK government commissioned Walker Report, examined corporate governance in banking. Some of them such as the FSA report on RBS, looked at corporate governance in particular banks. Others such as the Department of Finance's Neiberg report and the Commission's De La Rossier report took a broader review of the crisis. But what all of them had in common was that they identified failings in corporate governance. And these are the typical failings which have been identified. Issues in relation to risk management and internal control, deficiencies in the profile and practice of board management, perverse incentives, failures in disclosure and transparency, and then of course the complexity of the structures which are operating. Not so much a question of too big to fail, but too complex. It's true to say that corporate governance wasn't the cause of the problem, but I think it's also true to say, and the Commission, the European Commission, identified this in their working paper on the crisis, that timely and effective checks and balances in the governance system might have helped mitigate the worst effects of the crisis. And that's something that's worth thinking about. Okay, just to introduce briefly the CRD package. CRD-4 refers to a regulatory package which in practice actually includes a directive and a regulation. They came into force in July. It amended earlier provisions notably CRD-3. And in addition to implementing the Basel-3 requirements on liquidity and capital requirements, it also introduced a number of corporate governance provisions. The directive is directly applicable to firms in the EU. And sorry, the regulation is directly applicable. The directive will need to be implemented by member states. Most of the rules I'm going to talk about will apply from January 2014. Now CRD-4 will apply to all member states, not just those in the EURA system, but it does allow a discretion to member states as to how to apply various provisions, and that's very much to be welcomed. Also, when the single supervisory mechanism becomes effective, as was mentioned earlier, the SSM will take over regulation in this sense of the significant institutions. Another point to mention is the issue of proportionality which is built into the directive and the regulation in relation to some of the remuneration provisions. What that means is that the provisions should be applied to reflect the differences in different types of institutions, the different types of size, the different organisational structures, the nature of their business, the complexity of their business. And how that's going to work in practice really remains to be seen. Whether it's going to, in effect, neutralise some of the provisions is possible, but again, we need to wait to see how this is applied. The most controversial part of CRD-4, and possibly the one you've been reading about, is remuneration. And again, looking back to the crisis, inappropriate remuneration structures was identified as a problem. The FSA talked about risk being a driver, sorry, payment of remuneration being a driver for excessive risk, and that's true. Now, something which has also been discussed in relation to remuneration and corporate governance is the issue of excessive pay. Now, to my knowledge that hasn't been attributed as a cause of the crisis or a problem in the crisis, but again, it's something that you'll see as you look through the regulation, something that clearly was a cause of concern. CRD-4 hasn't been introduced on an empty playing field, so prior to it, CRD-3 indicated a number of issues which needed work. It suggested that significant institutions have a remuneration committee. It suggested that there would be a remuneration policy. It also emphasised the importance of aligning sound risk management with long-term growth with remuneration, and that's key, and that's really the focus, the undercurrent of all the regulation in CRD-4. It talked about a balance of fixed and variable pay, and again, that's something worth looking at because as we'll see in CRD-4, that's the area that was subject to the most change. Looking at variable pay, they also talked about looking at the constituent elements of variable pay, so they suggested that at least 50% of variable pay should consist of shares or share instruments. They also looked at the duration, so again, 40% they suggested should be deferred over a three to five year period. Now, those provisions, those last two blocks on the chart, they're still in effect, and they've been carried through in the current CRD. CRD-4 applies to identified staff as the earlier provision did, and identified staff is anyone who's likely to have effect on the risk profile of the bank, so it's been defined as senior management, risk takers, staff engaged in control function, and also any employee receiving remuneration in the same package. Now again, that provision has been carried over in CRD-4 from CRD-3, but the new part of it is that the EBA have been asked to produce some draft guidelines on this to determine how institutions should interpret what constitutes an effect on risk profile, and they haven't issued their definitive paper on this, but they have issued a consultation paper, which gives us an idea of some of the matters they're talking about, so they're suggesting, for example, that you would be one of the identified staff to list the remuneration measures apply if you're being paid more than 500,000 a year. If your variable pay constitutes more than 75% of your total package, if you have authority, which will have an effect on credit risk exposures, all those will take you into now a category which requires disclosure, and which requires restriction in terms of the provisions I'm going to talk about now. This is the most controversial aspect of it, the bonus cap. The term bonus cap isn't used, of course, but this is the provision which is deemed necessary to curb bankers' risk appetites. It talks about, for