John Tiner
Managing Director, Financial Services Authority
Central & Eastern European Regulators Seminar
Wroclaw, Poland - 27 June 2003

I am honoured to have been asked to make this presentation to the annual seminar of Central and East European Regulators. It is also a delight to come to a city as beautiful as Wroclaw. For me, it is a particular pleasure to return to this region of Europe as following the Velvet Revolution in Czechoslovakia I had the pleasure of spending a number of years working in Prague with the major banks that emerged from the old state banking system. It is a great tribute to the spirit, will and application of the people of this region that so much progress has been made since that time.

For my own part, I joined the UK Financial Services Authority in June 2001 having spent 25 years as an accountant, auditor and consultant in professional services, but always working with the financial services industry. My arrival into the world of regulation coincided with the deepest bear market since the 1930’s and of course the tragic events of September 11th…..

My speech today is titled "The FSA’s radical new approach to insurance regulation". Later in my remarks I will go on to describe the many steps we are taking to reform fundamentally our approach to regulating insurance companies in both the life and property and casualty sectors, but I think it is important that I first put these changes into context.

The decision to combine financial services regulators into one authority was made by the Government shortly after their election to power in 1997. This decision was one of a package of measures that also saw The Bank of England given independent responsibility for setting interest rates. So the FSA came into being in June 1998 and, for the mathematicians amongst you, you will realise we have just celebrated our 5th birthday. The FSA is now the single regulator for the financial services industry in the UK and is the combination of 10 regulatory authorities which operated under the former system. All of the FSA’s work is governed by an act of parliament, called the Financial Services and Markets Act which was passed in the year 2000 and which became operational on 1 December 2001. Crucially, the Act sets out four objectives which the FSA is required to work towards and to report publicly each year on its achievements against those objectives. Just last week we published our report for 2002/03. These objectives are: to maintain market confidence to secure an appropriate degree of protection for consumers, to promote public understanding of the financial system and to reduce financial crime. In making our regulatory judgements about policy decisions, sector wide issues or individual company problems we are required to have regard to what the Act describes as principles of good regulation. These principles cover the need to be efficient in our use of resources, emphasise the importance of the responsibilities of the senior management of firms as the first line of defence in respect of our statutory objectives, require us to consider the implications of our decisions on competition, innovation and the competitiveness of the UK market and the need for our actions to be justified on the grounds that they are proportionate to the risks to the objectives I have just referred to. We have worked very hard to operationalise these objectives and principles of good regulation so that day in day out they are the benchmark against which we consider all of our key decisions, such as where we prioritise resources and the possible actions we may take against individual firms.

The FSA is sometimes referred to as an integrated regulator. Many people think of this as a single regulator across the various sectors of the financial markets – banking, insurance, securities, exchanges, advice - , and indeed that is true – with one qualification that I shall refer to in a moment – but, importantly, it also represents the integration of prudential supervision and conduct of business regulation. Some commentators suggest that there is an inherent conflict between the market confidence aspects of prudential supervision and the consumer protection aspects of conduct of business regulation. My view on this is that it is true there are times when there is a tension between them, although in a market economy it is by no means inherent in the system and I believe that any such tensions are best resolved within a single regulatory body than between multiple bodies who can only deal with the tension in a way which damages the very objectives the regulatory systems should be working towards.

The two areas of the industry not presently under the statutory regulation of the FSA are mortgages and general (property and casualty) insurance. But don’t worry, they will be soon. The British Government has announced that all mortgage lending, administration and advice will be regulated by the FSA starting on 31 October 2004 and, in accordance with the Insurance Mediation Directive, the sale of general insurance will be regulated from January 2005. This is a substantial widening of the FSA’s regulatory responsibilities. Presently, the FSA regulates some 12,000 firms and our estimate of the number of firms conducting mortgage and general insurance business is a further 30,000 firms. So it looks as though I shall be busy for a while longer yet. For the sake of completeness, I should mention that the regulation of occupational pensions is the responsibility of a separate body, OPRA, which is soon to be beefed up following a Government review of pensions; and consumer credit is the regulatory responsibility of the Office of Fair Trading.

I have spent a few minutes covering the background to the FSA, as I think it provides an important context for what I will go onto say about insurance. In particular, it was clear to us that out of the 10 regulators which merged to form the FSA, it was those that covered the prudential and conduct of business aspects of the insurance market which were in most need of urgent modernisation. As it turns out, there have been one or two individual firm cases, Equitable Life and Independent Insurance, which have provided the imperative from a political and public confidence point of view to progress rapidly with the modernisation. Before I go on to talk about this agenda of reform, I think it is important that I mention five key principles that are at the heart of the FSA’s regulatory regime.

  • The first is that we are risk-based regulator. This means that in respect of consumer and industry-wide issues and for each individual firm, we make an assessment of the risks they present to our four statutory objectives. We then allocate our resources according to that analysis and, therefore, determine the intensity of our regulatory effort at a firm level . We analyse on a quantitative basis, the impact that each firm could have on our ability to meet our statutory objectives and then we do a qualitative assessment of the probability of risks arising, which may threaten that ability. I would note, that our risk-based approach also encompasses enforcement, where we plan to take cases forward which have a significant market impact.
  • The second key plank to our approach is the "principles of business" that we require all regulated firms to meet. Essentially, these set standards for the way in which we expect firms to operate and include such things as:
    • A firm must conduct its business with integrity.
    • A firm must maintain adequate financial resources.
    • A firm must deal with its regulators in an open and co-operative way.
    • A firm must treat its customers fairly.
  • The third principle concerns senior management responsibilities. Here, we look to senior management to satisfy themselves that their firm is operating in a way which meets our requirements. It is not something that can be delegated to a compliance department or, in the case of insurance, to an appointed actuary. It is senior management and not the regulator who is engaged in their business day in day out and it is right that they should accept the regulatory responsibility which comes with managing their business.
  • My fourth principle is that the FSA does not aim to operate a zero failure regime. That means that firms will fail and consumers will be mis-sold, but neither can necessarily be regarded as a failure of regulation. We believe strongly that operating a zero failure system would operate contrary to the principle of caveat emptor, would have seriously damaging effects on market competition and innovation and would incur undue and unreasonable cost.
  • The last point of principle I would like to mention to you is to do with consultation and cost benefit analysis. Under the Financial Services and Markets Act the FSA has rule making powers, but only following a process of comprehensive public consultation. In most cases, therefore, we consult on our ideas for new policy initiatives to address a particular concern, issue a public feedback statement on that consultation and then consult once again on the draft rules to be included in our rulebook. This can slow down the process of improving the regulatory framework and places an onerous burden on firms, their trade associations and consumer representatives in reading and commenting on the consultation documents. But I am quite sure that this kind of transparency in policy making is an essential element in building confidence in the regulatory system and in minimising unintended consequences of such decisions. It is pleasing to see that our colleagues in Brussels are increasingly moving towards this form of public consultation and it should be expected that we will see such a practice embedded in the modus operandi of the Lamfalussy Committee structure. Indeed, CESR, the Committee of European Securities Regulators has written into its charter provisions for public consultation along these lines, including public meetings. As part of this consultation process, before we are permitted to make any rules, we must first justify them publicly by Cost Benefit Analysis. Generally speaking, this means quantifying an estimate of costs we think will be incurred by the industry in implementing the new policy and describing the benefit in the context of our statutory objectives. Our Practitioner Panel would also like us to evaluate the CBA on the cumulative effects of policy development and we are thinking how we might address this point.

Moving on then to the question of insurance. I think it is fair to say that the role of insurance companies within our economies, and indeed in the global economy is becoming increasingly noticeable, if not increasingly important. We are all aware of the demographic challenges facing Governments around the world and the implications for the funding of an ageing population. Falling equity markets and lower long term interest rates have further underlined how important a well capitalised and thriving life insurance sector is to the long term savings of hundreds of millions of people. The general insurance industry continues to be an essential mechanism in the transfer of risk from individuals, corporations and governments who do not have the knowledge, appetite or finances to take certain risks, to institutions that do. So, claims as diverse as the victims of the severe floods in the Eastern parts of Germany in 2002, to workplace diseases which may have incubation periods of several decades, to claims emerging from the terrorist attack on September 11th have all tested the resilience of the general insurance industry. Similarly, the re-insurance industry, which is un-regulated in many countries, has come to the fore in the light of its crucial role in supporting catastrophe risk and in maintaining financial stability.

It is probably true to say that, in the past, insurance regulation has been more focussed at the national level than, say, banking, although the work of the International Association of Insurance Supervisors has recently been more pro-active in working up standards on solvency, re-insurance and financial instruments such as securitisation and credit derivatives. We very much support this stronger co-ordination at an international level and, within the European Union the establishment of the Committee of European Insurance and Occupational Pension Supervisors (CEIOPS) which has a very important role to play in advising the Commission on implementing measures for the Solvency 2 Directive.

The insurance market in the UK comprises some 600 life and general insurers plus 200 small mutuals that we call Friendly Societies. Of course the UK also hosts the Lloyd’s of London Insurance Market, which has an underwriting capacity of some £15 billion.

