1. The text of this speech is set out under five different headings. The first contains general notions on financial stability and the Eurosystem, i.e. the words that constitute the title of this speech. The second deals with the changing context in which the pursuit of financial stability is taking place in our time, both inside and outside the European Union (EU). The third heading covers specific elements of the European and euro area scenario. Next, I shall take a look at the policy side of all this: what are the problems and the configurations within which the policy objectives related to stability are to be pursued in the present institutional setting of Euroland? Finally, future developments are highlighted.
2. Let me first comment on the notion of financial stability. At first sight, it seems to be a straightforward notion, but from the point of view of policy-making, the notion of financial stability has much more problematic aspects than one would imagine.
We talk often about financial stability, but less often about stability in other important fields of economic activity. Hence, there must be something specific about financial activity which raises public policy concerns. Financial activity is about promises, confidence, expectations; all factors which may be subject to deception, sudden changes or possible collapses.
At the same time, the economic activity in any market economy is so fundamentally based on money and credit that, if major disturbances hamper the ability of the financial system to fulfil effectively its basic functions, the whole economic system is affected.
3. When dealing with stability issues, other aspects are problematic too: what are the relevant boundaries of the financial system? Banks should certainly be included, as a fundamental component, but other financial institutions are also increasingly relevant. Financial markets - in a very broad sense, including for instance foreign exchange markets - should also be included, together with the technology and the infrastructures which support market transactions.
The Treaty on European Union or "Maastricht Treaty" explicitly mentions "prudential supervision and financial stability". The latter notion is broader, and embraces prudential supervision as just one tool, albeit a fundamental one, within the overall set of policy functions aimed at achieving financial stability. This brings us to another problematic aspect, since the links between the macroeconomic and microeconomic aspects are particularly complex and sometimes even hard to detect. Bank supervisors all crucially refer to what is called "systemic stability". They pursue this goal mainly through microeconomic tools, focused on the risk-taking activity of individual institutions. Yet, if one looks at any important episode of financial instability or any systemic threat, one finds that macroeconomic factors were almost always at work in determining or amplifying the disturbance. The collapse of the savings and loans in the United States, the problems of the Japanese banking system and the crisis of the Scandinavian banks are among the most relevant examples of the last 20 years. In all these episodes the problem did not lie with a single institution, but with an entire sector of the industry, or with the whole industry, and each time it had a macroeconomic dimension. In addition, in Italy during the 1990s most of the banks in the southern regions ran into difficulties, largely reflecting the critical macroeconomic situation in that part of the country: the economy was not growing, the recession had lasted much longer than in any other country or region, and the quality of local banks' portfolios was deeply affected. When asset values undergo a sharp or prolonged downward trend, very often there is a general cause.
4. The matter becomes even more intricate when we consider how financial stability can be best achieved. We know that it is risk that threatens stability, but at the same time we also know that risk-taking is an essential aspect of banking and finance. Attempts to eliminate risk, for instance through full protection by the public authorities, would suppress the very function of the financial system and would severely hit the very foundations of the economic system as a whole. So, we have to conceive instruments able to deal with something which is both necessary and dangerous at the same time.
In this respect the most difficult task is to draw a line between potential or actual risks threatening the stability of individual institutions or of limited segments of the financial market, on the one hand, and truly systemic risk, on the other. While in the end the latter is the real source of our concern, in pursuing financial stability, and particularly in conducting banking supervision, usually more effort is devoted to monitoring individual institutions and to undertaking preventive action against local, functional instability, rather than in tackling systemic threats.
5. In dealing with all these relatively unsettled issues, three points need to be stressed. First, banks occupy a special place within the financial sector. Even though major innovation and structural changes are altering the way they conduct their business and the set of intermediaries with which they compete, banks are and remain special and form the core of the financial system. Second, preserving confidence is absolutely crucial for the stability of the financial system, even though, again, what is relevant is confidence in the system, not necessarily confidence in every individual institution. Everyone agrees that prudential regulations and control are essential for strengthening confidence in the safety and soundness of the financial system. Third, and finally, failures are necessary. An industry where freedom to access the market is not coupled with effective mechanisms for exit from the market is not a healthy industry. Failures are not an unhealthy phenomenon, and supervisors should not be blamed for letting banks go bust, just as they should not be praised for having no failures within their jurisdiction.
6. Let me now turn to the other part of the title, namely "Eurosystem". The term "Eurosystem" cannot be found in the Treaty. It has been coined to describe the central bank of the euro, just as the Federal Reserve System is the central bank of the US dollar. It consists of 12 central banks, the 11 national central banks (NCBs) of the countries participating in Stage Three of European Monetary Union, and the European Central Bank (ECB).
7. Both outside and inside monetary circles the debate on financial stability and the Eurosystem revolves, among other things, around the question of what should be done, in the pursuit of financial stability, at the national as opposed to the Eurosystem level. When addressing this issue, it is very important to distinguish between two different notions: subsidiarity and decentralisation. There is often confusion between these two principles, because they both concern the distribution of functions between the European and the national level. Subsidiarity relates to the decision on which competences can be effectively left at the state level, and which other functions should be assigned to a "federal" level. The choice between centralisation and decentralisation is concerned with the best way in which to organise the operational fulfilment of a competence that has been entrusted to the "federal" level. Thus the principle of subsidiarity was applied when deciding whether the competence for monetary policy or other aspects of central banking should be raised from the state level to the EU level. The Treaty provides that the principle of decentralisation must be applied as far as possible in the exercise and implementation of this EU competence. This means that very often the federal competence can be implemented or executed at the "periphery", i.e. by NCBs.
The distinction is conceptually important and also difficult to grasp, because NCBs are - and in the field of financial stability and banking supervision this is very relevant - "dual" institutions: they are a component of a European institution and, at the same time, a national institution. They literally have a composite nature. For instance, as supervisors they are fully national institutions, they take autonomous decisions and are fully accountable to national bodies; yet in the monetary policy field they participate in the implementation of decisions taken for the whole euro area and, in that respect, they act as part of the Eurosystem. In the first case subsidiarity applies, in the second decentralisation applies.
8. Finally, in discussing financial stability and the Eurosystem one should bear in mind that central banking has three major components: monetary policy, the payment system and the stability of the banks. No central bank in the world can afford to neglect any of these three functions. In fact, what is now considered by most to be the main function, i.e. monetary policy, has historically come relatively late, and is not that which was at the origin of modern central banks. The origins of modern central banks are rooted in the payment system, and can be traced back to the transition from a commodity to a paper currency. From a historical perspective, the second function is the pursuit of bank stability. It came about as a result of a further development, when the amount of money was increasingly taking the form of liabilities of commercial banks rather than notes issued by the central bank. This made it necessary to be sure of the quality of this commercial bank money, so that attention had to be focused on the quality of the banks' assets and on bank stability. These two functions of central banks underwent fundamental changes at the beginning of the 20th century. Monetary policy came later. I think this aspect should be borne in mind when looking at financial stability and the Eurosystem.
9. In discussing the changing context in which financial stability is pursued today, I shall first look at the elements which affect financial institutions and markets globally. There are four major transitions, which have not yet been completed and the full consequences of which are not always foreseeable: the transition from banking to finance, the transition from national to international finance, the transition from paper intensive processes to information technology, and the transition from command to market-friendly methods in public policies. None of these transitions had significantly got under way three or four decades ago.
10. In continental Europe, financial activity was fundamentally based on banks, which played a dominant role in channelling savings towards investments; in the United States and in the United Kingdom financial markets were much more developed, but even there banks were considered to be the cornerstone of the financial system. This is no longer the case. A growing portion of financial flows are currently channelled through markets, a growing share of financial assets consists in negotiable securities or contracts, a growing proportion of financial activities are performed by non-banks. The distinction between the three traditional branches of finance - banking, securities and insurance - becomes increasingly blurred.
