The Maastricht Council of December 1991 was both the end of a long process of study, discussion and negotiation and the beginning of a complex evolution of the European Community's institutions. Economic analysis has informed and learned from Europe's monetary unification, from the Werner Report to the Delors Committee's proposals for EMU. A joint CEPR conference with the Banco Pastor on `The Monetary Future of Europe', held in La Coruña on 11/12 December, examined the intellectual history that led to Maastricht and the prospects for monetary unification after the September crises on Europe's currency markets. The conference was organized by Guillermo de la Dehesa, Vice-Chairman of CEPR and Conseyo Delegado at the Banco Pastor, Alberto Giovannini, Jerome A Chazen Professor of International Business at Columbia University and member of the Italian Treasury Ministry's Council of Experts, Manuel Guitián, Associate Director of the Monetary and Exchange Affairs Department at the International Monetary Fund, and Richard Portes, Director of CEPR and Professor of Economics at Birkbeck College, London. Financial support and hospitality provided by the Diputación Provincial de La Coruña are gratefully acknowledged. A volume of the papers presented at the conference is now available from CEPR.*

Presenting the first paper, `European Monetary Union: The Intellectual Pre-History', Max Corden (Johns Hopkins University, Washington) reviewed the academic literature and policy proposals of the 1960s and 1970s in the light of more recent theoretical advances. Mundell's pioneering work on optimal currency areas had emphasized the importance of the exchange rate instrument to countries facing asymmetric shocks but also highlighted the benefits to trade of stable exchange rates. The Werner Report in the early 1970s inspired a small literature in which Phillips curves reflected countries' differing preferences between inflation and unemployment. The BarroGordon rational expectations approach introduced a discussion of `policy credibility' that was wholly absent from the earlier literature. Whether monetary integration can strengthen the credibility of the authorities' commitment will depend on the constitution of and political influences over the common central bank. But if the UK, France and Italy join a monetary union to gain credibility from an institution modelled on the Bundesbank, Germany should not; its own anti-inflation reputation can only suffer.

Barry Eichengreen (University of California, Berkeley, and CEPR) noted that the early literature had considered how industrial structure affects the nature of shocks, and hence the costs of monetary union, but Krugman and others had more recently related shocks endogenously to the exchange rate regime. Guillermo de la Dehesa stressed that US historical experience of large-scale depopulations demonstrated the role of labour mobility in coping with shocks.

David Begg (Birkbeck College, London, and CEPR) then presented his joint paper with Charles Wyplosz, `The European Monetary System: Recent Intellectual History'. The EMS was originally intended to be `symmetric', but German leadership and the ability of non-German members to `borrow' credibility were firmly established by the mid-1980s. Germany could tolerate this while effective capital controls remained. The commitment to unlimited intervention to preserve the parities remained untested but never fully credible for some years after the BaselNyborg agreements. The Delors Committee was established to determine how not whether to achieve monetary union, and its proposed `gradualist' approach created several `end-game' problems: national governments could invoke a `final realignment' the day before the union, run up excessive fiscal deficits in the transition expecting a bail-out, or print money to compete for seigniorage. The Maastricht Treaty sought to deal with these end-games but could not cope with German unification Europe's largest asymmetric shock since 1945.

Manuel Guitián noted that an important opportunity had been missed, since markets had for a long time given greater weight to EMS members' political commitment than to economic factors. Also, monetary policy had not been systematically asymmetric but historically dominated by Germany, which was perceived as the lowest-inflation EMS member even when it was not. Peter Kenen (Princeton University) argued that the capital controls in force in France and Italy before 1990 could not have averted the September crisis, since they applied only to residents; the admission of sterling a second `world-class' currency was also inherently destabilizing.

Francesco Papadia (Banca d'Italia) then presented his joint paper with Tommaso Padoa-Schioppa, `The Transition to EMU: Two Central Issues in the History of the Maastricht Treaty', which outlined the history of intergovernmental negotiations over the conduct of monetary policy in Stage II and the procedures for moving to Stage III. Unlike the Werner Report after the breakdown of Bretton Woods, the Delors Report conceived monetary union as a `national' system and focused on designing a common central bank rather than a framework for exchange rate interventions; it also concentrated on the final phase rather than the design of a technically sound and politically feasible transition. While it advocated establishing the European Central Bank (ECB) from the start of Stage II to strengthen policy coordination, German and Dutch objections led to the compromise of creating the European Monetary Institute (EMI). The appointment of an external President and provision of `own resources' avoided the risk of a weak monetary institution in the transition, but they did nothing to prevent the development of inconsistencies between exchange rate stability and autonomous conduct of monetary policy.

