A joint CEPR conference with the Irving Fisher Society was held in Frankfurt on 9/10 June on `What Monetary Policy for the European Central Bank?'. The conference was organized by Peter Bofinger (Universität Würzburg and CEPR) and Richard Portes (CEPR and London Business School), and received financial support from Dresdner Bank, the Fritz Thyssen Foundation and the Deutscher Sparkassen- und Giroverband, Bonn.

In `How Can the ECB Become Credible?', Alex Cukierman (Tel Aviv University) examined the effect of independence on credibility. Empirical studies show that inflation is lower the higher a central bank's degree of legal independence. But legal independence alone will not assure credibility as legal and actual independence are seldom perfectly correlated. Discussing possible conflicts between financial and price stability, he concluded that where a central bank prefers to achieve both there is no long-run trade-off. But, due to free rider problems, guidelines are necessary for the distribution of seignorage among countries. As the establishment of credibility is enhanced by fiscal constraints, the Maastricht convergence criteria for fiscal policy point in the right direction. The inflation and interest rate criteria are, however, of secondary importance because in the long-run they are determined by the stance of fiscal and monetary policy. As credibility is fostered by nominal targets, he recommended an ECB rule that combines a base target and an inflation target.

Lorenzo Bini-Smaghi (European Monetary Institute) commented that bank independence cannot be assessed only on objective criteria. The inflation criterion is important since credibility would be jeopardized from the start if the primary objective of maintaining price stability is not achieved. Financial markets must believe in the process, and hence strong political support is essential. Accountability is also vital. Manfred Neumann (Universität Bonn and CEPR) pointed out that even without legal accountability the Bundesbank is accountable. Otmar Issing (Deutsche Bundesbank and CEPR) agreed: since the law could be changed by a simple majority of parliament, the Bundesbank has to explain its policies. Stefan Collignon (Association for the Monetary Union of Europe) discussed the issue of becoming versus being credible. If the ECB takes over the Bundesbank model than it is credible, but how you get from here to there? Leonardo Leiderman (Tel Aviv University and CEPR) thought accountability not necessary for price stability since any policy-maker is accountable. This is a dynamic game with three parties: the central bank, the public and the government. Mervyn King (Bank of England and CEPR) referred to the problem that one country might be affected by a severe recession. Political legitimacy would be questioned and politicians may clamour for exit. This divorce danger should be discussed now.

In `The Strategy of Monetary Targeting: Can The German Experience Provide a Model for the ECB?', Manfred Neumann and Jürgen von Hagen (Universität Mannheim and CEPR) analysed the Bundesbank´s monetary policy and its transferability to the ECB. The Bundesbank announces a M3 monetary target range derived from the equation of exchange. Combining the price gap approach and the derivation of the monetary target shows that the actual inflation rate will exceed the medium-term targeted inflation rate in times of business cycle upswings, when actual output lies above potential output and actual velocity lies below its trend value. On this basis the authors discussed the actual performance with the help of estimated reaction functions for the period 1979-94. The two undershootings and five overshootings during that period happened in times of strong exchange rate movements. In addition, the Bundesbank deviated from the monetary target to fight excess inflation or to smooth the business cycle. The estimation results also show that the mid-point target determines the long-run growth of money supply, but it has little predictive power for the money growth in any particular year, and that the Bundesbank tolerates on average an inflation bias. Nevertheless, the authors recommended the concept of monetary targeting for the ECB: it anchors inflationary expectations, provides sufficient flexibility to react to unanticipated shocks, it can easily be understood and it can be used as a shield against demand for easy money.

Otmar Issing agreed with the description of Bundesbank monetary policy and its application to the ECB. But he warned of the latter's potential lack of credibility since it is a new institution. To overcome this, it should take over as much credibility from its predecessors as possible, including a clear concept of monetary targeting. Issing disapproved of the concept of direct inflation targeting: non-monetary causes of inflation in the short-run would endanger the ECB´s credibility. Leiderman remarked that the M3 target does not seem to be a very serious target, because of frequent deviations; for the ECB he fears a serious credibility problem as the cost of target deviations. Ernst-Moritz Lipp (Dresdner Bank) stressed the transition phase between fixing parities and the conversion to EMU: there may be major shifts of assets, which would make monetary aggregates an unreliable basis. Bini-Smaghi asked how the Bundesbank could remain credible in spite of missing half of its monetary targets. The ECB is not in the same position but will be under very strong scrutiny right from the beginning. Cukierman suggested combining monetary and inflation targets, costless as long as both variables move in the same direction. For contradictory developments a rule has to be formulated to stick to the low end of the target. King said comparing central bank council meetings of the Bundesbank, the Fed and the Bank of England would reveal a lot of similarities despite the different monetary concepts.