When the FSA took over banking supervision from the Bank of England and insurance regulation from the Department of Trade and Industry, we took over two very different kinds of business. One very simple point is that the resourcing levels were very different indeed. We inherited almost 400 people for nearly 600 banks and just 70 people for over 800 insurance companies. There were some reasons to explain this remarkable difference. The insurance supervisors’ efforts were supplemented by the work of around 20 actuaries in the Government Actuaries Department, who have since also moved to the FSA. But the differences the staffing numbers, and the intensity of work, were nonetheless very striking. To simplify somewhat, the relationship between insurers and their supervisors was less close than was the case with banks. Insurance supervisors rarely visited the firms they oversaw. They typically did not know the senior management well, if at all. And the relationship was based more on lengthy statistical returns, analysed by actuaries, than on an understanding of business strategy and risk management, which is the foundation stone of banking supervision, at least its practice in the UK post BCCI and Barings era.

The problems of Equitable Life and the lessons for the regulators are identified in a public report by the FSA’s internal auditor, Ronnie Baird. This report highlighted a number of deficiencies with the system of insurance regulation that we inherited. Under the auspices of the Tiner Project we have looked under the stone of almost every aspect of how we regulate insurance companies and our proposals for change are set out in our report on the future regulation of insurance of October 2002 and the several consultation papers we have published both before and after that report. Today, I would like to touch on three of these:

  • risk assessments of insurers;
  • governance in life insurance companies, including the role of actuaries;
  • and risk-based capital for insurers.

We have now completed risk assessments of the largest 200 life and general insurance companies. These risk assessments are conducted by both on-site visits to firms, including interviews with all levels of management, and desk-based analysis of information, both quantitative and qualitative, that we have gathered from firms and other sources. Our risk assessment focuses on both business risks and control risks. On business risks, we want to understand firms’ strategies and their process for developing and approving strategy. We will want to assess how they manage their underwriting risk, including pricing, excesses, terms and conditions and exclusions and how they manage their net underwriting risk through the re-insurance market. We will also want to understand how they identify and manage investment and credit risk. So for example what are their investment strategies about asset mix, duration, concentration and what risk perameters do they set for credit exposures for example on their bond portfolio and, importantly, in respect of their re-insurance protection. In the area of control risks, we look at the governance arrangements of the company – the decision-making structure, independent oversight and, ultimately, how the Board exercises its responsibilities for the proper management of the institution. We then assess the firms’ systems and controls and management of operational risk, such as legal risk, systems failure risk and so on. Once we have completed this risk assessment we write to the governing body of the company, usually the Board of Directors, setting out our assessment and the areas where we think the firm or the FSA needs to take further actions to mitigate the extent of the risks identified. These actions may include firms agreeing to tighten procedures, say over the management of their outsourcing arrangements with third parties who process claims, or the FSA asking a firm of accountants to review the underwriting process where we may have concluded that controls may not be adequate and that a more detailed diagnostic assessment with recommendations for improvements needed to be made. Consistent with our focus on senior management responsibilities, we look to the Board to respond to us on all the points we make. Generally speaking, we have found this risk assessment process to be an effective way of understanding the extent to which individual firms may pose risks to market confidence, consumer protection or financial crime and that it provides a forum for an effective two-way dialogue with the firm and, for the 20 largest insurance firms, what we call a close and continuous relationship.

The second aspect of our reforms which we have announced this week, concerns governance in life insurance companies. Here we are implementing a package of measures to make more transparent how discretion in with-profit funds is exercised by Boards and senior management, bringing policy holder liabilities within the scope of the audit and introducing a public actuarial opinion to be published with the annual report. We are abolishing the role of the appointed actuary. We are introducing a new role called the actuarial function, which will act as an advisor to the Board on various actuarial issues and, specifically, will monitor the financial condition of the firm and advise on the level of capital needed to support the business. For with-profit funds, we will require a with-profits actuary to be appointed who will directly represent the interests of policyholders in ensuring they are treated fairly by the firm and that the fund is operated in accordance with the published principles and practices of financial management (the PPFM). The PPFM is a new document that we will require firms to publish from the end of March 2004. The with-profits actuary will not be able to be a member of the Board of Directors. I believe that these changes will reinforce the clarity of responsibility within firms, and strengthen accountability by the Board to its policyholders.