Santiago Fernández de Lis (Banco de España) argued that instability on Europe's currency markets was not due to the Maastricht Treaty's provisions for the transition but rather to uncertainty about the achievement and future membership of EMU after the Danish referendum. Horst Schulmann (Landeszentralbank in Hessen) disputed the authors' emphasis on institution-building and stressed the importance of economic convergence to monetary integration.

In his paper, `Economic and Monetary Union: What Happened? Exploring the Political Dimension of Optimum Currency Areas', Alberto Giovannini (Columbia University and CEPR) noted that `official' studies have focused on aggregate rather than distributive effects of EMU. Following the literature on trade liberalizations, he applied three well-known models of a three-sector economy with workers, employers and a financial sector to identify gainers and losers from different regimes. In the MundellFleming model, workers and employers prefer floating rates in face of real shocks and monetary union in face of financial shocks, but the financial sector is indifferent. In the BarroGordon, `rules-versus-discretion' model, workers prefer monetary union to hedge against the purchasing power uncertainty induced by high inflation but are otherwise indifferent, while preferences of employers and the financial sector depend on the distributional effects of inflation and taxation. In the `transactions costs' model, workers are indifferent, employers prefer monetary union, and the financial sector prefers floating rates.

Jeffry Frieden (University of California, Los Angeles) suggested pursuing Giovannini's approach further to consider explicitly the impact of inflation on relative prices. Also, workers' apparent preference for monetary union under `rules-versus-discretion' contrasts with empirical findings that `left-wing' governments are more inflation-prone. John Williamson (Institute for International Economics, Washington) attributed the growth of support for EMU to the ERM's perceived success in restricting realignments and reducing inflation during 1979-87. Its subsequent erosion reflected concerns about the `democratic deficit', the misguided attempt to freeze the parities before convergence had been achieved, sterling's entry at an overvalued rate, and German unification which presented the classic case for an exchange rate adjustment.

In `The Politics of EMU: A Selective Overview', Pierre Jacquet (Institut Français des Relations Internationales, Paris) considered the recent political processes in France, Germany and the UK. The traditional economic case for EMU that a single market requires a single currency to maintain stability had revived as capital mobility increased in the 1980s, while its political logic derived largely from the `bicycle' theory that each Commission needs a new initiative to maintain the momentum of integration. France had long been the leading advocate of a centralized and independent monetary institution, but French opponents of EMU now maintain that what works for federal Germany will not work for centralized France and that a political counterpart is required to balance the `economic government' of the ECB. German support for EMU has always been qualified, since it entails abandoning the Deutschmark for something much less certain, but the political case for uniting Germany within a united Europe has proved overwhelming. The UK proceeded with EMU solely because it feared isolation if the other members went ahead without it; the economic dimension was very strong on account of London's pre-eminent position as a financial centre.

José Juan Ruiz (Ministerio de Economia y Hacienda, Madrid) maintained that Jacquet had overstressed French leadership, citing Spain's proposals for the EMI to issue a `hard ecu' and its major role in establishing the Community's `cohesion funds'. Richard Portes noted that the French argument that a unitary state requires control of its central bank goes unheard in the UK, whose political establishment is now leaning towards central bank independence.

Manuel Guitián opened the Round Table Discussion, `Should EMU End Inflation?', noting that allowing market forces a greater role had clarified the constraints on economic policy-making, and `no-inflation' remained a desirable objective. Michael Bruno (Hebrew University of Jerusalem and CEPR) agreed but stressed that there are others, which include not only employment but also structural adjustment and growth. The ERM almost achieved zero inflation without EMU, which was also intended to eliminate transactions costs and enhance political unification. Otmar Issing (Deutsche Bundesbank and CEPR) pointed out that the proposed ECB will have greater powers than national central banks including the Bundesbank do now, but governments will still have the final say; and the commitment of those that are not yet ready to accord their own central banks independence must remain in doubt. Antonio Borges (Banco de Portugal) observed that `fixed-but-adjustable' exchange rates entail multiple equilibria which impede trade and lead to inefficiencies in long-term resource allocation. Exchange rate variations present particular barriers to entry in the financial sector, currently the least integrated in Europe.

Assessing the Future: The Outlook for EMU

The second day opened with a debate on `The Outlook for EMU'. In `EMU: A Solid Framework for Maastricht', Niels Thygesen (Kobenhavn Universitet) noted that doubts over Maastricht centre on whether a single currency will be adequately underpinned by economic and political union. The EMS was initially intended only to limit currency instability; extrapolations of the convergence already achieved suggested a single currency would yield only small additional gains, although later studies incorporating enhanced efficiency and reduced price discrimination estimate much larger benefits. Shocks are more highly correlated in Europe than in the US, and geographical concentration of production is lower, so the loss of the exchange rate instrument should be manageable. But Europe's regional specialization may increase, and its economic regions do not correspond to its nations, so the Community should adopt a single currency and effect adjustments by other means. Labour mobility is lower than in the US, but fiscal transfers could enhance EMU's robustness. The ERM parities are now much closer to equilibrium than before September, so this is a good time to press ahead with EMU at least for a core group rather than allowing the break-up to continue.

Patrick Minford (University of Liverpool and CEPR) replied that we now know that the ERM is unstable but not whether proceeding to EMU could offer an effective guarantee against shocks which reflect inevitable political and social uncertainty in democracies. The microeconomic benefits cited by Thygesen were unlikely to warrant its macroeconomic costs: EMU may be less unstable than the ERM, but it is twice as unstable as a pure float.

In `Does European Monetary Union Have a Future?', Martin Feldstein (National Bureau of Economic Research) claimed that any decision to move to EMU will depend overwhelmingly on political factors, since its strictly economic benefits are relatively small. Even ratification of the Maastricht Treaty by all twelve EC member states cannot guarantee that the world it envisioned will materialize. Few meet the convergence conditions; the UK retains an opt-out clause which Denmark will probably emulate; EC enlargement makes monetary and especially political union less likely; and public support for European integration is weak. The generation that embraced union to avoid repeating World War II is ageing; the completion of the single market will soon demonstrate that transactions costs are exaggerated; and there are to be no large-scale fiscal transfers as envisaged by the Delors Report.

Joly Dixon (Commission of the European Communities) elaborated on the many positive reasons for the post-war formation of the European Community; these have also driven its subsequent integration as a voluntary association of member states.

Guillermo de la Dehesa then presented his joint paper with Paul Krugman, `EMU and the Regions', which examined whether the integration of EC markets and monetary union would lead to `cohesion', or the long-run convergence of per capita income levels. With allowance for increasing returns, scale economies and falling transactions costs in the single market, market access may be more important than relative labour costs. And if `agglomeration' effects then exacerbate existing differences, integration may make monetary union more difficult as regions' increased specialization makes them more vulnerable to asymmetric shocks. Fiscal transfers are much larger within than among EC countries and are unlikely to provide an adequate adjustment mechanism following the US model; after transferring some 20% of Italian GDP to the Mezzogiorno for more than 20 years, agglomeration effects still dominate. And labour mobility is almost zero among EC member countries, since regional policy has been almost entirely devoted to the maintenance of existing population levels.

Daniel Gros (Centre for European Policy Studies, Brussels) disputed the importance of fiscal federalism in mitigating regional shocks: per capita transfers are larger in the US but in the `wrong' direction and largely concentrated in agriculture. Also, EMU must be compared with countries that currently exist, not with theoretical optimal currency areas. William Branson (Princeton University and CEPR) pointed out that US fiscal federalism involves transfers based on direct relationships between individuals (or corporations) and the central authorities, which are not intermediated by states or regions. Also, some two-thirds of the Community's 12 million immigrants in late 1989 were nationals of non-member states, whose movements may provide substantial labour mobility without migration by EC member nationals.

Shang-Jin Wei (Harvard University) presented his paper, `Trade Blocs and Currency Blocs', written with Jeffrey Frankel, which used a gravity model to examine the emergence of the European Community, the Western Hemisphere and the `Asia/Pacific Economic Community' (APEC) as trade or currency blocs and in particular whether exchange rate stability promotes intra-bloc trade. He reported evidence of trade bias within all three blocs, while the EFTA countries do not form an effective bloc and the US belongs both to the Western Hemisphere and also to APEC. Trade within the Community and Western Hemisphere roughly doubled during the 1980s, but there was little change to intra-APEC trade. There has also been some intra-regional exchange rate stabilization, as European countries have increasingly pegged to the Deutschmark and those of the Western Hemisphere to the dollar. Within APEC, however, only Singapore has consistently tracked the yen, and most currencies have pegged to the dollar. Real exchange rate stabilization appears to account for up to half the increase in intra-European trade observed in the 1980s, but this is quantitatively small.

Tamim Bayoumi (IMF) noted that the crude measure of `distance' in the gravity model was misleading for intra-APEC trade, since transport across the Pacific Ocean is relatively cheap. Richard Portes suggested that the European simulations should replace the Deutschmark with the ecu, to which many EFTA countries had explicitly pegged during the 1980s.

Carlo Monticelli (Committee of Governors of the EC Central Banks, Basel) and José Viñals (Committee of Governors of the EC Central Banks, Basel, and CEPR) then presented their paper, `European Monetary Policy in Stage Three: What are the Issues?'. On strategic matters, Viñals stressed that the primacy of price stability and establishing the ECB's statutory independence were insufficient to maximize its anti-inflationary credibility, since the well-established correlation between low inflation and central bank independence tells us nothing about its causality. Inflation convergence conditions may ensure that only the `successful' join the union, but fiscal conditions are also important. Turning to tactics, Monticelli noted that EC capital markets were effectively integrated already, but further harmonization of banking practices would be desirable: to establish common conventions on signalling to private agents, and to deal with differential subsidies from central to commercial banks in the form of common services.

Jean-Jacques Rey (National Bank of Belgium) questioned whether the authors' empirical evidence that money demand is now more stable for an EMU of all EC member states applied to the more likely outcome of a core group of a two-tier Europe, with the outer tier perhaps remaining in the ERM. Frederic Mishkin (Columbia University) stressed the important signalling role of intermediate monetary targets in establishing credibility, although even the Bundesbank has departed from these when there were sensible reasons to do so. Monetary integration will also require closer involvement by some central banks in supervision and as lender-of-last-resort.

In the final paper, `EMU and the International Monetary System', Robert Mundell (Columbia University) noted that the arguments for establishing the widest possible area of free trade also apply to fixed exchange rates, so the theory of optimal currency areas may argue against floating rates. The European countries were unable to establish a common float against the dollar in the early 1970s and now face a choice between a monetary union for the continent as a whole and a `DM-zone'. This may prove a more natural OCA, but countries west of the Rhine that remain outside will then float with the dollar instead. Historical evidence on earlier currency areas indicates that the dominant country can always veto alternatives and tends to reject reform; leadership rivals seek reform while smaller countries acquiesce in the current leadership; and changes to international monetary systems are characterized by evolution, not revolution, and are therefore predictable. Europe will not have a unified strong government even in the year 2000, so if EMU goes ahead with a flexible dollar/ecu rate, it will lose any struggle for clients with the US; it would thus do much better to reach a fixed-rates agreement with the US and Japan.

Andrew Crockett (Bank of England) noted that the strains in the ERM differed from those under Bretton Woods: all ERM member countries envied the Bundesbank's anti-inflationary credibility, while the dollar was not a benign anchor in the early 1970s. Also, the Bundesbank was prepared to share monetary sovereignty, conditional on its institutional form, in response to the stronger political will for unification. Charles Wyplosz (INSEAD and CEPR) proposed basing the eventual monetary union on a renamed Deutschmark rather than a basket currency such as the ecu, to replace the potentially unstable Stage II with the well-established, credible monetary policy managed by the Bundesbank.

The conference closed with a second Round Table Discussion on `EMS, EMU and the Rest of the World', led by Luigi Spaventa (Università degli Studi di Roma, `La Sapienza', and CEPR), who maintained that the clause of the Maastricht Treaty allowing a `last realignment' in 1999 was largely responsible for the ERM's break-up. `Market ratification' anticipated and accelerated what would have happened anyway: the ERM can survive without EMU, but not vice versa. He contrasted the UK's `liberation' from the constraints of the ERM with Italy's humiliation: both countries abandoned their parities but the temporary nature of this arrangement was as prominent in Italian official statements as it was conspicuous by its absence from those of the UK. The markets now assume that the across-the-board realignment of the Deutschmark that was rejected in 1990 is incomplete, and if France maintains its current parity it will take time to adjust to the devaluation of its major non-German trading partners. The immediate future of the ERM will depend critically on whether it is still intended to effect the transition to EMU or rather to revert to a simpler, less ambitious system reflecting economic choice rather than political will.