Leonardo Leiderman presented a joint paper with Lars Svensson on `Inflation Targets', stressing the need for nominal anchors for price-setters and the disappointment with monetary and exchange rate targeting. The main characteristics of inflation targets are an explicit quantitative target and the absence of intermediate targets. Their main advantage is their high transparency and the crystallization of price stability as the main goal of monetary policy. On the other hand, the controllability of inflation targets is less than of monetary targets, so a trade-off exists between transparency and controllability. He showed that so far the performance of this approach is positive, but there has not been a test of the inflation target regime over the business cycle, so it is too early to provide a complete judgement.

Mervyn King said that the shift towards inflation targeting has often been undertaken to constrain discretionary behaviour and give policy-makers the right incentives. He agreed that the main argument for inflation targets is the clear obligation to promote price stability, but suggested that they promise more than can be achieved. Monetary targeting might be better since the central bank is able to control monetary aggregates and thus can be held responsible for failing. Cukierman said that there will always be an inflationary bias in expectations because of public uncertainty in judging whether inflation target preannouncements are a commitment of policy-makers or just a forecast. Frederic Mishkin (Federal Reserve Bank of New York) pointed out that information about monetary targets is available earlier than information about inflation, so central banks are able to adjust their policies early. Georg Rich (Swiss National Bank) emphasized the important role of transparency of inflation targeting to the public. Targeting forces the central bank to tell the public not only what it wants to achieve but also how to achieve it. This raises transparency, explaining the necessities of monetary policy in order to achieve its objectives. Bofinger said that a theoretical framework for targeting is lacking so its transmission process is a black box.

In `The Term Structure of Interest Rates and its Role in Monetary Policy for the ECB', Arturo Estrella (Federal Reserve Bank of New York) and Frederic Mishkin presented evidence on the applicability of the term structure within a core-set of European countries. Their econometric results suggest European central banks can largely control the term structure through interest rate policy instruments. Hence, the spread between the long- and short-term interest rates contains valuable information about the monetary stance. The spread also has predictive power for both real activity and inflation, with horizons of one to two years for the former and longer for the latter. The term structure does play a useful role as an indicator of monetary policy tightness for the ECB, and of future real activity and inflation in Europe. But the ECB should not depend on it as its only indicator, and, according to the Lucas critique, a policy change, such as establishing an ECB, is likely to alter the structure of financial markets.

Eduard Bomhoff (Nijenrode University) stressed that the ECB should use the term structure as one of many indicators for monetary policy. He mentioned three problems: first, testing monetary policy within the fixed exchange rate system of the EMS; second, the lack of an underlying theory of why the term structure is a good indicator for monetary policy; and third, how to differentiate interest rate volatility. Warren Oliver (S G Warburg) stressed the need to analyse aggregate interest rate data and compare it with other indicators like money, as well as the high level of sensivity of the results. Jürgen Kröger (European Commission) asked why France is part of the core countries although direct credit controls imposed in the mid-1980s make the term structure less reliable. Issing explained why the Bundesbank resisted using the German term structure as policy advice, because it is partly influenced by US or world interest rates and not by the Bundesbank itself. Cukierman ascribed importance to the way the public interprets a central bank policy change: lowering short-term interest rates can either be viewed as a permanent or a temporary loosening of monetary policy. For Leiderman, if the term structure is only useful for obtaining information, a less restrictive approach (for example, through stock market data) should be taken. The predictive power of the term structure is, according to Neumann, rather poor: probabilities of less than 50 per cent are not significant enough. Instead of using the term structure for forecasting purposes, Bofinger suggested employing a real short-term rate, because its effects on the real economy are much clearer.

In `Should the ECB Adopt a Divisia Monetary Aggregate?', Peter Spencer (Kleinwort Benson) showed the strong theoretical necessity for monetary targeting because of ECB independence and EU federalism. For the selection of a monetary aggregate as an intermediate target, he listed four criteria: admissibility, stability, simplicity, and controllability. An aggregate can be admissible if the marginal rates of substitution between any two components of the aggregate are independent of the values of any variable outside the aggregate. Using a nonparametric test to check whether a subset of national financial assets forms a weakly separable group in the potential EMU countries, he found admissibility only for German M3 and UK M4 components. He concluded, however, that this ensures admissibility for the whole EMU bloc because of their dominant size.

Karl-Heinz Tödter (Deutsche Bundesbank) showed that the identification of an admissible asset group is not so easy. The controversial empirical results in the literature and the somewhat questionable conclusions give evidence for the low statistical power of the nonparametric tests. Mike Dueker (Federal Reserve Bank of St Louis) asked if an admissible flow of monetary services is of relevance for an intermediate target especially if there are no exact rates of return for the various components available. Cukierman argued that the main interest of central bankers should lie in the explanatory power of the monetary aggregate for the price level. The admissibility of an asset collection in terms of a representative consumer utility function or the violation of a nonparametric test has no meaning for this central point. It in no way answers the question of the best nominal aggregate to achieve price stability.

In `Which TARGET for Monetary Policy in Stage Three? Issues in the Shaping of the European Payment System', Curzio Gianinni and Carlo Monticelli (Banca d'Italia) examined the strong link between the payment system and monetary policy. The establishment of an integrated market for central bank money across member countries is an indispensable requirement for the conduct of the single monetary policy in Stage III. This, however, can only be attained if it is supported by the integration of national payment systems. To ensure the minimal degree of integration and harmonization, the European Monetary Institute has just designed a system called TARGET (Trans European Automated Real-Time Gross Settlement Express Transfer). Implementing TARGET successfully requires the active support of central banks, a clear perception by market participants of the benefits, and consistency with the instruments and procedures of monetary policy. The key issue is how to satisfy the increasing liquidity needs which such systems are likely to imply.

Eduard Hochreiter (Österreichische Nationalbank) emphasized the need to establish the integrated payment system at the beginning of Stage III: if that case, different interest rates will be possible because the arbitrage process will not work. He assumed that some fragmentation in the financial markets will remain because of specific knowledge of local banks and local preferences. Cukierman questioned whether an integration of international capital markets is different from national integration only in a quantitative or also in a qualitative way. Collignon recommended making payments in ecus when TARGET is implemented. Neumann suggested the potential disadvantage that intraday operations may induce higher volatility on money markets. Klaus Friedrich (Dresdner Bank) remarked that there is a danger of too much regulation.

Georg Rich presented `Does the ECB need Minimum Reserves?', focusing on the monetary policy aspects of reserve requirements. They may strengthen the central banks' ability to control the money supply but they can also cause several distortions. He concluded that they do not necessarily strengthen central bank control of either the money supply or the ultimate target variables of monetary policy. The reason banks insist on them is to faciliatate the operational aspects of monetary policy. Ample commercial bank holdings of reserves enable central banks to limit their role to a lender of truly last resort. For the ECB, he came to the conclusion that modest reserve requirements might be needed in order to facilitate ECB operations.

Martin Klein (Universität Halle and CEPR) agreed that reserve requirements should belong in the ECB's toolkit, emphasizing that they reduce the probability of central banks turning into lenders of last resort. He also suggested that they might have positive effects for industrial policy: since small banks hold excess reserves while reserve requirements tend to be binding on large banks, the latter could compensate for the potential disadvantages of small banks in the larger European market. Spencer mentioned that the UK succeeded in dispensing with reserve requirements, and significant moral hazard problems did not arise. Bini-Smaghi questioned the applicability of the UK system to Europe. Axel Siedenberg (Deutsche Bank Research) maintained that there are no convincing reasons for required reserves, and if they are introduced, that they should be interest-bearing. Volbert Alexander (Universität Giessen) emphasized the close connection with the dis cussion concerning the optimal European payments system: knowledge of the monetary policy instruments available to the central bank is an essential prereq uisite for this issue. Jacques Mélitz (INSEE, Paris, and CEPR) stressed the reserve ratios trade-off. They should be low in order to avoid distortions, but, at the same time, they must provide adequate buffer stocks. They do not solve the moral hazard problem: even in the case of existing reserve requirements, the central bank might be forced to become a lender of last resort.

In the last paper, Jürgen Jerger and Lukas Menkhoff (Universität Freiburg) investigated `Refinancing Policy Options of the ECB'. They evaluated the available instruments with respect to the goals of efficiency, competitive neutrality and subsidiarity, and used these results to formulate the outlines for a ECB refinancing policy concept. The most important element of their proposal was a recommendation for standing facilities. To ensure the necessary flexibility in controlling unexpected money market situations, some open market operations are helpful. In addition, they recommended an unlimited marginal facility to accommodate an upward liquidity demand shock. Limiting the ECB to these three instruments would contribute to an understandable and credible stance for European monetary policy.

Mélitz said that the question of how to control the money stock is not investigated in the paper: there are different techniques of intervention in Europe and these need to be unified. Once EMU has taken place, decentralization is not sensible. The subsequent discussion focused largely on interpretation of subsidiarity with Jerger and Menkhoff discussing operational subsidiarity and subsidiarity within the banking sector. Portes and Bini-Smaghi stated that the principle of subsidiarity is not correctly interpreted by the authors in this context. Detlev Rahmsdorf (Landeszentralbank in Hesse) stressed the role of commercial banks as well as the role of a volume tender, and favoured decentralization. Neumann noted that decisions at the local level and the interbank market are not investigated in the paper.

becoming a book?