11. The second trend is the shift from national to international finance: the Bretton Woods system was built on the idea that flows of private capital across countries should not be encouraged, that they should primarily consist in official transactions, and that they should be of a limited amount. There was almost an implicit encouragement to set restrictions on international finance. And this was the case in the 1950s and the 1960s. By the end of the 1970s, the two key functions that the Bretton Woods system had entrusted primarily to the official institutions - fixing exchange rates and governing capital flows - had been taken over by the markets. The internationalisation of financial activities complicates the accomplishment of supervisory tasks: in an international financial environment the currency used for transactions, the legal system on which to rely in the settlement of disputes, as well as the exercise of prudential controls, may belong to different countries, or to different jurisdictions. The use of the instruments created for pursuing public policy objectives is immensely complicated by these changes.
12. The transition from paper-based to electronic payments has a number of implications which we are only just beginning to understand. Among other things, it means a transition to real-time finance and to a situation in which the counterpart in the transaction is very often unknown. In a way, this makes it almost impossible to regulate and to monitor the process through the traditional arrangements. The largest market in the world, i.e. the foreign exchange market, is already a market that exists nowhere. It would be impossible to regulate this market simply because the market has no address, there is no organisation running it and there is no membership of the market. Just a few years ago, such a market did exist and the exchange rate was "fixed" there. This shift in technology is also producing immense economies of scale and consequently an enormous push towards industry consolidation.
13. Finally, there is the transition from command methods to market-friendly instruments in public policies. This has taken place in a gradual way. For a long while, supervision was mainly based on authorisation: even specific operations were scrutinised by the responsible authorities, and could be prohibited on largely discretional grounds. Banks had to ask permission to grant a particular loan, or to open a branch. This supervisory framework was laid down after the banking crises of the 1930s, and was aimed at achieving stability by limiting competition and by reducing opportunities to do business and take risks according to market criteria. Things have gradually changed. Today, the market is seen as a mechanism which itself contributes to financial stability. A change in ideas has significantly fed into this shift of emphasis in supervisory controls. However, the ability of market participants to circumvent prohibitions and strict controls by going international, by using new technologies, or by shortening the period of operations and acting between two reporting periods rather freely, was an important, and perhaps even a predominant, factor. The transition was forced by events, perhaps more than by a change in ideas.
14. All these four trends have a global reach. In the changing context in which financial stability is being pursued by the Eurosystem, there are, however, also some specific European factors. In Europe there are two main factors at work: the Single Market and the single currency. They are two separate elements which, as we all know, do not coincide geographically either.
The Single Market has already produced, probably to a fairly large extent, the effects intended in the original project. From the second half of the 1980s onwards, the banking and financial system has gone through a massive process of legislative change and reorganisation in view of the Single Market. All the fundamental legislation concerning financial activity has been redrafted in Brussels and then transposed into national legislation.
15. What is really new, therefore, is the introduction of the single currency. Only now do we see how effective the existence of different currencies was in preserving segmentation in the Single Market for banking and financial services, in spite of all the harmonisation of rules and of the lifting of regulatory barriers to entry in national markets. Basically, even within a single bank, activities were segmented by currency. Financial centres were specialised in financial products denominated in their national currency. The introduction of the euro is bringing about an enormous change in the industry, because it is making currency borders disappear. In many fields, the market was almost instantaneously unified on 4 January last year. In the money market, a single interest rate appeared during the afternoon of that same day. The unification of the bond market is proceeding a little more slowly, and in the equity market even more time will be needed. For many of these instruments, unifying the market without unifying its infrastructure, i.e. the trading platform or the securities settlement systems, is almost impossible.
The shock of the euro certainly affects banks directly, in a variety of ways. If we look at banks as multi-product firms we realise that some services which they supply have already been completely transformed into euro area activities, while other areas of business may never expand beyond local boundaries. The market for corporate finance services, e.g. financial advice and support for large scale mergers and acquisitions, already covers the whole euro area. If a large firm decides to make a take-over bid, it can choose between a limited number of investment banks active in international markets. The market for retail banking will remain national, regional or even sub-regional in many respects.
16. Can it be said that as long as there are no cross-border mergers, there will be no euro-area wide banks? I believe the answer to this question is negative. We have a Single Market in Europe for automobiles, even though no single major cross-border merger has taken place in the industry for decades. In order to assess the area-wide expansion of banking institutions, balance sheet composition is much more relevant. We can probably already identify a limited number of banks which, regardless of the extension of their branch network and of their ownership structure, can be truly be defined as euro area institutions. At the same time, a very large number of banks in Europe also remain locally based with regard to the distribution of counterparts. Furthermore, "local" most often does not even mean national, but regional or provincial. In Italy and Germany, for instance, there are no more than half a dozen banks operating on a national scale, with an aggregate market share which probably amounts to less than 30% of the domestic market. All the other banks are local.
17. Can it be said that with the introduction of the euro, there will no longer be segmentation in the financial market? Here too, the answer is negative. Important segmentation factors are still present. For instance, the rules and practices governing the working of the labour market are a very significant obstacle to further consolidation and restructuring in the banking sector, and may contribute to maintaining some degree of segmentation. Proximity to customers is a crucial element in the service industry. If banks are tied to adopting national contracts, the possibility of significantly altering the features of the service supplied is severely limited. The Deutsche Bank cannot apply in its Italian or Spanish branches the labour contract it uses in Germany. The working of competitive mechanisms to shape a single market may be considerably affected. Segmentation might also be induced by other, remaining differences between national regulation and legislation in Member States. This holds true not only for supervisory matters, but also, for instance, for bankruptcy legislation or for company law, since these both have a substantial influence on lending activities.
Finally, even if those barriers disappeared, there would still be substantial cultural and linguistic barriers. Even in the most far-reaching hypotheses, the shake-up of the industry is not likely to occur quite in the same way as that which we have seen in the United States.
18. Let us now turn to the policy side, i.e. the pursuit of the public interest in financial activity. Authorities have to cope with a number of challenges. Their task is even more complex in Europe, where the introduction of the euro and the specific trends affecting the financial industry are creating extra pressure for change. A variety of public interests are at stake. Besides the pursuit of the goal of financial stability, there is a public interest in preserving or encouraging competition, as well as in ensuring transparency and correct rules of conduct vis-à-vis counterparts and investors. These three types of public interest are generally entrusted to different authorities. Stability is normally entrusted to a supervisor, which may be a central bank or a separate authority; competition policy is normally entrusted to a special authority, whether national or European; transparency codes and rules of conduct are normally entrusted to securities and exchange commissions.
To some extent, it is efficient to separate these assignments, because some of these objectives may occasionally conflict. Historically, bank supervisors have not been the most keen to promote competition and transparency in the banking industry, because competition is bound to result in victims. At the end of the 1960s, the instinct of a bank supervisor was to work to solve problems at supervised institutions before the public and the market realised what was happening. All in all, this approach had some advantages, but it would not be feasible today. This may imply that nowadays there is less scope for conflicts. The more transparency is deemed to be desirable by supervisors, the less forceful is the argument for separate agencies.
19. The introduction of the euro has not altered the pre-existing institutional structures for prudential supervision. Different national arrangements prevail in Member States: the three fundamental segments of the financial industry (banking, insurance and securities) are regulated and supervised by separate agencies in some countries, while in other countries a single agency has been entrusted with responsibilities covering the whole range of institutions and markets. After the move towards a "conglomerate" type of agency in the United Kingdom, it is now rather fashionable to think that this may be the right formula. The formula, however, is still untested.
20. In the field of banking supervision, the entry onto the scene of the Eurosystem had to be considered. In fact in the Delors Committee, which drew up the plan for European Monetary Union, there were long discussions about the whether banking supervision should be included among the activities to be entrusted to what came to be called the "European System of Central Banks" (ESCB). It was decided not to entrust the ESCB with the task of supervision, and as a result a dual separation was introduced into the institutional framework: first a functional separation, i.e. between central banking and banking supervision and, second, a geographical separation between the jurisdiction of central banking (which is the euro area), and the jurisdiction of banking supervision (which is confined to the domestically chartered institutions).
The first type of separation is not uncommon, although it is not the approach prevailing in the 11 countries participating in the EMU. More than half of the 11 national central banks have supervisory responsibilities in their respective countries, while the functions are separated in other countries. It is impossible to say whether one of the two solutions is superior. Banking supervision has functioned well in each of the two systems, while there have been some unwelcome incidents in both frameworks. What is new is the second type of separation; the "geographical separation". Indeed, there is now no other example of a single currency area where banking supervision is not performed at an area-wide level. In the past, the significant exception was the German system where, during the 1950s, supervisory competences were entrusted to the "Länder"; however, this set-up was quickly abandoned and supervision was transferred to the federal level, although not entirely to the Deutsche Bundesbank.
Thus, the original features of the institutional framework for financial stability in the euro area constitute a special challenge, since all the mechanisms for the smooth functioning of the new system must be in place, and there is no previous experience which can provide guidance.
21. The IMF and the group of ECB-watchers which has been set up at the Centre for Economic Policy Research have raised some doubts about the effectiveness of the arrangements for prudential supervision and financial stability in the Eurosystem.
Most of the criticisms have been related to the lender of last resort issue. This is not surprising, since the supply of emergency liquidity lies precisely at the intersection of the three functions of central banking. It relates to monetary policy, because it creates liquidity; it also concerns the payment system, because most of the time the emergency liquidity is needed for closing the payment process; and finally it relates to banking supervision, since it requires an assessment of the solvency of the ailing institution. In a way, this is the classic case in which central bank intervention may be needed to ensure bank stability. Given the geographical separation I have just referred to, the issue naturally arose of the level at which this function would be performed, if necessary.
This criticism was used to cast doubt on the overall adequacy of the institutional arrangements in the Eurosystem. I maintain the view that the European legislation, including the Treaty and the relevant Directives, is neither incomplete nor contradictory. For every bank the supervisor is identified: every bank is licensed in one country, and every home country has a well-defined home supervisor. The co-operation among supervisors is provided for by the relevant Directives. A full framework for central banking is set out in the Treaty and in the Statute of the European System of Central Banks and of the European Central Bank (the Statute of the ESCB). The Governing Council of the Eurosystem specifically dealt with the problem of "emergency liquidity assistance" in the first half of 1999. The necessary links between the responsibilities of national central banks for emergency liquidity assistance and the possible impact of such operations on the management of liquidity in the euro area have been established.
22. The system has its own specific features, but the originality of its design does not mean that it is either incomplete or unworkable. It is certainly necessary to make this institutional setting work in a way suitable to accommodate the new and fast changing situation in the marketplace. First, functional separation calls for extensive co-operation among bank supervisors, since developments taking place in the industry increasingly have an area-wide (or Single Market-wide) dimension. Second, the Eurosystem needs to enhance its instruments for monitoring the relevant developments in the banking industry, as is the case for any central bank, even when it is not in charge of banking supervision.
These are thus the two courses of action which will have to be encouraged in the years to come, in line with the evolution of the European banking system. This is perfectly feasible within the existing legal and institutional structure. It will take years, given that the whole process has only just started. However, there is no other, easier, way forward.
Reproduction is permitted provided that the source is acknowledged.