The third area of perhaps the most significant reform is in the field of solvency or what we might in the future refer to as risk-based capital. The drivers for change here are international accounting standards, the Solvency 1 and Solvency 2 EU Directives and the new Basel Accord for banks. Work on international accounting standards for insurance contracts and for financial instruments more generally, including progress towards fair value accounting will have a profound effect, especially for life insurance firms. The main effect is likely to be a shift away from creating reserves through a conservative valuation of liabilities, and away from the use of limits on the admissibility of assets as a means of making up for deficiencies in valuation. This will then require greater emphasis on capital to meet unexpected claims or losses and this in turn should create: greater clarity in assessing the true financial condition of an insurer; greater convergence between the accounting and regulatory treatments of balance sheet items, closing the gap between accounting and economic reality; and greater cross-sector harmonisation and international convergence.

The Solvency 1 life and non-life directives have made relatively modest updates to the existing directives, ahead of the more fundamental and wide-ranging Solvency 2 review. Good progress is being made in the EU towards the Solvency 2 directive, even if it is unlikely to be implemented until 2007 at the earliest. The current minimum EU standards are certainly out of date and seriously flawed. Notably in the non-life sector it is widely accepted that the margin of solvency is set too low and therefore most firms choose voluntarily to hold capital well in excess of the required minimum. The solvency margin is determined by reference to volume and not risk which can have perverse outcomes when premium income is falling. It provides no incentive for non-life insurers to practice good risk management and see a payback in the level of capital it is required to hold. We hope – and are actively engaged to trying to ensure – that the new Solvency 2 directive will be a major step in the direction of a genuinely risk-sensitive and properly calibrated approach to solvency requirements for both life and non-life insurers. We would also like to see a much greater role for scenario and stress tests in the framework. But at this stage it is impossible to be certain about what the Solvency 2 directive will contain. You might be surprised to hear me include a proposed international banking standard in the list of key international drivers towards modernising insurance regulation. But we do think that the framework provided by the Basle accord, consisting of three mutually reinforcing pillars – minimum capital requirements: supervisory review of a firms own internal capital assessment: and market discipline is one which has clear application in the insurance sector. That is not to say, I should emphasise, that we think the prudential system for banking should simply be transported across to insurance. Insurance is a different business to banking and we need a fit for purpose risk sensitive system of prudential regulation consistent with the three-pillar framework.

In the life insurance sector, we have observed during the recent equity market volatility that the way in which the statutory basis of solvency works in falling equity markets can force perverse decisions on asset allocation. Indeed, given the relatively inelastic relationship between asset prices and mathematical reserves, and the materiality of life insurers as investors in the equity market, that the current rules can push the market to overshoot at both the top end and the bottom end of a cycle. We, therefore, plan to introduce a new approach to measuring the solvency of life insurers based on "realistic" assessment of assets and liabilities together with a "safety margin" or "capital buffer" to cover adverse market developments. The key factors in the realistic approach are:

  • Stating liabilities at the amount a firm expects to have to pay, discounted to a present value. Importantly, for with-profits business, this means quantifying both guaranteed and discretionary benefits, the latter being more sensitive to movements in asset values, including equity prices.
  • Using capital markets techniques to quantify the value of options and guarantees that maybe embedded in policyholder contracts. This would mean taking into account the time value, intrinsic value and volatility relevant to the policyholders contractual rights. Firms are developing their valuation and risk management techniques in this area and the more advanced firms are now able to assess and value these commitments on the basis of stochastic models.

Life insurance firms must also be able to demonstrate that it has sufficient financial resources to meet its realistic liabilities under stressed circumstances, as well as under normal conditions. This is why a safety margin is needed in excess of normal realistic liabilities. We expect to require firms to assess a range of factors including key market risks such as equity values, property values and movements in the yield curve, as well as credit risk shocks in the asset portfolio.

Until the solvency system for life insurers is modernised at the European level, through Solvency 2, we will operate an amended version of the current statutory solvency calculation alongside the new realistic liabilities plus capital buffer test. The key amendment to the statutory solvency basis would be a shift from the current net premium method to a gross premium contractual method in the calculation of mathematical reserves.

I have referred previously to the controls we expect firms to have in place in managing credit risk and our proposal to include a credit risk shock in the assessment of the capital buffer. Recently, we have conducted the first ever detailed study on some key aspects of credit risks among UK life insurers and, if I may, I would look to take a minute or two to summarise the results of the study.

Credit risk is one of the many risks run by life insurance firms. It is the risk that a counterparty that owes them money will not pay the full amount promptly when it falls due. This risk exposure does not come from their direct customers; they pay premiums in advance to insure their benefits. Credit risk comes mainly from the investment operations of insurers, for example in investing reserves in bonds and other assets to build up a fund to meet future claims, or through hedging via derivative contracts. Credit risk exposure also arises through the placing of reinsurance contracts and cash management operations.