the first time, if you recall, the earlier slide talked about just a balance. Now the idea is, well, let's tell people what the balance should be. So they're talking about a ratio of one to one fixed to variable pay. They suggest that where shareholders agree that that can be changed, you can increase the level of variable pay, but only up to one to two percent ratio. Now the idea of giving shareholders an opportunity to comment on this, again, is very much part of the direction the EU commissioners are moving. They're moving towards a more shareholder-centric view. And again, we've seen that for all companies, not just in terms of banks, if you consider, for example, the changes which are introduced in Switzerland, the MINDER proposal in March on mandatory sale and pay, look at the changes which have been introduced in the EU in terms of, again, mandatory sale and pay, and look at the changes which have been suggested in the EU action plan in terms of shareholder voting on remuneration issues. There's also a possibility that the one to two ratio can increase again. Member states are allowed to alter that to discount the variable rate by about 25%, up to 25%. Member states can choose a smaller amount if they choose. And again, that will be on the basis that the variable pay is deferred for not less than five years. So again, the whole focus is on putting this into a long-term strategy for the directors of the companies. The directive is very clear. The new rules will apply to those paid in 2015, which means that it will apply to performance in 2014. Again, this will be a respective of the contractual obligations which have been applied. This has been the subject of a lot of debate, and again, you'll have been reading about it in the papers, not just from the institutions themselves, but also from regulators. Martin Wheatley, who's the CEO of the Financial Conduct Authority in the UK, said that it would lead to increased in the overall levels of pay. He said that it would be very clear what would happen. It's happened before. He said, when attempts have been made to do this type thing, there'll be an increase in the fixed pay in order to ensure that staff will enjoy the same overall level of pay that they'd enjoyed before. And that's exactly what's happening. A review which was reported by Robert Taff recruitment in August indicated that for UK financial services companies, pay had gone up by two-third since this measure was mooted. Two-third. He said, on average, there was an increase in 75% of the firms in terms of pay. There's been a cost analysis of this. Again, by the regulators in the UK, they expected that this cap is going to cost £500 million sterling in the UK banking industry alone. So again, it's extraordinarily controversial. The UK government felt so strongly about this that they refused to accept the CRD. They voted against it. CRD 4 was actually accepted on the basis of the qualified majority. And on top of that, they've challenged it. They've instituted a challenge to the European Court of Justice. They said that it wasn't fit for purpose. They said it was rushed through without any analysis as to the effect of it. And also, they said that the effect would be to do just what's happened to push up bankers' pay. And therefore, that would make the institutions more risky, not less risky. There's also provision in relation to clawback. Up to 100% of variable remuneration should be subject to a clawback or a malice, which is a negative bonus. And the institutions, when they're drawing up their provisions on this, have to set specific criteria relating to whether staff members participated in or responsible for conduct, which involved substantial loss to the entity. They also have to bear in mind that clawback should be based on a failure to meet standards of probity and fitness. Is this likely to be effective? Again, there's been a lot of talk about clawbacks. It's again supposed to be something of a panacea in this area. It hasn't worked hugely in practice. If you look at the statistics, for example, of annual reports in the last year, and again, year where we've had LIBOR scandals, year where we've had money laundering scandals, and the clawbacks have actually been tiny in practice. An example, HSBC last March indicated that they clawed back £700,000 of bonuses, but that was from a £20 million total salary bill. So again, nice idea. It may not be working in practice as much as we'd anticipated. They're increasing in CRD4. There's increasing transparency and disclosure requirements. So firms are required to indicate the ratio of fixed variable pay. There's also a greater granularity in the disclosure of pay for high earners, which would be somebody earning more than £100,000. The EBA benchmark these, and the competent authority in particular jurisdictions have an obligation to give this information to the EBA so they can benchmark, and an element of that has already been undergoing. So you can have a look, for example, at the pay ratios in all the member states for 2010, 2011 for the high earners. There's actually a note in the Irish Times. There's information on that over the weekend. And again, that was based on this survey, which was produced, I think, in July. So when you look at that, you see that there's a huge disparity between banks in different jurisdictions. So for example, in the UK, there were 2,436 high earners in the banks and financial institutions. The next highest was Germany. They only had 170. We came in at 21. Of our 20, and you won't be surprised to hear this, none were in retail banking. They were in investment banking in asset management and in certain other businesses. What you can also see looking at the statistics for Ireland is both the fixed and the variable rates components of remuneration have been falling. So they're falling on the whole. In terms of the ratio between fixed and variable, Ireland pretty much follows the UK in the moment at that particular balance. So our ratio of variable to fixed for 2011 was 336%, and it was very, very similar to the UK. This provision for periodic reviews of the directive, and I think this is very much to be welcomed in light of the concern about the lack of analysis, which was undertaken before introducing the remuneration provisions. I think it's very important to see this. The review is required not only to look at the effectiveness and the implementation of the provision, but they're all supposed to look specifically at whether the bonus caps have had an effect on financial stability. When we talk, though, about performance pay, when we talk about variable rate remuneration, the assumption is that if we designed it better, we'd achieve the outcome we desire. If we just got the balance between fixed and variable rate, that the incentives would be sufficient, and that shareholders, therefore, and companies and banks would benefit. That may not necessarily be the case, and there's a lot of research done from the field of cognitive science, which looks at this. So what I want to do is just take a minute or two, just to give you a flavor of the research which has been undertaken there. So the first research which was done in this area was done over 40 years ago. It was done by a psychologist looking at blood donations. So what he found was that entities which were looking for blood donors, when they offered to pay for blood donations, the number of people volunteering to give blood donations didn't increase, it fell. So it suggested that people were more willing to grant or to give donations when they were doing it for an intrinsic motivation. And an intrinsic motivation is something we do because we want to do it, or we feel responsible for doing it or just as bound to do it. And that research has been added to over the years, and it's been developed and one of the interesting aspects of it is what's called crowding out. And crowding out theory suggests that intrinsic motivations will actually be crowded out or destroyed if you introduce financial incentives. And there's a very celebrated study on this which is interesting, which looked at a crash in Israel. So in this particular crash, the owners were concerned because the parents were constantly late to pick up their kids. Stressed, hassled, came late. Apologetic, but what was late. So what they decided to do was they decided to introduce a monetary payment, a fine for parents who came late, and they thought this will stop it. What they actually found was exactly the opposite. Parents came later because they felt entitled to come later. They felt that this was a service in effect that they were paying for. And so what happened was that instead of worrying about the poor teachers who they absolutely respected, they felt, well, we're paying the poor teachers and we can do this. So monetary payments changed the social norms which were involved in the whole setting. Now some people have said, look, that's a social setting. It's not going to apply. But again, research suggests it does apply in terms of economic relations. Now there are various reasons for this. One suggestion is that people perceive that if there is no financial incentive, they perceive that they are doing their job. They're putting in effort because of social norms, because this is the right thing to do. If however you introduce a financial incentive, then people expect that what they're required to do is what they are paid to do, and nothing more than that. So the perception shifts in favor of the monetary incentive and again against the social norm or the intrinsic incentive. There's a different strand of research which relates to gaming of targets. Now again, that's something that most of you think, well, we would never do that. Jensen suggests most of us do it all the time and we do it when we set budgets. And he suggests that he describes budgets as a game that pays people to lie twice. So the idea is you lie firstly when you set achievable targets and then you lie secondly when you indicate that you have met those targets. And sometimes again, the suggestion is that the targets are met to an extent that may not necessarily be in the interests of the company. The last point relates to an idea that Tali raised in his book, The Black Swan. And he referred to the idea of overestimation of the effect and the influence we have on success. So if something goes wrong, we had nothing to do with it. If something goes right, yes, we had a huge element in making that go right. And this is an idea that you set a narrative. You draw your own narrative in a very self-serving way. So if you take the idea again of performance-related pay, you take a situation where the company is doing well, then the assumption is because the company is doing well, the director's pay should be based on the value and on the company profits. If the company isn't doing well and therefore they can't point to that sort of a value, then they say it should be based on the merit on what they have actually done in working very hard despite not reaching the targets. And if you look at the benchmarks of success of company profits during the year, you can see that replicated. Pay will follow those benchmarks if the companies are doing well. If the companies are not doing well, they don't follow them at all. So what this suggests is that performance-related pay isn't a solution for the alignment problem, not because we haven't designed it properly, because we haven't got the component parts correct, but really just of who we are and the way our minds work. The second topic I wanted to mention is boards. And boards, as you know, have been criticized roundly in the crisis for poor monitoring, for lack of challenge. The Walker Review, for example, said that non-executive directors failed as individuals to challenge the executive on substantial issues. So what the CRD does is it sets out a number of provisions