AIRMIC ANNUAL LECTURE
RAC, PALL MALL 29 JANUARY 2002
Howard Davies
Chairman, Financial Services Authority

Thank you for the invitation to deliver your annual lecture. I was honoured to be invited. I was also quite impressed by the prescience of the powers that be in the Association. Because when you asked me to speak I was not the regulator of the insurance broking industry, nor was it planned that I should be. Since then, however, the Government have announced that general insurance broking will, in future, be looked after by the FSA. That came as something of a surprise to me. But it is clear that you, as buyers, managers and brokers of insurance were well aware of what was going on in the bowels of the Treasury.

My subject matter this evening is reflected in the title of tonight’s lecture: "rational expectations" what should the market, and policyholders, expect from insurance regulation?

I assure you that I will give you a direct answer to that question before I finish. I would not wish to be accused of misrepresentation or mis-selling.

There has already been a lot of debate about the rôle of regulation in the insurance industry, particularly over the last 18 months, stimulated by the Equitable Life affair in particular. There was some positive news from the Society yesterday, when the result of the policyholders vote on the compromise scheme was announced. The support was overwhelming, though court approval is still required before the scheme can go ahead. And, on the implications for public policy, we await the outcome of Lord Penrose’s enquiry, expected later this year.

Most of the focus of public debate has been on the regulation of the life sector. Tonight I propose to dwell more on the non-life side of the business. We have focused less on that so far, partly because we have had no remit in relation to the broking of general insurance products. Now, following the Government’s December announcement, we are beginning to work on the design of a regulatory régime for general insurance broking, and later on I will say a little about how we propose to proceed.

But before I begin to answer the question, I think it would be sensible to say a few words about how we see the state of the industry at present, both internationally and domestically. In designing a regulatory régime one should begin with an analysis of the state of the market, the dynamics of competition within it, and the areas in which there is the greatest a risk of damage to the interests of consumers.

The year 2001 was, not to put too fine a point on it, a challenging one for the general insurance industry. Indeed, for many of you, it was certainly a year to forget.

Clearly the biggest event was the 11 September attack on New York, whose consequences for the insurance industry worldwide have still not been fully digested. Estimates of loss range from around $20 billion upwards. And there is an uncomfortable gap between the "bottom up" estimates which can be derived from the provisions made by individual insurers and reinsurers, and the "top down" guesstimates of loss, based on what claimants and potential claimants have said. The bottom up estimates produce a figure closer to the $20 billion end of the range, while most of the top down estimates are now around $40 billion. There are good reasons to explain some part of this gap, to do with double counting and discounting, but I nonetheless find it striking.

And the tragic events of 11 September came on top of what was otherwise, for other reasons, also a difficult year. Underwriters have been making sizeable losses. One of the great mysteries of the insurance sector, to an outsider, is the so-called premium cycle. But however obscure the drivers of that cycle may be, it seems that over the last 12 months we have been at or near the bottom of it.

The year also saw some high profile and troubling failures. Independent closed its doors in June of last year, when its capital raising efforts failed. The circumstances of that failure are being investigated by the Serious Fraud Office. A much larger failure, albeit with smaller ramifications in London, was that of HIH, Australia’s second biggest general insurer. The Australian Government have set up a Royal Commission to investigate and we will be among those particularly interested in the Commission’s conclusions.

By the end of the year we were beginning to see some positive signs for the industry, if not necessarily for all its policyholders. As premiums began to rise, new capital has been attracted. The latest estimates we have seen suggest that perhaps $10 billion of new capital has gone into the Bermuda market so far, and perhaps about $2 billion into the London market. Premiums have begun to rise very sharply to the great concern of many companies, who have seen their commercial property premium rise very sharply. Some estimates suggest rises of around 50% across the board, and in many cases premiums have doubled, or even more than doubled. On the domestic front, industry estimates suggest that motor and domestic property premiums may rise this year by somewhere approaching 10%.

The FSA is not an economic regulator, and is not for us to determine whether these premium rises are in all circumstances justified. But I would say that we believe it important for the long-term health of the industry, and its clients, that there is some strengthening of the industry’s capital base, and that is bound to require some increases in premium rates. However, given the sizeable impact of these rises on the profitability of industrial and commercial customers in particular, the industry will clearly need to be able to satisfy those customers that it is doing what it can to control its own costs. My suspicion is that this pressure on costs, and the need to show stronger capital backing, may drive further consolidation both domestically and internationally. But that would be the result of commercial decisions which are not a matter for us.

