A quarterly magazine of the IMF

December 1998, Volume 35, Number 4

The introduction of the euro is expected to spur the creation of pan-European financial institutions and markets, with considerable benefits for consumers and investors. Challenges remain within the euro area institutional framework for the management of systemic risk and financial crises.

Ensuring financial stability within the European Economic and Monetary Union (EMU) will be challenging in the early years, in part because there are a number of aspects of EMU that might increase the potential for systemic events. First, there is the possibility that TARGET (the Trans-European Automated Real-Time Gross Settlement Express Transfer system)—the settlement system slated to take effect on January 1, 1999 that will link the real-time gross settlement systems of European Union countries—will be used less extensively than expected and might therefore not reduce systemic risk as much as has been anticipated. Second, as new pan-European financial markets emerge, the growth of cross-border unsecured interbank lending could increase the risk of contagion, at least until an EMU-wide repo (repurchase agreement) market is created and use of secured (collateralized) interbank credit lines becomes more widespread. Third, the introduction of the euro is likely to encourage further bank restructuring and consolidation, but in an environment where it may be difficult to close banks and to reduce costs through downsizing. Inefficient and unprofitable institutions may therefore continue to operate, engaging in increasingly risky activities.

Box 1: Prudential supervision and financial stability The European System of Central Banks Statute and the Maastricht Treaty assign to the European System of Central Banks only limited functions related to prudential supervision and the stability of the financial system, but it does have an explicit role in promoting the smooth functioning of the payment system. The flow of supervisory information between the ECB and the competent authorities is also regulated by the BCCI Directive (Directive 96/25/EC of June 29, 1995). The European Monetary Institute's 1997 annual report provided some clarification on how these provisions will be implemented in EMU. In EU legislation relating to prudential supervision of credit institutions and the stability of the financial system, the ECB has the option of playing an advisory role and must be consulted only on draft EU and national legislation that influences the stability of financial institutions and markets. To ensure effective interaction between the European System of Central Banks and national supervisory authorities, the Maastricht Treaty stipulates that the European System of Central Banks "shall contribute to the smooth conduct of policies pursued by the competent authorities relating to the prudential supervision of credit institutions and the stability of the financial system." The European System of Central Banks will not systematically receive supervisory information; its requests for such information will be considered by national banking supervisors, who will inform it, on a case-by-case basis, in the event of a banking crisis with systemic implications. The EU Council of Ministers, upon the initiative of the European Commission, may assign specific tasks to the ECB related to prudential supervision; however, the European Monetary Institute's annual report indicates that any transfer of supervisory powers from national authorities to the central bank is considered premature at this stage.

The possibility of heightened systemic risk may not be apparent in the early days of EMU because market integration and bank restructuring may occur slowly. The limited number of cross-border mergers of European banks that have taken place thus far, the gradual increase of competitive pressures in the retail sector, widespread public ownership of banks, and the underdeveloped state of capital markets in many EMU countries may provide some EMU countries with more time for restructuring banking systems. Decentralized arrangements for market surveillance and crisis management (based on home country supervision, for example) may be allowed to continue temporarily, providing some time for adjustment. Eventually, however, pan-European capital markets and banking systems are likely to develop, creating a need for centralized mechanisms for financial surveillance, systemic risk management, and crisis resolution. Institutional arrangements in advanced countries indicate that the central bank may be a natural place to centralize some of these functions, but such centralization would require simplification of the current division of responsibilities—as mandated by the Maastricht Treaty and European Union legislation—between the European Central Bank (ECB), the national central banks, national supervisors, and national treasuries (Box 1).

The current framework

Against this background, the thinking and planning about crisis management are still evolving. Whereas some understanding is likely to be reached before EMU takes effect, important decisions have yet to be made about how EMU countries would resolve a bank liquidity crisis that occurred, for example, at the fine line between monetary policy operations and liquidity support for systemically important private financial institutions. The lack of clarity on how problems will be dealt with reflects, in part, the "narrow" price stability mandate of the ECB spelled out in the Maastricht Treaty. The mandate of the ECB calls for it to focus on monetary policy and gives it only a limited, peripheral role in banking supervision and no responsibility for providing liquidity support to individual financial institutions. In accordance with the limited role the treaty envisions for the ECB, the European Monetary Institute—the ECB's precursor, which was dissolved on June 1, 1998—organized its work so that a clear separation has been maintained between monetary policy operations and the provision of liquidity for reasons not having to do with the conduct of monetary policy. No institution in EMU has been designated as a lender of last resort; thus, no central institution holds the responsibility for providing, or coordinating the provision of, liquidity in a crisis.

It is unclear how a bank crisis would be handled under the current institutional framework (which is composed of the Maastricht Treaty, the Statute of the European System of Central Banks, and the regulations and guidelines issued by the European Monetary Institute), especially if the crisis were to involve a pan-European bank for which several countries shared supervisory and regulatory responsibilities. The main issue is whether the European System of Central Banks or the national central banks have effective mechanisms and understandings in place for taking action if a particular financial institution is having difficulties in financing payment instructions sent either across TARGET for real-time settlement or across one of the alternative netting payment systems. It appears that EU supervisors have reached several understandings about how to deal with cross-border crises and that discussions about the lender-of-last-resort function are under way, but no final decisions have been taken.

In fact, practitioners and academics do not agree on a conceptual framework for dealing with the immediate consequences of a banking crisis. Some observers have argued that, to avoid moral hazard, central banks should use only open market operations to deal with a liquidity crisis. Others have argued that when a systemic event occurs in which there is little or no doubt about solvency—as was the case with the Bank of New York's computer failure in 1985—the central bank should have the possibility of discounting assets (for example, loans or commercial paper) other than eligible collateral. There is a diversity of experience and practice among the major central banks. For example, in the United Kingdom and the United States, as well as in some other advanced countries, central banks have considerable discretion in deciding what kind of collateral to accept in exceptional circumstances to provide liquidity to the banking system. By contrast, Germany's Bundesbank has almost no discretion about what kind of collateral it can accept, and there has been no instance in which uncollateralized intervention was necessary.

The German model

The German system is an important benchmark for examining how crisis management might take place within EMU, because the mandate of the European System of Central Banks is similar in many respects to the Bundesbank's. The Bundesbank—like the ECB—has no explicit responsibility for safeguarding the stability of the financial system and does not act as a lender of last resort. Indeed, the German framework for dealing with crises seems to be constructed so as to avoid a role for the Bundesbank in providing funds in rescue operations. The system, in effect, has three lines of defense: (1) supervision and regulation by an independent body; (2) short-term liquidity assistance from the Liquidity Consortium Bank (a specialized institution—30 percent of whose capital is held by the Bundesbank, with the remainder held by all categories of German banks—that ensures the timely settlement of domestic and external interbank payments), combined with brokered market solutions; and (3) deposit insurance and, if necessary, injections of public funds. In practice, thanks to close cooperation with the independent supervisory authority and the Bundesbank, the Liquidity Consortium Bank has been able to identify solvent institutions to which short-term liquidity assistance should be provided. This close cooperation has also allowed the Bundesbank to be involved in resolving problems by encouraging strong banks with ample liquidity to purchase illiquid but sound assets from troubled institutions in need of liquidity. Deposit insurance and public funds have been used to deal with insolvent institutions.

Is the German model applicable to EMU?

There are a number of reasons why such a framework (three lines of defense, with no central bank funds) might not be applicable in the event of a crisis within EMU. First, no institution corresponding to the Liquidity Consortium Bank exists in other EMU countries, nor is one planned at the EMU level. Second, even if such institutions existed, they would be inadequate in relation to the size and the cross-border systemic implications of a liquidity crisis involving a major pan-European banking group unless they were endowed with considerable resources and had much greater access to supervisory information than national supervisors are likely to provide to the ECB. Third, the current agreement about sharing information between the ECB and the national supervisors—which can be summarized by the formula of no real obligation, no real obstacle, and some understanding—would probably not give the central bank the same authority to broker a solution to a banking crisis at the EMU level as the Bundesbank has at the national level. The ECB could play such a role only if it were perceived to have the same access to supervisory information at the EMU level that the Bundesbank has at the national level, or if it had the authority to inspect counterparties in order to assess their creditworthiness. Fourth, the German system has worked well in an environment characterized by relatively underdeveloped capital markets and a large share of public ownership in the banking system, in which a crisis would unfold in "slow motion," compared with the speed with which a crisis would probably spread through EMU-wide capital markets and banking systems. Finally, in an integrated EMU banking system with several EMU-wide institutions, the use of deposit insurance schemes and treasury funds would add to the time needed to determine how financial responsibilities should be shared among national authorities and could delay a decision about how to deal with a problem bank.

Box 2: Lender-of-last-resort operations The question of which institutions in EMU will have the authority and responsibility to act as lenders of last resort during liquidity crises is still ambiguous. Whether or not the Governing Council of the ECB will choose to maintain this ambiguity remains to be seen. In a local liquidity crisis (that is, one affecting a large institution located in an EMU country), the key issue is whether the national central banks will be able, without authorization from the ECB, to provide liquidity support to troubled institutions. Although the national central banks have scope for such operations, the ECB's Governing Council could, by a qualified majority vote, prohibit them from purchasing ineligible collateral from illiquid institutions. Indirect means of assisting banks experiencing liquidity difficulties are also open to national central banks—for example, they could swap some of the liquid assets in their balance sheet for some of the troubled banks' illiquid assets and assume the credit risk on the latter, or they could guarantee the troubled institutions—but the Governing Council may issue guidelines prohibiting such on- and off-balance-sheet operations or specify that they require prior authorization. If guidelines are so strict as to prevent the national central banks from providing direct or indirect liquidity assistance to troubled banks, they may be able to provide assistance by opening up the definition of eligible Tier II collateral to include a broader range of assets, but this would require approval by the ECB's Governing Council and could delay resolution of a crisis. In the event of a general liquidity crisis that would affect the entire EMU—for example, gridlock in an EMU payment system or TARGET—the ECB may need to provide liquidity to avert a systemwide crisis. Collateralized intraday credit and extraordinary open market operations may be sufficient to inject the necessary funds in some instances, but in others these measures may not suffice, owing to a lack of eligible collateral. At such times, the risk of a systemic crisis would be high, forcing the European System of Central Banks to accept ineligible paper as collateral for payment system overdrafts or open market operations. For example, when a general liquidity crisis occurred in the United States during the stock market crash of October 1987, the U.S. Federal Reserve System gave banks unrestricted access to its discount window so that they could keep their credit lines to brokers and securities houses open.

Under the current institutional framework, considerable uncertainty also remains about the scope that national central banks might have for providing emergency liquidity assistance to troubled banks (Box 2). In all relevant crisis situations involving pan-European markets and institutions, the ECB appears to have to decide either to inject extra funds into the system (in the event of a general liquidity crisis) or to allow national central banks to intervene in a local liquidity crisis. Such intervention would require intimate knowledge of counterparty institutions, however. Supervisory information would be necessary to assess the credit risk that such operations would involve in the event that the European System of Central Banks were forced to accept ineligible collateral. Moreover, the ECB would be unable to rely on market assessments to distinguish between a liquidity crisis and a solvency crisis. (In most liquidity crises, the markets would question the solvency of the institution in difficulty because a solvent institution would have been able to borrow from the money markets to meet its liquidity needs.)

Even if the ECB's involvement in the management of liquidity crises is to be minimal—possibly only to authorize, or to refuse to authorize, the national central banks to act as lenders of last resort—the current arrangements between national supervisors and the ECB governing the exchange of supervisory information seem to be too limited to allow well-informed decisions during a fast-breaking crisis. An arrangement in which the ECB does not have independent access to supervisory information on a systematic basis and in which banking supervisors will inform the European System of Central Banks "on a case-by-case basis should a banking crisis arise" makes the ECB entirely dependent on national supervisory authorities for the information it needs to make sound decisions. In addition, the new framework is not clear about the understandings between the ECB, the 11 national central banks, the 11 supervisory authorities, and, possibly, the 11 treasuries in EMU. In the event of a crisis involving a European banking group, clarity and transparency about the sharing of information would greatly facilitate coordination and management during the early stages of a financial problem or crisis.

Constructive ambiguity?

The limited agreement on information sharing probably reflects the fact that no clear lender-of-last-resort function has been attributed to the European System of Central Banks and that, at present, there does not seem to be a fully worked-out framework for crisis management in EMU. Current understandings indicate that crises might have to be managed through ad hoc arrangements to do whatever is necessary to avert systemic problems. The idea may be that, in the event of a crisis, a national central bank or a national authority would find a way to provide liquidity support, and then central banks and supervisors would quietly pursue longer-term solutions, including finding buyers. (The role that national treasuries would play in crisis management in EMU is another open question. Whereas treasurers may ultimately provide the funds for bank rescues, it is unlikely that they could be the immediate source of liquidity.) Whereas this lack of transparency may be interpreted as "constructive ambiguity" intended to reduce moral hazard, the current understandings and arrangements within EMU would need significant further development before they would be workable in an environment in which speed is increasingly critical in the handling of financial and systemic crises. Some European authorities believe, however, that, once established, such arrangements may well not be disclosed to the general public because to do so would increase moral hazard.

The current decentralized approach does not assign responsibility for supervising pan-European banks or for ensuring EMU-wide financial market stability either to national central banks or to national governments. As European banking groups emerge, the questions of whether national central banks could adequately assess the risks of contagion and whether the home country central bank of a bank in difficulty could be easily identified will become increasingly relevant. In addition, decentralized lender-of-last-resort policies may create an uneven playing field and introduce different levels of moral hazard across EMU. At the same time, the ECB will be at the center of European financial markets without the full set of tools necessary for independently assessing the creditworthiness of counterparties or a framework for rapidly providing support to solvent but illiquid institutions. This is not likely to be sustainable, and the ECB may be forced to assume a leading and coordinating role in crisis management and banking supervision.

Alessandro Prati is an Economist in the Capital Markets and Financial Studies Division of the IMF's Research Department. Garry J. Schinasi is Chief of the Capital Markets and Financial Studies Division of the IMF's Research Department.