to deal with this. Now bear in mind these are in a directive. Many of you are familiar with reading these type provisions that you can see on your screen, because you'll have seen them in the corporate governance codes, comply or explain codes. This is different. This is in a hard piece of legislation. So members should be of sufficiently good repute, possess knowledge, skills and experience. They should be honest. They should have integrity and independence, and that will allow them to challenge it. Again, the EBA are producing guidelines on this. I think we'll await with interest the contents of those guidelines. There's also a concern that directors failed to spend enough time really looking at their board papers. They failed to spend time looking at the financial statements and the risk measures. The reason for that, it was said, was that many of them were over committed. So in order to ensure that that over commitment doesn't occur, there is a restriction on the amount of directorships that can be held for identified staff. So, for example, in the guidelines which have been published, they suggest that you should hold a maximum of one executive directorship and two non-executive directorships, or else four. But that should be the balance. No more than that, or else you're not giving sufficient time to this. Again, we're all familiar with the provisions in the central bank, Corporate Governance Code for Credit Institutions and Insurance Undertakings, because they have provisions there to do exactly the same thing. And so you'll see a restriction. It's slightly different. In Ireland, a director in a significant institution should not have more than three non-executive directorships in an equivalent institution and no more than five additional directorships. That's eight in total. The central bank has published a consultation paper on this just indicating some of their ideas in relation to the application of the CRD-4. And so, again, these issues will remain to be resolved. What you can see, though, when you look at this, is there's an immense distrust. The fact that this is put in hard law, there's an immense distrust for the managers of the board and the directors of the board. It also sends a very strong signal as to what's expected. Hard law provisions on this. Will it work? Putting something like this into a piece of hard law, will it ensure that things happen? This is the headline that I saw during the week, which caught my eye. Is it better to be a cokehead than a blockhead? And this refers, of course, to Reverend Paul Flowers in the UK, who was the ex-chairman of the Cooperative Bank. And he was filmed in what looked like a drug deal, which came to media attention. It was a media sting operation. But in addition, he oversaw the management of the co-op and, at the same time, he was the manager, the chairman of the co-op. A large, large financial, a large amount of money went astray and was misreported. The sub-heading to Lisa Kelleway's article said, who makes the worst company chairman? One who's always scratching his head or one who's off his head? And again, this related to a leaked report from his interview in the Parliamentary Committee in the UK, the Treasury Select Committee, where he had misunderstood the amount of assets in the co-op. He said that they had about three billion in assets. At the time, they had 47 billion in assets. Now, he was interviewed by the FSA for this job. Now, he wasn't interviewed as a chairman, but he was interviewed for a non-executive director. When he was appointed as chairman, when he took over, it was on the basis that two other banking experts were appointed at the same time. But again, it appears that those two banking experts, in terms of strategic decision-making on the board, were outfoted by flowers in certain cases. So I think you have to ask yourself, would CRD have made any difference to this sort of a case? Is it going to mean that this isn't going to happen? And Brendan referred earlier to a lot of debate in the UK about how the FSA or the new FSA should operate. And one suggestion has been made, is that there should be annual reviews. But would an annual review even catch this? Would anyone ask the type of questions that would indicate this lack of knowledge? And again, the point that anyone going for an annual review with the FSA or, dare I say, the central bank will swat. You'll prep up. You will absolutely know what the balance sheet is before you go into an interview. So can you actually regulate these issues? There's also the idea of board challenge. And again, the manner in which decisions are taken on a board and the research into the human heuristics of decision-making is very interesting. Again, it's not new. In 1957, Herbert Simon talked about bounded rationality. And the idea of bounded rationality decides how people operate, how we make decisions. And he suggested that we make decisions based on the information we have. We also make decisions based on the limitation in our minds. And then we make decisions based on the time we have to make those decisions. So his work has been developed by a number of psychologists, notably Daniel Kahneman. And his book, which was out in the last year, which I'd recommend to you, Thinking Fast and Slow, he suggested a number of biases which work in boards, which again are familiar to us in the concept of a board. People believe a conclusion is true. When they believe that that conclusion is true, they're likely to believe arguments to support it. People are inclined to match their view of the qualities of a person to their judgment of one attribute that is particularly significant. Again, these are issues we can relate to in the context of the crisis. Groupthink. We all know about groupthink now. Groupthink is something that the regulators have taken into account. It's something that you'll see in the European Commission's review on corporate governance. It's something that you'll