There are, nonetheless, some trends in the insurance industry which do cause us some anxiety.

The first is what some have come to call "insurance-based investment banking".

There are a number of aspects to this, some of which have been well documented by the Bank of England in their recent Financial Stability Review. We have seen a rapid growth in credit risk transfer of various forms between the banking or investment banking sectors and insurers or reinsurers. More recently these transfers have taken more exotic forms, such as collateralised debt obligations (CDOs) or indeed synthetic CDOs. One investment banker recently described synthetic CDOs to me as "the most toxic element of the financial markets today".

When an investment banker talks of toxicity, a regulator is bound to take a heightened interest. That is why we and the Bank of England have been looking hard at the nature of these risk transfers, and importantly at the motivations for them. In principle, if the consequence is that risks are borne in the market by companies best equipped to hold them, or with the best ability to diversify those risks within their own balance sheets, that should be a good thing, with positive effects on the stability of the financial system worldwide. But there are two other hypotheses. First, that insurance companies may not be pricing these risks appropriately, perhaps because they lack the sophisticated technology to price them which investment banks possess. Or, and perhaps most likely, there may be some tax, accounting or regulatory reasons for risk shifting across sectors.

The taxation issues are not for the FSA, but the regulatory capital point is. If there are curiosities about the relationship between the capital régimes for banks and insurance companies which are driving this risk transfer, then we may be creating, not reducing market instability. Our general view is that the capital treatment should in principle be the same, where the risks are the same. But achieving that form of level playing field is not straightforward in an environment where the international capital régimes for banks and insurance companies have evolved in very different ways. I shall say more in a moment about our new approach to thinking about the prudential régime for insurers.

The second source of anxiety for us arises from our analysis of the circumstances surrounding the failure of Independent, and indeed of some other companies here and elsewhere in recent years. One aspect, which was also a feature of the Equitable’s reserving arrangements, was the existence of financial reinsurance treaties, of doubtful value, with unregulated reinsurers.

In the case of Equitable we documented publicly one case where a side letter casts serious doubt on the validity of the reinsurance treaty which, in the round, looked to be little more than window dressing designed to impress, or perhaps even to bamboozle the regulator. Like most of you, I imagine, bamboozlement is not something which I especially enjoy. So we have been investigating financial reinsurance with some care. And I have to say that we have found some more examples of arrangements where it is wholly unclear whether any risk has in fact been transferred, and where the motivation seems purely presentational. We have already required a number of companies to renegotiate these arrangements, where they seem to us to be unsound. But I believe that a number of companies should be asking themselves some hard questions about the prudence of their re-insurance practices, and indeed on both sides about the business ethics of arrangements of this kind.

My observations about the changing industry so far have focused largely on the underwriting side. But there is change under way in the broking community as well.

The UK broking industry went through a period of rapid consolidation at the end of the 1990’s, which resulted in the formation of the three huge groups: AON, Marsh and Willis. Now there is consolidation activity under way among the medium-sized brokers, in response to the changed competitive environment. We are seeing some interesting new developments as brokers attempt to put together regional networks. And of course brokers are having to come to grips with the implications for their business of the development of alternative risk transfer. The major brokers are having to see themselves as risk managers, in a broader context than they have previously done. That is causing some to develop new operating models in search of competitive advantage.

We take no particular view on these developments, as long as sound business principles are followed. But it is something in which we shall have to take an increasing interest as we begin to define our régime for general insurance broking of which more in a moment.

At the retail end of the market, there remains a huge number of small companies. No one seems to have a very clear idea of the precise numbers. We are working on the basis of an assumption that, as we take on mortgage and general insurance broking we shall probably acquire at least another 10,000 firms to regulate, on top of the 11, 500 who are lucky enough to be authorised by the FSA already. These firms come in all shapes and sizes, though the majority of them are small, with all kinds of combinations of businesses offered, and all kinds of different relationships with suppliers. Planning a régime which is sensitive to this variety of life forms will be quite a challenge.

So against that fluid background, of an industry under pressure on the capital side, in the throes of structural change, and in the process of redefining its business offering and its distribution, how should the regulator behave? To come back to my introductory question, what is it reasonable to expect a regulator to be able to achieve?

In determining our regulatory approach in any market, we must now begin with our statutory objectives, as set out in the Financial Services and Markets Act, which was finally implemented on 1 December last year. That Act tells us that we have four clear objectives.

First, we must maintain market confidence.

Second, we must secure the appropriate degree of protection for consumers, but bearing in mind the general principle that they should take responsibility for their own decisions. In other words, there is an explicit ‘caveat emptor’ clause in the legislation, although we are also told to have regard to consumers’ experience and skill, their legitimate needs for information and advice, and the degree of risk they incur.

Third, we must promote public understanding of the financial system. I hope that does not mean that I am required to describe the intricacies of synthetic CDOs in words of one syllable to primary school classes. But it does mean we should be working to ensure that financial markets are as transparent as possible and, in particular, that retail consumers are given clear information which allows them to understand the decisions they are making. Without that it is not reasonable to require them to take responsibility for those decisions.

Fourthly, and lastly, we are required to work to reduce financial crime. That is, you may think, an heroic objective. But in the aftermath of 11 September it is clear that work to reduce financial crime is an integral part of maintaining market confidence. Our, and your vigilance against money laundering, in particular, has been heightened.

We are also told, in the legislation, that we must pursue these objectives in a manner compatible with what we call the principles of good regulation. They require us to regulate in a way which is proportionate to the risks in each business, in a way that minimises the impact on competition and innovation, and which takes account of the competitive position of the UK’s financial markets. That points us, for example, towards a lighter touch régime in wholesale than in retail markets.

These are the objectives and principles that will guide us in the future in our relationships with the insurance industry and its customers. But the legislation does not put regulation on autopilot. Within this general framework there are some tough decisions to take. The biggest decision of principle is just how safe one wishes to make the industry. Let me put the point starkly in relation to the life sector.

It would be possible to reduce the risk of failure in the life insurance industry to approaching zero by requiring all life companies to keep their assets in the form of cash and short-dated government obligations. In those circumstances there would be next to no liquidity risk and almost total certainty about future returns. Those returns, however, would be unexciting, limited to the risk-free lending rate, with a reduction for the costs of marketing selling and portfolio management. In real terms, it is unlikely that life insurance companies would be able to show a positive return to investors.

So as soon as one moves away from this unexciting, but certainly safe model, one is allowing risk to enter the system. For quite a long period customers have benefited from the fact that life insurance companies have been able to keep a large proportion of their assets in equities. Just at the moment, that does not look such a brilliant idea, and bonus rates are being cut across the market in response to depressed equity market conditions after 2 years of market falls, the first such period for 25 years. But allowing the possibility of higher returns inevitably brings in its train the possibility of failure.

In the case of general insurance, too, an attempt to eliminate failure could have other damaging consequences. There would possibly be greater industry concentration, reduced product choice, less innovation and higher costs. A dynamic industry cannot be one in which all win prizes.

So we have said, from the moment I assumed responsibility as Chairman of FSA, that we are not aiming at a zero failure régime, which would be undesirable in theory and unachievable in practice.

It is a depressing, perhaps inevitable feature of my life that while I can secure broad theoretical agreement across the political spectrum for this general proposition, whenever there is an actual company failure, that consensus seems to fade away, and the immediate assumption is that there must have been some inadequacy in the regulatory régime itself, or in the way that régime was administered. I recognise that this comes with the territory. And I would certainly not argue that failures should simply be accepted without any assessment of why they occur. That is why we carried out, promptly and transparently, our own assessment of the circumstances surrounding the closure of Equitable Life to new business. But I hope it is recognised that a corporate failure is not necessarily a failure of regulation. It may simply be a sign that market forces are working. As Gore Vidal once observed, "it is not enough to succeed : others must fail".

But as we look at the insurance industry in the round, can we draw any conclusions about the appropriateness of the current régime? Is it sufficiently tight to reduce the rate of failures to an acceptable level, while allowing adequate room for competition, innovation and risk-taking? Or is there an argument for tighter discipline in some areas? This is a difficult question, one on which I think it is appropriate for us to have an open dialogue with the market itself, and its customers.

A few facts might help to illuminate that debate.

There are currently 39 general insurers in a formal insolvency process in the UK. Most of them failed in the 1990s. The gross insurance liabilities of these companies, as it has been quantified so far, amount to some £12½ billion. Liabilities for some recent insolvencies including Independent, Drake and HIH’s UK operations have not yet been quantified. On past experience we can expect that perhaps 40% of these liabilities will eventually be paid, so a reasonable estimate of losses to policyholders, both corporate and personal, might be around £7½ billion over the last decade or so. The Policyholder Protection Board’s accumulated net costs up to the end of the last financial year on general insolvencies was £291 million.

It is interesting to note that over the last 20 years the net cost of bank failures to the Deposit Protection Board, after recoveries, was £25 million. There were, of course, some other losses to depositors not compensated by the Deposit Protection Board but, with the single exception of BCCI, these have been very small over the last two decades.

Is this level of failure among general insurers acceptable? Answers on a postcard, please, to me at Canary Wharf. My own instinct is to think that this failure rate is possibly higher than we might ideally wish to see. Certainly the cost to both retail consumers and to the corporate sector have been non-trivial. At the least, I think it amounts to an argument for looking hard at the prudential rules, to identify whether they are sufficiently risk-based. That will be a central task for John Tiner and his team, who are reviewing the prudential rules for all insurance companies, among other things.

We have already said that we would like to move to a régime which is closer to that being developed for banks by the Basel Committee. In other words one which is more forward looking and involves firms themselves making their own specific assessment of the overall level of financial resources they need to meet their liabilities, and regular stress and scenario testing. We would envisage setting individual capital requirements for higher risk firms in future, and providing a consistent set of guidance on systems and controls, covering all sectors. All of these changes must be undertaken bearing in mind, as I have said, our principles of good regulation, and also taking into account developments under way in the European Union and the wider world.

We have to recognise that UK companies face global competition and that we should not tie their hands behind their back in that contest. But success in the insurance market is certainly heavily related to a reputation for safety and soundness, and for maintaining reliability in paying out claims. So a more robust prudential environment can be a competitive advantage. That is one rational expectation you may have of the regulatory régime.

We normally think of the prudential régime as something which sits under our market confidence objective. But it is clear from what I have said about losses over the last 10 years that prudential failure has been an important source of consumer detriment. So a robust prudential régime is part of the consumer protection work of the Authority, too. But we must now also bend our minds to the way in which the insurance broking community relates to its customers.

The insurance mediation directive, and the way in which the Government have chosen to implement it by giving the job to the FSA requires us to take far more interest in the front end of the business, so to speak, the policyholder interface, than we have in the past. So far, we have taken no direct responsibility for selling and marketing of general insurance. In my view, there will be spin-off benefits in terms of our understanding of the overall business, which will help us on the prudential side, too. It may help us to identify companies who are mis-pricing risk, and whose business practices are questionable. But the prime justification of the change is to enhance consumer protection. Against that objective, what should be our principal areas of concern?

So far we have done no research of our own to identify areas of potential consumer detriment at this interface between policyholders and the industry. That is an omission we shall certainly wish to correct before too long. For now, therefore, we are reliant on work done by the industry itself, or by consumer advocates. That work suggests a number of areas which are worthy of further examination.

The general question of transparency is one which causes concern among many consumers. Transparency both about the relationship between brokers and the companies they represent and transparency about the terms of policies and, in particular, policy exclusion. That appears to be a particular concern in relation to travel insurance, which accounts for 1 in 8 of all complaints to reach the insurance division of the Ombudsman scheme. Many of those complaints involve people who have had problems in claiming, particularly because exclusions about medical conditions have not been brought to their attention.

A second related area is suitability. And, of course, pressure selling of cover which is redundant or duplicative. That can be the case in the housing market, and also in relation to product warranties. Much of that selling takes place through retailers, rather than brokers. There is Consumers Association research which provides evidence of salespeople deliberately talking down the reliability of the product to generate commission on insurance.

A third area is claims-handling. The Ombudsman service had around 6,500 new general insurance cases last year, the large majority of which are about claims or the interpretation of contracts. Around 2,000 related to motor insurance. Of these general insurance cases 30% were decided in favour of the consumer and 25% were resolved by conciliation or mediation. (It follows that 45% were decided in favour of the insurer: when it comes to complaints, the customer is not always right). These figures suggest to me that there is some scope for an improvement in claims-handling both by companies and brokers, but that is something we shall need to investigate.

Then, lastly for this evening, there is the issue of fraud and negligence. This works in both directions. Fraud against insurance companies pushes up business costs and hence prices to consumers. The ABI estimate that fraudulent claims push up premiums by 4%, which amounts to a cost to honest policyholders of around £1 billion a year. But there is fraud by companies, too. We have encountered cases of intermediaries taking premium income but failing to place cover in the market, or placing cover negligently, so claims are subsequently denied by the underwriter.

There are other forms of dishonesty by intermediaries in the wholesale market. The commonest failure is misleading the insured by one means or another about the amount of commission taken by the broker, and hence about the net premium paid to the insurer. Others involve accepting corrupt inducements of one kind or another, for the benefit of either the firm or the individual broker. In addition it has been alleged, albeit never proved, that some of the spirals that have been created were driven by brokers seeking additional commissions by continually recycling the same business.

I should emphasise that we cannot say now how serious or widespread these dubious practices are. But I note that the industry itself was sufficiently concerned about the standards of business being done that it established the General Insurance Standards Council to use self-regulation to raise quality and confidence. As we begin to put together our statutory régime, we shall want to learn the lessons which GISC and other comparable industry self-regulatory bodies have themselves learned over the last 2 or 3 years. We will try as far as possible to build on the work they have done, and to ensure a smooth transition from self to statutory regulation. That is not an entirely straightforward issue. It will not be possible, as we have done in relation to the self-regulators which had statutory backing the PIA etc to arrange a grandfathering process whereby companies authorised by GISC can automatically be authorised by the FSA. But we have said that we will try to give credit to companies who have made efforts to be in good standing with GISC or another comparable industry self-regulator.

We are keen to ensure that GISC remains in existence between now and the introduction of the new statutory régime, which is likely to be some time in 2004, though the Directive has not yet quite completed its passage through the exotic European machinery. It would be most unfortunate if all the work put in over the last couple of years as allowed to fall away. We do not want to see a hiatus, in which standards are allowed to slip back.

So let me now try to give the second part of the answer to my headline question tonight, in relation to conduct of business, rather than prudential regulation. You should expect the regulator to set up a régime which minimises disruption and cost to firms, but which strengthens the confidence of their customers. A régime in which the consumer oriented regulations are proportionate to the risks and to the consumer detriment identified. And it should be a régime which does as little violence as possible to competition in the marketplace. We shall not, for example, be copying across to general insurance anything like the current polarisation régime with its market-distorting effects, which applies to advice on personal investments. We are indeed consulting on dismantling that regime and in general insurance, as elsewhere, we shall aim to minimise our impact on competition.

If we can meet these rational expectations, then the prize is potentially great. Both you, as practitioners and users of insurance, and we the regulators, have an interest in ensuring that the general insurance industry based in the United Kingdom has a reputation both domestically and internationally as a well-regulated yet flexible and competitive business. The international reputation is, of course, even more important in the wholesale than in the retail area. It was considerations of international reputation which, for example, caused Lloyd’s itself to ask for statutory regulation by the FSA something which the Government were prepared to incorporate into the Financial Services and Markets Act.

With a reputation for sound regulation, and for treating customers fairly, there is the prospect of an industry in which we see fewer failures, and greater consumer confidence. That could be one of those fabled ‘win win’ arrangements. They are, I grant you, hard to find in practice. But here there is the genuine prospect of a good result all round because, as we know, this is not a zero sum game. There is evidence of under-insurance in some parts of the market, both in the commercial and in the retail sector, under-insurance which in some cases arises from a lack of confidence in the industry and its intermediaries. Good regulation could raise confidence, expand business and reduce the pain and hardship experienced by individuals and businesses who face large uninsured losses.

But these high-minded aspirations of course need to be put into practice, and, in regulation, the devil lies always in the detail. We shall therefore need considerable help from the industry and its representatives as we move forward to design a new general insurance régime. I hope that we can count on your co-operation and collaboration over the next two years.

More Speeches: