The European Community's progress towards economic and monetary union has provoked much debate over the proposed common central bank's relationship to existing national central banks and the Community's other institutions, the international role of the ecu, and the implications of monetary union for the coordination of member countries' fiscal policies. Many of these issues were addressed in papers presented at a CEPR joint conference with the Center for German and European Studies at Georgetown University and the International Monetary Fund. This conference, on `Establishing a Central Bank for Europe', which was held in Washington DC on 1/2 May, focused on the theoretical and institutional issues raised by the Draft Statute of the European System of Central Banks and the European Central Bank (ECB), which was sent to EC ministers in November 1990. Many of the papers related proposed designs for the ECB to the US experience. The conference was organized by Bradley Billings and Matthew Canzoneri, Professors of Economics at Georgetown University, Vittorio Grilli, a Research Fellow in CEPR's International Macroeconomics programme, and Paul Masson, Advisor at the Research Department of the IMF. The conference took place as part of CEPR's research programme on `Financial and Monetary Integration in Europe', supported by a grant from the Commission of the European Communities under its SPES programme. Further funding was provided by the Ford and Alfred P Sloan Foundations as part of their support for CEPR's International Macroeconomics programme, and by a grant from the government of the Federal Republic of Germany to the Center for German and European Studies at Georgetown.

Institutional Design

The Draft Statute leaves many questions concerning the exact form and role of the ECB unanswered. In `Designing a Central Bank for Europe: A Cautionary Tale from the Early Years of the Federal Reserve System', Barry Eichengreen (University of California at Berkeley and CEPR) cautioned against potential instabilities during Stage II of the Delors plan, when national central banks will continue to issue their own currencies and intervene domestically but exchange rates will be fixed, so money supplies and interest rates will be determined by the market. Under the similar US Federal Reserve System of the 1920s and 1930s, the unclear definitions of the roles of the Board of Governors and the Reserve Banks from the outset led to major delays in the implementation of policy and in some cases to the adoption of wrong policies altogether.

Eichengreen used time-series data for 1919-38 to test for structural shifts and effects of institutional changes on policy. The results of tests with changes in Fed holdings of government securities and bankers' acceptances as the dependent variable suggested that changes to institutional forms do affect policy; while changes in unborrowed reserves did not reflect such influences. For a simple analytical model of the interaction between the New York and Chicago Reserve Banks, where each had a private goal of attaining a target level of earnings and a public goal of stabilizing output, Eichengreen showed that non- cooperative Reserve Banks will not stabilize enough, but the Board of Governors may force the banks to cooperate to supply the correct amount of stabilization. Eichengreen presented a series of case-studies to demonstrate the problems that arose when the two Reserve Banks conducted contradictory polices or their objectives differed from the Board's. He concluded that if the roles of national central banks and the ECB remain unclear, their struggle to determine which controls policy may lead to the union's dissolution.

Bennett McCallum (Carnegie Mellon University) found the case- studies compelling, but he expressed doubts about the econometric evidence and questioned the implications of the structural shifts. Eugene White (Rutgers University) noted the difficulties of determining whether the Fed's early problems were due to its institutional structure or to wrong policy choices.

Alberto Alesina (Harvard University and CEPR) and Vittorio Grilli (Birkbeck College, London, and CEPR) presented their paper, `The European Central Bank: Reshaping Monetary Politics in Europe'. They examined the Draft Statute's proposals in terms of Rogoff's `conservative central banker' analysis, assuming initially the Community's complete political integration with a common legislature and executive. They found that an independent Central Bank set up by a Community legislature which appointed a Governor more `inflation averse' than the European median voter would be as independent of national and Community political institutions as the Bundesbank is now. For the case where the Community's economic and political integration is incomplete and specifically member countries' preferences over the trade-off between low inflation and output stabilization differ they found that peripheral countries have more to lose in terms of stabilization than those in the core, but they stand to gain the most in terms of credibility.

Finally, the proposed systems of appointment to and and voting rules of the ECB's Board based on votes weighted by country may produce decisions far from the preferences of the European median voter. At present, the left would be grossly under-represented relative to its shares of votes in the member countries' last national elections.

Peter Bofinger (Landeszentralbank Stuttgart and CEPR) asserted that the proposed structure of the ECB will more closely resemble that of the Bundesbank immediately following World War II, when the banks of the Länder retained greater independence than they do today. Luigi Spaventa (Università degli Studi di Roma, `La Sapienza', and CEPR) noted that uncertainty remains whether the ECB or the finance ministers will control the Community's external exchange rate, and he suggested extending the analysis to assess the ECB's effective independence by incorporating exchange rate shocks vis-à-vis the dollar.

Central banks provide the financial sector with an efficient and stable payments mechanism and also control inflation; but the Draft Statute adopts a `narrow' view by specifying monetary stability as its sole objective. In a paper on `The ECB: A Bank or a Monetary Policy Rule', David Folkerts-Landau (IMF) and Peter Garber (Brown University) investigated the effects of this approach on the development of European financial markets. Increasing securitization frees up bank reserves and increases bank credit to issuers of less liquid securities, thus making them more efficient than those dominated by bank intermediation. The authors argued, however, that increased securitization raises the likelihood of liquidity crises, since the failure of an agent in a securitized market due to a sudden price collapse or some exogenous factor may generate a systemic failure if banks withdraw their lines of credit en masse. If most short-term funding remains on bank balance sheets, in contrast, financial markets create few liquid claims and fewer liquidity problems. Moreover, the central bank's role as `lender of last resort' is essential to maintain liquidity. Folkerts- Landau and Garber concluded that as well as preserving monetary stability the ECB should provide traditional banking functions to enable financial markets to develop into highly liquid, securitized markets. The ECB is better able to perform this role than the national central banks. Their open market operations would interfere with the ECB's role of monetary stabilization, and they would also find it more difficult to resist local banks' calls for assistance.

John Pattison (Canadian Imperial Bank of Commerce) argued that concern over liquidity may be exaggerated since European banks unlike their US counterparts have extensive branch systems which provide access to large pools of liquidity. Graham Bishop (Salomon Brothers) argued that there is no need for the ECB to fulfil a supervisory/regulatory role, since the prices of assets on securitized markets are well known. Richard Sweeney (Georgetown University) noted that expectations play a major role in payments crises and suggested enriching the analysis with an explicit model of securities pricing.

The Transition Period

In her paper, `The Impact of Monetary Unification on the Composition of Markets', Alessandra Casella (University of California at Berkeley and CEPR) noted that when transaction costs are significant, a common currency implies both lump-sum savings and a change in the division of domestic and international trade. A common monetary policy and inflation rate may also affect market structures if inflation was previously distortionary.

For a two-country model in which heterogeneous agents sort themselves among markets, Casella showed that the lower-inflation country provides the more stable national currency, which will be used in all international transactions. If not all citizens engage in international trade, establishing a common currency may trigger distributional effects which suggest internal disagreement over the desirability of joining a union. The results of national referenda will generally differ when the two countries' initial conditions are different; and a common currency will secure favourable votes in both countries only for intermediate values of an exogenous development parameter. This suggests that political support for a common currency will be contingent on its timing.

Torsten Persson (Institute for International Economic Studies, Stockholm, and CEPR) agreed with the paper's general approach, but he proposed an alternative model whose results depend upon whether the policy-maker is utilitarian or majoritarian. In both cases, he found Casella's result that a common currency wins a majority in both countries at intermediate stages of development did not necessarily apply. William Branson (Princeton University and CEPR) disputed that a common currency implies a common inflation rate, since inflation rates differ across regions of the US.

Presenting a paper on `European Monetary Union and the International Role of the Ecu', written jointly with George Alogoskoufis, Richard Portes (CEPR and Birkbeck College, London) examined the implications of moving from the present ERM to Stage III of EMU for the ecu: as a means of payment, a unit of account and a store of value. The ecu's importance as a means of payment will grow once it substitutes for the dollar in the official interventions of EC member countries and those that peg to the ecu particularly if this practice is adopted by the countries of Eastern Europe and the European Bank for Reconstruction and Development. Its potential as a unit of account will depend on whether EC external trade is invoiced in ecus, which will depend in turn on inflation and the historical pull of the dollar. The authors argued that the dollar will dominate on account of `hysteresis', or historical inertia. The ecu's role as a store of value will depend on its holdings by private investors, which will increase only if Europe becomes clearly established as a safe haven. The benefits to the Community from reduced volatility of foreign exchange markets and increased seigniorage revenues if the ecu partly displaces the dollar must be weighed against the costs to the ECB of acting as a lender of last resort. Portes maintained that both the benefits and the costs are likely to be small.
Guillermo Calvo (IMF) noted that the length of the transition to Stage III which will not be short may critically affect the ecu's emergence as a vehicle currency. Jeroen Kremers (Ministry of Finance, The Netherlands) argued that the costs of the ecu's development as a vehicle currency may well outweigh the benefits and questioned whether the ECB would want it to perform this role. Matthew Canzoneri (Georgetown University) noted that market fears that the ECB will bow to political pressure to inflate away certain countries' debt may lead to the internationalization of ecu debt, in which case all member countries will have to pay an interest premium.

David Currie (London Business School and CEPR) presented a paper on `Hard ERM, Hard ECU, and European Monetary Union', in which he contrasted the evolution of the ERM with the move to EMU, viewing the former as a process shift and the latter as a regime shift requiring institutional changes, whose consequences are less certain. He maintained that the present `hard ERM' is unsustainable since countries such as France are reluctant to remain linked to the Deutschmark unless there is progress towards EMU.
Currie maintained that EMU's ability to deliver low and stable inflation will depend critically on the structures of institutional decision-making and of policy-makers' incentives, which are both neglected by the literature on optimal currency areas. Inflation-prone economies' policy preferences may still dominate those of Germany and other low-inflation countries, and if EMU leads to a loss of reputation in monetary policy a `hard ERM' may be superior. Further, the Draft Statutes do not guarantee full ECB independence, since finance ministers' management of the external exchange rate where exchange rate and domestic monetary policies are interdependent would compromise the ECB's independence vis-à-vis national fiscal authorities. Currie argued that fiscal policy coordination based on market mechanisms via risk premiums cannot adequately ensure the proper functioning of EMU, and he advocated instead applying a set of rules to limit national governments' fiscal actions.

Dale Henderson (Board of Governors of the Federal Reserve) stressed the importance of the ECB's ability to commit in choosing between EMU and a hard ERM. Andrew Hughes Hallett (University of Strathclyde and CEPR) argued that the main gain from EMU derives not from adoption of a common currency, but rather from the increased cooperation this implies.

Fiscal Policy Requirements

As EC member countries approach agreement on a common monetary policy and the institutional form of the ECB, they have yet to reach any agreement on the necessity, desirability or possible form of common fiscal constraints. But proponents of fiscal policy discipline argue this is essential to EMU's viability. In their paper, `Market-Based Fiscal Discipline in Monetary Unions: Evidence from the U.S. Municipal Bond Market', Morris Goldstein (IMF) and Geoffrey Woglom (Amherst College) provide an empirical test of an approach to fiscal discipline based on allowing private financial markets to constrain excesses through rising default premiums.

If countries with excessive deficits are eventually denied access to credit, the higher cost of credit and the risk of being cut off from credit markets may provide the discipline required for fiscal restraint. Goldstein and Woglom tested the first half of this hypothesis that financial markets impose higher default premiums on the fiscally imprudent by estimating the relationship between borrowing costs and default risks on 1982-90 data on the general obligation bonds of 39 US states. Their results indicated that states with conservative fiscal policies faced lower borrowing costs, and a state's stock of debt and current fiscal deficit both had significant positive effects on the interest rate it must offer. The authors cautioned, however, that the second half of the hypothesis that governments will adopt more conservative fiscal policy when presented with higher borrowing costs remains to be addressed.

Mervyn King (Bank of England and CEPR) questioned the relevance to Europe of analyses based on US data, since the maximum yield spread in the US is 84 basis points while in Europe it approaches 500, and the extent to which currency union will reduce that spread remains uncertain. Vito Tanzi (IMF) added that the US states' debt/income ratios were more symmetric than European countries', while Peter Garber pointed out that the observed yield spreads might be due to differences in liquidity, rather than default premiums. He suggested including bid/ask spreads as a regressor to test for an omitted variable.

Presenting a paper on `Federal Fiscal Policy and Optimum Currency Areas: Evidence for Europe from the United States', written jointly with Jeffrey Sachs, Xavier Sala-i-Martin (Yale University) noted that inter-regional shocks that generate unemployment in some regions and inflation in others may threaten a monetary union's survival. He argued that the economic cohesion of the US viewed as a collection of regions with irrevocably fixed exchange rates may be attributable to the role of the federal fiscal authority in insuring against such shocks. The federal authority maintains parity through tax and transfer policies to mitigate major imbalances, by redistributing income from `favourably shocked' to `adversely shocked' regions. This policy involves no aggregate fiscal deficits, and therefore contrasts with the fiscal stabilization policies of regional governments, which require them to run budget deficits.

Sala-i-Martin reported an analysis of data from the nine US census regions to assess the federal role as an insurer. The analysis indicated that an average region suffering a $1 adverse shock to its income could reduce its federal payments by 33-37 cents and received additional federal transfers of 1-8 cents, so its final disposable income fell by only 56-65 cents: the federal income tax system therefore acts as an automatic stabilizer. In contrast, an EC member country suffering a $1 adverse shock may reduce its tax payments to the Community by a negligible half cent. The authors concluded that Europeans wishing to view the US as a role model should consider creating a federal fiscal system at the same time as the proposed monetary union.

Behzad Diba (Georgetown University) argued that fiscal policy should not be used to counter permanent shocks to regions (or sectors) of the economy, although a federal fiscal system could act as a smoothing device if the financial system failed to provide one. The absence of an inter-regional insurance market from the US does not imply that no such market could emerge in Europe; and the case for a federal fiscal system amounts to a more general claim that governments should provide this type of insurance if the market cannot do so efficiently. Even without financial markets, regional governments may lend and borrow, so the automatic stabilizers may remain at the regional level. Alberto Giovannini (Columbia University and CEPR) added that a massive market failure would be required to prevent EC countries from smoothing their consumption over time. Matthew Canzoneri noted that a federal fiscal system could play the role of an automatic stabilizer if national governments were prevented from running large deficits by EC fiscal constraints.
Conclusions

In a concluding panel discussion on the problems of central bank design for EMU, William Branson (Princeton University and CEPR) noted that whereas during the 1960s and 1970s countries chose whether to enter a currency union depending on whether they were optimum currency areas, the Community's current plans for monetary integration reflect a political decision to reform its member countries economies' so that they become such an area. Further, he contended, the literature on game theory that is now applied to the ECB's design is static and therefore cannot capture its endogenous evolution. Antonio Borges (Banco de Portugal) cautioned against over-emphasizing the precise form of the Draft Statutes, since there is general agreement that the ECB should retain flexibility to respond to changing markets. Limits on budget deficits are also required, since a `no bail-out' clause will not be credible, and future research should therefore aim to define these feasible fiscal constraints. Graham Bishop (Salomon Brothers) noted that Europe's integration is largely driven by political factors and argued that EMU's continued existence with a politically independent financial system requires a credible `no bail-out' rule. Further, the markets have already anticipated and created a two-speed transition to European monetary integration. Massimo Russo (IMF) contrasted the national central banks' view that the ECB should be modelled on the current Bundesbank with EC member governments' preference for its policy to reflect an average of country preferences for inflation, which demonstrates that the ECB's independence remains an open question.

The papers presented at this conference will be published in early 1992 in a volume to be edited by Matthew Canzoneri, Vittorio Grilli and Paul Masson.