even see in the CRD paper itself. And these are some of the symptoms of groupthink, illusions of invulnerability, rationalizing warnings, self-censorship of deviating ideas. Sound familiar in the context of the crisis? Anyone who's read any of the reports on that, again, we'll see instances of this at the board level. Diversity. I mentioned that CRD4 referred to groupthink, and it did so in the context of diversity. It said that diversity was one way in which groupthink can be avoided, in which it can be challenged. Now, the definition of diversity in CRD4, again, is seen only in the recitals. And the definition suggests that it's not just gender diversity, which we tend to think of more often, but it's also a broad range of diversity measures, so age, geographical province, education, and professional background also included there. And the CRD suggests that a committee, a nomination committee, if there is one, and again, it suggests there should be one for significant institutions, should evaluate the balance of knowledge, skills, and diversity on the board in making its appointments. And then it talks about gender diversity. It says that institutions should set a target for the number of women they want on boards, and then they should prepare a policy and indicate how they intend to reach that target. And there's very little detail, even in the consultation, in terms of what that policy might have. They suggest that there should be an equal number of men and women put forward for interview, but that's about the only idea of policy we actually see. It's a lot less stringent than the proposed diversity on gender balance on boards, which is making its way through, it was accepted by parliament just two weeks ago, which requires 40% of women on boards by 2020. So it's a lot less prescriptive than that. Nonetheless, it's a useful measure. Again, it suggests that diversity be benchmarked, and that will be useful just to give us an idea as to whether it's accepted and whether it's working. The last issue I want to mention is risk oversight. And again, the problem with risk management is well known to everybody in this room. The fact that managers of companies failed to appreciate the risk involved with many of the strategies that they were accepting. The fact that they didn't have sufficient knowledge or understanding of what was involved, and therefore they weren't aware of the exposure of their companies. So these are the provisions which are in CRD4, which attempt to avoid some of these issues and to remedy the deficiencies. I suppose the most important one is the first thing, which says that the board is responsible for risk strategy and policies. Again, that's an issue which has been debated very much in the EU, whether there should be a legal responsibility. Again, it's obvious to lawyers that there's a fiduci duty to act in the interests of the company, which involves diligence, which in the case of a bank, involves informing yourself of risk management and attempting to mitigate it. This is the first time in this directive. It says that there should be information on the risk situation and that should be given to the board in sufficient time to make a decision. It says there should be a risk committee for significant institutions. Again, this should be full of experienced, knowledgeable experts on this area. It suggests there should be increased disclosure. Again, just a last look at what we need to learn from the cognitive sciences. Again, there are problems here. Studies will tell you that intelligent people have difficulties estimating probabilities, making predictions and otherwise attempting to deal with the whole concept of uncertainty. We're difficult at doing this. Some of the biases we've talked about earlier, but some of the additional ones I thought were worth mentioning just in the context of risk availability. People have a very limited memory and they tend to remember recent events rather than historical events. So when they're assessing risk, they take recent events into account to a greater extent. Anchoring the idea that we rely heavily on one piece of information for our decision making and once that anchor is set, once we have adopted a particular view, confirmation bias says that we really don't like moving away from that. We don't like changing a particular perception. We tend to rely more on a forecast than on a historic earning. Again, research suggests that it's an extraordinary idea. Relative positioning. Again, we know all about that here. The idea that people engage in excessive risk taking in order to outperform their competitors. Again, we know that's happened. Okay, the last point I want to make before letting you off for your lunch was just to note, does CRD4 in terms of governance make a difference? Is it likely to make a difference? Will it improve things? I think there are good aspects to it. I think the fact that it gives discretion to member states in order to deal with that flexibility, that proportionality, I think that's very positive. But as noted at the outset, I don't think it's an ultimate panacea, but on the basis that perfection is the enemy of the good, perhaps there's something in it worth seeing. So I want to leave you with a note that Professor John Coates made to the legislatures in the US Senate in 2009 when he was talking about corporate governance. And he said, corporate governance isn't rocket science. It's actually harder than rocket science. And he said it's harder for rocket science than rocket science because corporations in the absolute simplest sense are made up of groups of people who are coordinating their activities for profit. Rocket science involves looking at inert material moving through space, easy. Much harder is to try and predict how an individual, an actual person, will behave in any particular scenario. And even harder than that is to determine how a group of people will act. That's the challenge. Thank you.