The nature of UK Life Insurers credit risk exposures has changed significantly over the last few years as they have switched out of equities during the bear market, mainly into bonds. From information gathered by FSA we estimate that life insurance firms sold about £30 billion of equities from their with-profits funds during 2002. We anticipate that most of this was reinvested in bonds.

Detailed information on credit risk exposures from holdings of bonds and similar securities has not been routinely gathered through the annual returns so a survey of firms with large exposures in corporate bonds has recently been conducted. The survey covered about 50 of the larger life insurers in UK with bond portfolios in excess of £120 billion in total. The results of this survey indicate that the industry in the UK has generally adopted investment policies that set appropriate limits for credit risk appetite and for exposures to individual counterparties. The investment portfolios generally comprise good quality credit risks consistent with the stated investment policies, and the reserves set up for credit risk are, on average, what we would expect, given the loss experience that they have suffered.

The survey was conducted by sending questionnaires to the larger life insurers in the UK and analysing the data contained in the completed questionnaires. Information was requested on investment policies with specific focus on the appetite for credit risk, exposures to individual counterparties, and exposures to market sectors. Clearly defined limits were set by most companiesfor the maximum exposure to credit risk, and about half also set a target level for the credit exposure that they wanted to take on, within the maximum set. Clear limits were also set for exposure to any one counterparty, although only in the investment portfolio. Some firms had defined rules for exposures to market sectors, though most firms said only that they maintained well-diversified portfolios.

We asked for details of firms’ 2002 experience, including details of defaults and the top 10 exposures at the end of the year. The default experience was broadly in line with the statistics for 2002 provided by the credit rating agencies, although it is interesting to observe that there was quite a large variation between firms. The top 10 exposures contained few surprises, generally comprising good quality bonds, with bonds with government-backed guarantees or issued by companies in the financial sector predominating. 91% of the aggregate portfolios were invested in investment grade bonds (that is, with credit ratings of BBB or higher), and 23% were in AAA bonds. This might even understate the quality of the portfolios as 6% of bonds were un-rated but the other information provided indicated that these comprised good quality bonds for which an external rating had not been obtained.

[The breakdown or exposures by credit rating grade was:

Credit Rating Grade % of portfolio

AAA

23

AA

17

A

33

BBB

18

BB

2

B

1

CCC

0

Not rated

6

The part of the portfolios that was not rated seemed to be comprised of good quality bonds with only a small proportion being distressed assets.]

Another interesting feature to note was that there was not very much evidence of significant holdings of credit derivatives.

We will be writing to the CEOs of life insurers to provide detailed feedback on our survey including identifying best practices which firms will wish to benchmark their own practices against to consider what, if any, improvements in the control of credit risk should made. Information was also gathered on reserving. This is a very technical area. Assets are brought in to the balance sheet at market value, a valuation basis that should make appropriate provision for all risks, including credit risk. Firms must make a deduction for risk from the redemption yield on the asset in setting the basis for calculating technical provisions for statutory purposes. The bases described for making these deductions would produce, on average, deductions at a reasonably prudent level. Some firms also had procedures for limiting the risk-adjusted yields on bonds where these would otherwise be unusually high.

On the non-life side we will be publishing a consultation paper in the next week or two on a risk-based enhanced regulatory capital requirement. We intend to decompose risk into three main components: insurance risk, counter-party risk and market risk. Within these, different capital charges will be applied to different types or classes of business to reflect the different risk in each case. So, for example, within insurance risk, household, motor and travel insurance might be subject to a lower risk weighting than long term liability insurance. Similarly, asset risk charges will aim to capture, at least to some extent, counter-party risk and market risk – including interest rate risk – by applying different risk weightings to different types of assets. We are not, as I have said earlier, aiming for a zero failure regime but we do want the minimum requirement to reflect a reasonably high level of confidence that an insurer will remain solvent even if subject to a plausible range of adverse shocks. We also think our proposed approach introduces much greater risk sensitivity into the requirement, aligns regulatory capital more closely to economic capital and allows for a more transparent measure of capital adequacy. At this stage, we are simply consulting on an approach to the calculation of enhanced regulatory capital and we will work closely with the industry to determine whether our proposed calibrations are appropriate, what impact the new tests have on the level of insurance company capital and when the new tests should be introduced and in what form.

Looking forward, we will afford insurers the opportunity to use internal models and stress and scenario tests to determine their own level of capital required to support their business. We will take into account this internal capital assessment in considering whether we should grant waivers to allow firms to operate at a level of capital lower than that determined by our rules.

All of this presents both the regulator and the industry with significant challenges. There is an urgent need to deepen and broaden the pool of resources within the insurance industry who are able to lead the development and implementation of these new concepts.

More Speeches: