TUESDAY 23 OCTOBER 2007

Professor Gerry Dickinson

  Q1  Chairman: Professor Dickinson, it is very good of you to come. I will start by asking you to give us a bit more detail about the Geneva Association, but before I start any of this I need to explain to you that this session is on the record and is being recorded for a web cast. You will get a transcript of everything that is said during the session. We would like you to tell us a bit more about the Geneva Association and its work and you can either start by making a general opening statement, or you have seen the list of topics and we can start with those, whichever way around you feel would be helpful. This is our first shot at Solvency II and although I am generally familiar with banking regulations, insurance regulations are a stranger field for all of us.

  Professor Dickinson: Thank you very much, it is very nice to be here and to be invited. I hope that I can share some views with you and I hope to get some feedback, Baroness Cohen. You have mentioned the Geneva Association and perhaps I can address that first: what is the Geneva Association; what does it do? The Geneva Association is a club of CEOs of 80 of the largest insurance companies in the world, European, American, Japanese, Chinese, etcetera, so it is very influential in terms of the industry. It meets just to exchange ideas, it is not a pressure group as such; it exchanges ideas with each other as in a club, and there is a mutual learning process taking place. It sponsors various academic events—we have two academic journals to try and progress the thinking on insurance and risk and there is a global network of academics that we are linked into. We are also closely involved with the European Commission in various ways because of the work we do partly because we set up the Chief Risk Officers Forum in 2004, which plays a big part on the technical issues to do with Solvency II. So the Geneva Association has a link with the regulatory side. We also have links with the IAIS, which I think is relevant for your discussions going forward. The International Association of Insurance Supervisors in Basle is housed in the Bank for International Settlements, and therefore it is very close to the Basle Committee and it is obviously briefed as to exactly what is happening on the banking side. The Geneva Association has recently had an agreement in principle with the IAIS that we would in fact interface with the international insurance industry on changes in the regulatory regime coming out of the IAIS. So they are international; we are international. It parallels the role of the CEA, the Comité Européen Des Assurances in Paris, which is a group of national insurance associations, the ABI being one of them. The CEA interfaces with the CEIOPS, on the technical side. So there is a parallel there which will go forward. So we have been very actively involved in the technical side of this and also on the policy side to. So, in summary, it is a club of CEOs of insurance companies that sponsors research on an arm's length basis with academic research institutions, publishes two journals, has conferences, has open debates and is very transparent in its dealings on policy issues and it tries to influence the direction of good practice, best practice internationally for insurance companies. It also indirectly tries to make sure that the regulatory regimes are workable and do not constrain the market. So that is the Geneva Association.

  Q2  Chairman: Do you wish to make any more of a general statement or shall we just start asking questions?

