It is a pleasure to be able to congratulate the Estonian Financial Supervision Authority on its fifth anniversary. I do so on behalf of the Financial Services Authority of the UK, and more generally on behalf of your colleagues in Europe concerned with the authorisation, supervision and regulation of financial services, and I do so most warmly. I wish you much success, and many more anniversaries.
The fact that the Estonian Financial Supervision Authority is only five years old should not in any way be regarded as a mark of immaturity, since youthfulness is a common feature of very many of the European financial supervisors: the UK's FSA has yet to have its tenth de facto or its sixth legal birthday; in Germany, BaFin goes back to May 2002; in France, the Autorité des Marchés Financiers to August 2003; and, in Poland, the Financial Supervision Commission was established only last year. So – certainly in comparison with central banks – we are all recent constructs. At my own personal advanced age, I find this youthfulness attractive.
One advantage of youth is that it encourages looking forward, not back. After all, those in their youth have a greater direct interest in the future than do those who have already enjoyed a high proportion of their life. On that basis, it is particularly fitting to be asked to initiate discussion on what developments we should expect in European financial markets and the consequences of those developments for financial supervisors and regulators. This morning, I want to concentrate on some of the issues which we as regulators and supervisors will have to deal with over the coming decade, and to consider how best we should tackle those issues.
I take as a given that we live in a world in which global competition is becoming ever more a reality, and are dealing with an industry which is internationally mobile: financial firms can – and do – move their country of operations relatively easily and quickly in response to what they see as the relative advantages and disadvantages – cultural, resources (particularly human resources), tax, regulation – of different countries. We are dealing with an industry which in many ways is becoming ever more virtual: as a sign of this, the building that housed the last physical trading floor of the London Stock Exchange is now being demolished (and physical trading ended there in 1986). But in all sorts of ways, physical transactions are giving way to electronic transactions. We are dealing with an industry that is increasingly international: I need not remind you here in Estonia of the reality of foreign banks and insurance companies playing a very significant part in the financial life of a country. But throughout Europe, increasing foreign participation in financial services is becoming the norm: the sixth largest UK bank is part of a Spanish banking group – Santander; and the sixth largest Spanish bank is part of a British group – Barclays. We are dealing with an industry which – in banking and insurance in particular – is undergoing profound changes through the pervasive effects of derivatives. The traditional lending activities of banks have moved on hugely, through a stage of origination and distribution of assets to an ability to analyse, divide and trade assets in a multitude of forms of often great intellectual and legal complexity – hence the growth of credit risk derivatives in all their forms. I expect all those trends to continue: increasing competition, ease of movement between countries, more transactions occurring electronically (and hence the growing importance of technology), products which rely on models, and offer novel means of dividing and spreading risks.
How much regulation?
Against this backdrop, let me set out some of the regulatory issues. The first, quite simply, is when should we regulate? How much regulation makes sense? And what should be the test which determines whether regulation is justified, or is likely to prove counterproductive or unduly burdensome? My starting point in answering these – linked – questions is that we should choose to intervene via regulation only where there is both a market failure and where regulatory intervention has the prospect of producing benefits which outweigh the costs which regulation inevitably imposes. This principle is enshrined in the policies and operations of the FSA, but it is not commonly accepted across Europe. It is remarkable, for example, that we were committed to the far reaching – and undoubtedly costly – changes set out in the Markets and Financial Instruments Directive (MiFID) without any such analysis of costs and benefits having been carried out. Indeed, in general I would say that the Financial Services Action Plan, which originated at the Lisbon summit, as a whole conspicuously lacked the sort of testing I advocate. It is much to be hoped that future EU initiatives will be subject to those tests. I am encouraged by Commissioner McCreevy's action in relation to clearing and settlement where he has promoted an industry solution rather than regulatory intervention; and I welcome his commitment to conducting impact assessments and cost benefit analyses.
Adoption of these criteria – identified market failure and positive benefits over costs – has considerable implications for regulators. To identify market failure you need to know about market success – more generally, how the market actually works: the added value at different points in the chain of producing and distributing financial products and services, and where those economic forces can be expected to produce acceptable results. This knowledge is not lightly acquired. It requires effort to gain and – in a fast‑developing industry – to keep it up to date. I think all regulators, regardless of size, find this a difficult task. Second, the analysis of costs and benefits needs to be done in a timely way, and with a level of precision – or approximation – which matches the stage which any proposal for regulation has reached: hence it should be approximate in its quantification at early stages of a proposal, and become more precise only as the proposal becomes better defined. But it should not be left until late in the process of regulation, for then it is often too late to move to a market rather than regulation‑based solution to the problem which has been identified; and the danger exists that the cost benefit analysis becomes an exercise in justifying decisions already made, instead of shaping those decisions.
I should add that proper analysis of costs and benefits – that is, analysis which identifies what the benefits and disadvantages are, and quantifies them with sufficient precision to shape decisions – neither is easy nor can it be done quickly. Proper process has implications for the timing, as well as the content, of regulation: if the Commission had been committed to this approach, CESR would have been given longer for its work on Level 3 of MiFID; as it was, the short timetable excluded serious analysis of this sort.
Home and host responsibilities
My second set of questions derives from the increasingly international nature of the financial services industry. What should be the responsibilities of the home regulator – the regulator in the country in which a bank, insurance company or securities company is incorporated – towards host regulators – those regulators in the countries in which firms have either branches or subsidiaries but are not headquartered? What should be the responsibilities of host regulators to the home regulator? And, from the firm's viewpoint, how can it avoid a mass of reporting and regulatory obligations, at best duplicated and at worst contradictory, in the various countries it operates?
To answer these questions, we need to develop a regulatory approach which recognises realities, not one based on the simple nostrum, favoured by some EU banks, that any financial institution should be supervised, for prudential purposes at least, pretty much exclusively by the home country supervisor – with the host regulators in other countries losing the scope to challenge the analysis and decisions of the home regulator or to act independently of them. Let me explain why I do not believe this is a real solution. There are at least four realities which the home country supervisor approach, at least in its simplistic form, does not acknowledge.
First, there is a question of legitimacy: regulators are accountable, and are properly expected to be accountable. It is simply unrealistic to expect a regulator in a country to answer questions about the activities of a bank important to that country by saying that as "host" regulator he or she has nothing to contribute and by referring those questions to a distant "lead regulator". It would, in my view, be unreasonable for the Estonian Financial Supervision Authority to respond to questions about the Estonian activities of Swedbank, by referring them to the Swedish Financial Supervisory Authority (Finansinspektionen) – just as it would be unreasonable for the Hungarian Financial Supervisory Authority (PSZÁF) to respond to questions about Citibank's activities in Hungary by referring them to the US Federal Reserve Board or Office of the Comptroller of the Currency. Swedbank is too important in Estonia, and Citibank too important in Hungary, for this to be possible.
Second, there are substantial differences between the legal powers granted to different regulatory organisations in different countries. Even within the EU there is no common basis of legal powers. And beyond the EU the position is even less clear.
Third, there are questions of political will: it is clear that the degree of independence of financial regulators is not uniform across the world, nor is it uniform even within the EU.
Fourth, there are questions of competence. Not all countries are able to devote the resources, or have the experience, to discharge all the responsibilities that might be expected of a lead regulator even with the benefit of some collaboration with host authorities. We should recognise that this is a real problem, which should not be glossed over.
So, for all those reasons, I think we should reject the simple "home country regulator" model. Instead, we need to establish a clearer and better understanding of the duties and rights of both home and host regulators, and devise working methods to make those operational. I believe that solving these home:host problems, and devising sensible solutions – sensible for both sets of regulators and sensible for the financial institution – will be one of the major tasks for regulators across Europe over the next decade. We have made some progress, notably through the work of CEBS on Pillar 1 of the Basel 2 arrangements, on devising solutions, but I think we have much more to do.
Operational issues
If we succeed in making progress on the two big principles I have described – a better and more selective choice of when to take regulatory action and a framework for home and host relationships – we will have more time to devote to what I call operational issues: the practical translation of legislation into what actually happens. I am a strong believer that it is important to concentrate on the effectiveness of legislation, not the existence of legislation. If I am allowed to stray into utility regulation by way of example, I am much taken by the fact that Commission reports on European energy markets two years ago stated that both the German and the British markets were 100 per cent open to competition at the retail level. Now, in a legal sense that was true. But in every other sense it was almost totally misleading: the German market was – and still is, as last week's report from Commissioner Kroes made clear – marked by continuing dominance by the traditional incumbents, vertical integration which deters new entrants, practical barriers to competition and an almost complete absence of switching by consumers of their supplier; the British market is in practice open, with substantial new entrants and up to 100,000 households switching every week. With experience like that, you will understand why I am concerned with implementation as much as – or even more than – I am with legislation.
Let me set out some of the practical issues which I believe we have to face in terms of implementation:
- How do we ensure that implementation is consistent across 27 member states? There is a particular challenge which arises from the accession countries, where the tradition of independent regulation is not universally entrenched. But it is not confined to accession countries. So we need a mechanism which allows some measure of monitoring of what is happening in various member states by way of implementation. How does this fit with the Commission's responsibilities for infraction proceedings? Should it act ex ante or ex post? And – since by definition it will be attempting to intervene in or at least influence national decisions – how do we avoid any process becoming politicised?
- How do we cooperate between regulators more effectively? I think that individual CESR members operate well on a bilateral basis when dealing with enforcement decisions: then we are at our best. But I am less clear that we work well on a cooperative basis to establish common practices, to learn from each other's experiences, or to establish a common pool of joint experience. For example, many of us worked two or so years ago to establish whether the US problems of market timing in mutual funds have also occurred in our own markets, and if so on what scale. But I was disappointed by the extent to which we have managed to exchange information, either about preferred methods of analysis or about the results of analysis.
These are but examples of implementation issues. There are many others – the whole of the Capital Requirements Directive being a particular issue for banks, MiFID for securities firms, and – stretching out over at least the next five years – Solvency 2 for insurance. Across financial services, consistent implementation will be an ever‑present concern.
Debt versus equity
Until now, I have discussed process points. Let me end with a more product‑related point. Securities regulators have traditionally concentrated, and to a large extent still do, on equities, rather than debt. This reflected two factors: first, the preferred position and lower risk of debt versus equity; and second the extent to which equity was traded more than debt. Neither factor carries the same weight today: the ability to take a debt asset and divide it through the use of derivatives into different tranches of debt with different risk characteristics makes the assumption that debt is less risky than equity at best simplistic; and the extensive trading of credit risk derivatives as well as changes in banking practices mean that debt trading is now commonplace. I do not believe securities regulators have yet got to grips with these changes, and that over the next decade we will move towards greater interest in debt instruments and debt trading. Our successors may well regard our present attention to equity products and transactions as disproportionate.
Conclusion
Those, then, are the issues which I believe will shape the work of financial supervisors in Europe over the next five or ten years: a better means of influencing when regulatory initiatives are taken, based upon a fuller understanding of markets and what they can deliver; the resolution of home and host responsibilities; the emphasis on working together to employ existing legislation rather than the introduction of new legislation; and, particularly for securities regulators, a new attention paid to debt instruments. These are issues which will test our abilities and I wish the Estonian Financial Supervision Authority well in its next five years in dealing with these issues.