  Professor Dickinson: If I could perhaps give you a background to the whole debate. The Solvency II initiative is obviously a major resource commitment by the insurance industry and by governments; it is a major change. The first question, is why is it necessary to change what we have? The current regulatory regime for insurance dates back to the first EU insurance directives in the 1970s but it has not really been changed since. These were capital based rules, rather static rules because in those days thinking was not well developed even in banking on risk-based approaches to capital. Nothing really changed much, partly because of the lack of some political will in the 1980s, but banking was similarly a little slow. So when the Basle I came in in 1988 there was a re-focus on having capital of a financial enterprise linked to the risks it takes, tailored to its risks. It costs to hold the capital, policy holders pay for this in the long-term. We have to have a matching of resources to regulatory cost with capital being one of those resources. So there was no change. After Basle I, in the 1990s—most regulatory change is reactive—there were some failures of US insurance companies in the late 1980s on the life side, linked to the junk bond crisis. With these failures in the US insurance market, and the US Congress said, "Why do we not have a risk-based capital system like the banks have?" Because the US insurance regulatory regime was not well developed as, it was at the state level and there was always the feeling that it was not up to best practice. So the Americans introduced the risk-based capital system and better practice and this was followed by the Japanese and other countries around the world—Canada and Australia. The Europeans were aware of this but did not move. They were aware, in the mid-1990s when this was happening and knew that the Basle Committee was looking at Basle II, a new framework which would be a second generation risk-based capital system for banks, under a Three Pillar structure. There was a committee of regulators under the German insurance regulator, Dr Muller, in 1997, which looked into this. The committee said "Shall we go with what the Americans have done and the Japanese? Or shall we wait and see." There were two other issues. One was of course that the banking regulation was changing with Basle II; "the other" was the IASB—IASC in those days—was developing International Accounting Standards for insurance, which started in 1997, so let us wait and see if the IASB comes up with International Accounting Standards that we can use for the balance sheets of the insurance companies. So those two were constraints on any change. I think there was also a lack of political will at that time and a little feet dragging. But there was renewed political will after 2000, and in 2003 the Commission set up CEIOPS which greatly enhanced the regulatory change, being driven now by a political process. Of course, the FSA in the UK—and I will come back to it a little later on—has always had a feeling that the regulatory regime was not up to muster for the insurance industry and of course with the convergence of regulators, it was aware that banking had a risk-based capital and insurance had not. So why cannot we have them similar. So there was a process where the two were brought together. The FSA played a big role, in bringing forward some of the ideas which found their way eventually into Solvency II. Of course, after the Independent Insurance collapse and then Equitable Life there was political pressure on the FSA to do something about the UK regulatory regime and they then brought in a risk-based capital system, but knowing that it would take time for the Europeans to get organised. The FSA wanted to move quickly but it anticipated, correctly as it turned out, what direction the European Commission would go later; it guessed right, but whatever happened subsequently it may have to change UK regulation. But it also influenced the process and as it started early. I mentioned the weakness of the EU current system, apart from being a static set of rules it does not look at certain areas of risks that the insurance company faces, such as asset risk—the risk of equity markets and bond markets moving up and down. The asset risk was not covered adequately in the EU insurance regulation, which was in effect Solvency I. Neither was credit risk—apart from reinsurance recoverables. Credit risk was not looked at properly; and of course operational risk came in with Basle II so this was also missing. So the new insurance regulatory regime that is coming in has had the benefit of waiting, because it has been able to capture the fact that Basle II has worked out in more detail. The IASB has come out now with a clear direction for International Accounting Standards, which has happened in the meantime; and we have in place corporate governance regimes which have encouraged better enterprise risk systems within companies. We also have better technology, financial modelling capabilities, so you can actually model things better than you used to be able to. So the combination of all these things has meant that Solvency II is not only a new system for Europe, but I would argue will be the first of a second generation risk-based capital system for insurance, and in line with best practice in banking. In some areas, it may be ahead of it, not least because it is European, whereas the Basle II is global, and we have had to focus not only the technical matching of capital to risk but also on creating a more efficient European insurance market which has meant that it is market congruent. It is consistent with market change—international trade, international takeovers, cross-border business, etcetera. This is not in the Basle II yet—but it is in the Solvency II. Because it has this non-national focus, it is a model that is transportable globally. So Solvency II is now seen this way, I was in China recently and in Korea talking to the regulators there, and they see that Solvency II is the way to go. They will take Solvency II, adapt it. The IAIS, which is in Basle, is also learning from the process within Solvency II. So Solvency II is influencing the global market. That is an overview.

  Chairman: Thank you very much for, Professor Dickinson, and I think you have shot my fox in answer to question one! Before I go on to ask Lord Jordan to ask a question does any colleague want to ask a question at this stage?

  Q3  Lord Cobbold: Yes. Do we take it that it is popular with the insurance business? And who are going to be the winners and losers in this? It seems to be very important that it should be popular. Also, is it going to affect small businesses and stifle competition?

  Professor Dickinson: Those are important questions. The insurance industry has been very involved in this whole process. It has been a very good partnership between the insurance regulators, CEIOPS, and the insurance industry—there have been differences, of course, but I will discuss those later. There has been a consensus that people are generally happy with what has come out of it. The problem, as you say, is who will be the winners and losers. It is complex. One of the worries I have had throughout is that we do not want to have regulation that is too complex because regulation that is too complex cannot be implemented; it cannot be monitored by the regulators. It is an issue of enforceability. The second issue is that the smaller companies do not have the modelling capabilities or internal expertise or data to actually do the modelling. So there is also a problem here. I think the industry as a whole benefits—although some will benefit more than others. The bigger companies will gain more at the expense of smaller companies, I believe, and it will be a force for more market consolidation over time.

  Q4  Lord Cobbold: There are some pretty large figures mentioned in the papers for the cost of this operation and that will again affect small businesses.

  Professor Dickinson: Absolutely. We do not have the model yet, or rather what the model will be in detail; we just have the framework. We do not have the detail of what the model will finally be—it is still being worked out. But it will be costly; we have to watch the cost of the regulation because the more complex it is, the more costly it will be. The cost issue will weigh more heavily upon the smaller companies so it could affect their competitiveness because it will be proportionally more relative to their premiums. There is a fixed cost element in this. At the same time the cost of regulation, if it is expensive, will be passed on to customers in higher prices, so there is a balance between cost and efficiency of regulation.

  Q5  Lord Steinberg: Thank you very much for the résumé that you gave at the beginning, which was helpful to everybody. I have a few comments and questions to ask but I will take the last one first, which was where you said that the costs would be greater and affect more the smaller companies rather than the larger companies, and that worries me quite considerably because as you will probably know if you only look at the insurance industry in Britain at the moment there is consolidation already taking place. We expect this week a bid by Standard Life for Resolution, backed by Swiss Re, which is again taking out the smaller company, which has developed quite well over the last number of years, and it is being taken out and going into a larger conglomerate in this area, and that all affects the consumer, does it not?

  Professor Dickinson: Absolutely.

  Q6  Lord Steinberg: And prices will go up. The other question is that there has been hitherto a cycle when premiums are higher and premiums get lower and the consumer is the person who is normally affected. I am gathering—and you will correct me if I am wrong—that Solvency II is more likely to increase the costs to the consumer rather than reduce the costs. Finally, the whole thing seems to me—and I should say that I am a retired bookmaker—a question of odds, that what you are looking for are the companies with the biggest asset backing versus the least risk, and you generally find that the smaller companies have a higher degree of risk and a slightly lower asset base, and that surely is going to cause a greater squeeze again, going back to the first part. Can you try and give me some answers to that?

  Professor Dickinson: Let me deal with the last one first. I think you are right, in fact, smaller companies clearly do not have the benefit of spreading the risk, as bookmakers do, of spreading them across the markets to make sure they can afford to pay. They do not have the diversification benefit that large companies have, hence they are riskier in general and therefore need proportionately more capital because of this. You have two types of insurance company, life insurance and non-life insurance—motor, property etcetera. The life insurance industry is slightly less risky depending on the nature of the contracts. If you are guaranteeing a lot, then obviously it can be very risky. But on the non-life side the big risks, commercial risks go to the big companies; there is a matching of types of customer to the market supply—BP and Marks & Spencer go to big insurance companies. Smaller companies often go to smaller insurance companies. The retail insurance market is shared. So it is not necessarily true that the bigger companies are less risky because they are bigger since they may be taking bigger risks on the commercial side. But in general I think that the smaller companies will be slightly more risky and they will be penalised a little bit more in terms of their capital requirement. Let us look at the other side of the coin. Many insurance companies, historically, within Europe started locally—in Northern Ireland, Scotland, Poland etc. Historically insurance companies, especially mutuals, have often had a local characteristic and serve the local populations there, especially Scandinavia, which has many mutuals. So there is a danger that these smaller companies that know the local market will be bought out, and this may impact to some extent on the quality of service provided to local customers. There is a cost issue and there is also a service issue. I am not saying that it is necessarily true that smaller companies are always better than the large ones—people move around a lot more now so the mobility of the population means that it is less important to have a local company—but I think there is an issue often balance between the large and small. If I can turn it back a little bit, if I may? If you look at the insurance industry over 20 to 30 years in Europe, the bigger companies have grown through acquisitions and they have tended to grow by buying—not very small companies, because it is not worth their while, but the medium-sized companies. So if you look at the structure of the insurance industry, it is estimated—and I think it is a reasonable estimate—that the top 20 insurance companies, the big ones, control about half of the European market. These are the global players. The number of companies that are licensed in Europe is about 5000. The effective number in terms of being under one management may be about 1,500. So you have a lot of small companies and a few big ones. The big ones are very big because they have bought the medium-sized companies, so we have a world of supermarkets and boutiques—we have little in the middle. It is common in most markets, and insurance is no different. So going forward we have customer benefit issues and the question of competition—to allow the smaller companies to survive. They are too small for the big ones to buy, but with private equity coming in now they can be bought and turned around and sold. So there is a danger, I should say—with some of this hot money that one can buy a company and turn it around, putting two or more companies together, package them together and sell to a big company. Maybe this process is a little bit too aggressive, and I think that one should look at the smaller companies across Europe providing a local or specialist service and some which are mutual companies. We do not want them to be all conglomerates. So there is an issue here. Customers may lose out, as they would have already left to go to bigger companies if these smaller companies were not providing a good service.

  Q7  Chairman: If I may draw the moral from that, if little companies are more heavily regulated and if the costs of regulation bear heavily on the little companies then the temptation to merge, to put themselves into handily packaged sized groups for the big companies becomes very real?

  Professor Dickinson: Yes, I agree. To some extent—and the FSA is doing a good job—I am not saying that is wrong, but the cost of regulation, the complexity of regulation makes smaller companies say, "Is it worth all our time at the board level doing this paperwork and not doing our business? I must sell and do something else." So there is a pressure at the board level if their time is being taken up, especially if you have extra responsibility for the governance. They may say, "Is it really worth it?" Regulation often has a bigger impact than you think. A small change in regulation can be disproportionate and I think in the smaller companies—and we have seen this in the insurance broking industry as well—you can have too much regulation. So regulation has to be proportionate, allowing for the effect on the smaller companies. It is something that I am sensitive to, because I think that the smaller companies can grow into big ones and we should given them the chance to grow.

  Chairman: Lord Trimble.

  Q8  Lord Trimble: From what you have been saying the one thing that stood out when you talk about the cost of regulation, if I understood rightly you are saying that we do not actually know what the cost of regulation would be at the moment because the model has not yet been done.

  Professor Dickinson: Yes.

  Q9  Lord Trimble: This strikes me as rather strange because the cost of regulation is going to be hugely important.

  Professor Dickinson: Absolutely.

  Q10  Lord Trimble: How is this going to work out and is it going to have an impact on the success or failure of the regulatory framework?

  Professor Dickinson: I think that we do not know. We know roughly what the model is going to be but the degree of calibration—the detail in the model is undecided. Of course the devil is in the detail—the cost is in the detail. We have a broad idea, but we will not know the exact final model until 2009. So I think a lot of effort should be put in to make sure of the model that is coming out is complete and theoretically sound but it is also workable.

  Q11  Lord Trimble: But who is going to take decisions or make judgments about this?

  Professor Dickinson: I suppose at the end of the day whether the Commission itself or the European Parliament actually says—perhaps it should come from the European Parliament to say, "How much is this going to cost?" The question should be put is—is it proportionate to the benefit? I think someone has to put that question. It is a little bit of a blank cheque at the moment.

  Q12  Lord Cobbold: The figure of £1.3 billion to £2 billion has been mentioned—quite a formidable cost.

  Professor Dickinson: That is just a guess, I suppose, of the cost. I have not seen that particular number; has that come from the Commission?

  Q13  Lord Cobbold: It is in the Government's Explanatory Memorandum.

  Professor Dickinson: I have not seen that particular number.

  Chairman: This would seem to be the moment, Lord Jordan, for you to ask about how all this is going to be put together.

  Lord Jordan: Professor, I would like to ask the question on some of the detail. Under the Lamfalussy arrangements the framework directive only sets out the principles that constitute the core of the new prudential framework. What will be the European insurance industry's priorities as the technical detail is agreed and what are likely to be the sticking points?

  Q14  Chairman: This might answer the questions on cost.

  Professor Dickinson: First of all you might say what are the priorities or perceived benefits? The first thing from the insurance industry's point of view is the calculation of capital, what they are considered to have in the regulatory regime. That is how much capital they are deemed to have, the permissible capital, and is this enough to meet the minimum that is required, given the risks that the company takes. So there are two sets of calculations: one is a balance sheet calculation, how much capital do they have; the second is, is it sufficient to meet the capital required on a risk basis? One of the benefits coming out of Solvency II from an insurance company point of view is that the capital on the balance sheet will be calculated on an economic basis, a realistic balance sheet. So the capital will be deemed to be higher than a conservative view, which was traditionally the case. So the capital will be higher, consistent with what it is in reality. If the capital is measured in a way that is consistent with the internal firm view on the capital and not understated by using conservative assumptions under a worse scenario. That is one of the first benefits—economic capital will be used. The second is that there will be a risk-based system, tailored to the company's risks. A major priority is that whether they will get the full benefit of diversification across the group. All companies want to have the full benefit of the diversification that they actually have. The problem is when you have subsidiaries within a larger group, the capital is locked up in the subsidiaries because they are independent legal entities. This is one of the priorities for the industry and also one of the challenges for regulators: how to view the insurance company with a number of subsidiaries in different countries, or even within one country on a "see-through" basis. So the priority for the bigger insurance companies in Europe is that they want to see the full diversification benefit that they have for the group enterprise on a "see-through" basis—as if the subsidiary companies do not exist. That obviously gives them the benefit of diversification—less capital is needed, the more the diversification, i.e. the large numbers. The regulatory challenge in this, is how do you make sure that the capital in the subsidiary, say in France, is sufficiently available to meet the liability in Germany or the UK? So is the capital fungible, i.e., can capital be moved around, and can one waive some of the limited liability issues that arise from having a subsidiary? So if there can be a diversification for the whole group, as though there were no legal entities within it, then this would give a benefit for capital. But that benefit should only accrue if that capital can be moved around.

  Q15  Lord Jordan: But does that not imply that you could have a failure in one country that could suck in and damage the company in other countries?

  Professor Dickinson: Yes. So the way the regulation is evolving is that there will also be group regulator. For example, Aviva is a UK-based group. Aviva will be regulated by the FSA, but it has subsidiaries in France, Germany, etcetera. The Aviva subsidiary in France would be regulated as far as its local adequate reserves, (funds to pay the liabilities) and the minimum capital, the MCR—absolute minimum. These funds will be held in France in the subsidiary. The extra capital that is thought to be needed above this is, called the SCR, a target level. The extra capital requirement can be held anywhere in the group. So under the system, funds will be held in France, enough assets to cover local liabilities, plus an extra margin for risk and the minimum capital requirement will also be held in France. The French regulator will also be monitoring the group's subsidiary there. There would not be the requirement, though, to hold in extra capital, which is the target capital, called the SCR—Standard Capital Requirement—in France. This could be held in the UK. So it allows the company to recognise this central capital. The funds that will be accounted for at the centre of the group would be the difference between what is called the total capital for the SCR, which is for the whole European group or global group, and the minimum capital. A significant amount of money. So some of the capital is held locally and that is the minimum, but the extra capital—for that rainy day—can be portable across Europe. It is not locked up in each entity, which fragments the capital and raises the cost of capital and the cost to customers. The other thing I should say in this is that, apart from the diversification benefits, under the new regulatory regime, an insurance company can use its own model like banking, rather than rely on the government's own Standard Capital Requirement, risk-based capital requirement. It can work out its own capital requirement. Solvency II says, "You can work out your own capital under our framework rules," and the regulator's job is to test whether that model is appropriate. So you can create your own threshold. I think this is good in many ways; the companies have to model properly internally, have better management internally, But there is a problem for the regulator, how do they monitor those models? Are the regulators up to it? One benefit is that you can set your own capital standard but the regulator comes and checks your model. The final point I would like to say is that in the past insurance companies have had to hold equity capital—and equity capital is expensive—and some degree of subordinate debt. So the amount of capital you can raise to grow the business is very limited and not competitive with the banks or other types of financial institutions. Under Solvency II, the eligible capital will extend beyond equity capital and subordinated debt to other types of capital, so there will be a wider range of capital available to the insurance companies. This will involve some contingent capital—that is, if you need the capital you can call on that capital—this has a lower cost than raising the capital and holding it. Debt capital or hybrid capital, provided it meets the regulatory constraints, is also cheaper. So the cost of capital will be lower and there will be more sources of capital available to the insurance industry. This lowers the cost of capital and makes companies more competitive. The other side, there are the mutuals, which are often the smaller companies, cannot raise equity capital. They can have calls on their members if they are short of capital, so the regulator recognises as part of the capital of a mutual that these callable funds, from their members, so this now counts as capital. So they do not have to hold too much capital. But if in measuring their adequacy of capital, they can say that they will be able to call capital from our members. This will count as tier two capital. Also, widening the scope of non-equity capital also helps mutuals. So within the regulatory regime we have more flexible source of financing, which helps the big ones but also helps the smaller ones, and mutuals too. This is already in the framework directive and this is one of the key benefits. The sticking point will be how much debt capital or contingent capital rather than paid up share capital, which is expensive.

  Q16  Lord Steinberg: May I come back in again, please? On the notes that we have here under Article 73 valuation of assets and liabilities, own funds Articles 85 to 98 and Minimum Capital Requirement at Articles 125 to 128, no mention has been made of free float. The person that I admire most in the entrepreneurial and in the insurance world is Warren Buffett of Berkshire Hathaway. Berkshire Hathaway has a free float of $50 billion and this ensured that his insurance company were able to handle a lot of the recent weather problems quite easily, whereas other companies did not. Could you talk to me about free float and what that means in relation to the capital aspect?

  Professor Dickinson: Warren Buffett views an insurance company as a source of funds. When he sells an insurance policy he does not see it as providing protection—he does indirectly—he sees it as a source of funds. Insurance premiums are paid in advance of claims so he can issue insurance contracts and effectively borrow funds—he calls it free float. If you look at Warren Buffett, as you do, he always benchmarks the cost of underwriting the loss as being the cost of borrowing. He benchmarks against the five-year US bond rate. It is a leverage; I borrow the money at five per cent, or whatever it is, and I then buy equities. What does he do with the money? He buys equities because if he likes risky assets. So he borrows from the policyholders—he is obviously careful—but he invests the money in risky assets-equities. The policyholders are safe because there is a large amount of capital, but if he did not have 50 billion of capital and that pool of cash he would find it difficult. Under the Solvency II Warren Buffett will have to have more capital for his equity risk; he in fact would have to put aside some of his capital.

  Q17  Lord Steinberg: I accept that he uses that as an arbitrage but, nevertheless, it is still an asset of the insurance company and as such the question I am asking is will that be taken into account in the assets and liabilities and so on in relation to the articles in the Solvency II?

  Professor Dickinson: The float concept you raise is his particular way of looking at how much is borrowed from the policyholders—that is the liability, his liability to the policyholders.

  Q18  Lord Steinberg: This is $50 billion in cash.

  Professor Dickinson: It is cash. First of all he has the float, the money he has raised in advance from policyholders, the premium, before he pays the claims. The premium on that float, with the money he might have borrowed. He is holding a lot of cash, it is equities and cash—he is in and out of the market, the stock market.

  Q19  Lord Steinberg: He is more in than out?

  Professor Dickinson: Yes. I think in a sense he is not ideal; he is an interesting case, he has a lot of capital, which means that he is obviously secure. It does not mean that his model is to be followed by anybody who is without that degree of capital.