EMU and the International Monetary System
Chapter

Comments on Ghironi and Giavazzi

Editor(s):
Thomas Krueger, Paul Masson, and Bart Turtelboom
Published Date:
September 1997
Share
  • ShareShare
Show Summary Details
Author(s)
André Sapir

One of the most crucial issues facing European policymakers concerns the composition of the group of countries that will enter the third phase of EMU in January 1999, and the process whereby those lagging behind will be able to join the monetary union as swiftly as possible. Luckily, economic analysis provides some guidance for solving this problem. The arguments in the literature can be divided into demand and supply factors.

On the demand side, the optimum currency area (OCA) theory offers reasons why countries may wish to join a monetary union, or stay outside. According to this theory, the demand for membership hinges on the economic structure of a particular country, and on the potential cost associated with the loss of the exchange rate instrument in the face of country-specific shocks. Based on this view, Barry Eichengreen and others have suggested that the EU is divided into two groups of countries: a hard core made up of countries whose disturbances are highly correlated with those of Germany, namely Austria, France, Germany, Denmark, Belgium, Luxembourg, and the Netherlands; and a periphery comprising countries with significantly more idiosyncratic shocks, namely, Finland, Greece, Ireland, Italy, Portugal, Spain, Sweden, and the United Kingdom. But studies based on past behavior may be of little value for predicting the future when it comes to a regime change such as EMU. Thus, countries may meet the OCA criterion ex ante, but not ex post. This is essentially the view held by Krugman,1 who believes that EMU is bound to foster specialization in Europe and, hence, increase the probability of idiosyncratic shocks. Conversely, countries may meet the OCA criterion ex post, even though they fail to do so ex ante, if EMU reinforces intra-industry rather than interindustry trade, as suggested by Bini-Smaghi and Vori.2

On the other hand, the Maastricht criteria may be interpreted as rules for supplying membership to countries able and willing to adhere to a particular brand of monetary union, one characterized by low inflation. Until recently, these criteria also suggested that the EU is divided into two groups of countries, roughly equivalent to the two groups identified by the OCA criterion. But the latest evidence shows a great deal of macroeconomic convergence among countries that are nearly EU members, which suggests that the Maastricht criteria may be losing some of their discriminating qualities.

The very interesting paper by Fabio Ghironi and Francesco Giavazzi attempts to provide yet another approach to the insiders-outsiders problem, which differs from the previous two approaches in two fundamental ways. First, Ghironi and Giavazzi are not concerned so much with which countries should be inside or outside the monetary union as with the desirable size of the two groups. In fact, in their paper, there are no differences between the countries that make up the two groupings: no difference in economic structure like in the OCA theory, nor any difference in macroeconomic conditions like in the Maastricht approach. Second, the authors rightly postulate that the relationship between insiders and outsiders needs to be analyzed within a framework that explicitly recognizes the existence of third countries not belonging to the EU.

The goal of Ghironi and Giavazzi is to highlight conflicts that may potentially arise at three separate levels: between different economic authorities within each country; between EU countries inside and outside EMU; and between the EU and the United States, the latter representing all countries outside the EU. To do so, they construct a very elegant three-country model, with two economic authorities in each country: a central bank in charge of monetary policy, and a government controlling fiscal policy. The three economies undergo a common, symmetric shock to which the authorities respond so as to minimize a loss function that depends on inflation and employment. Each of the two authorities has its own loss function, featuring specific relative weights for inflation and employment: the central bank is assumed to care more about inflation, while the government attaches more weight to employment.

The authors analyze different fiscal and monetary policy regimes, but emphasize in the conclusion the setup that, they believe, is most likely to characterize the working of EMU, at least initially: frozen fiscal policies in Europe and in the United States, noncooperative monetary policies under an ERM II regime in Europe, and noncooperative monetary policies under a flexible exchange rate regime between Europe and the United States. Given these assumptions, a number of conflicts emerge: (1) three of the four European authorities favor the largest possible monetary union, but the government of the outs prefers a small union; (2) if a full union is not possible initially, the ECB and the government of the ins are bound to disagree on the merits of enlarging progressively the union; and (3) a similar conflict holds for the central bank and the government of the outs.

I see two problems with the preferred results of Ghironi and Giavazzi. First, they are predicated on the assumption that the monetary authorities of the ins and the outs play a noncooperative game. Essentially, the paper postulates noncooperation and concludes, unsurprisingly, that it implies conflicts of interest. This assumption I find highly unrealistic not only because of existing institutional arrangements in Europe, but also because of the desire of (most of) the outs to join the monetary union. As a matter of fact, the results presented in the paper clearly indicate that cooperation is more desirable than noncooperation, since the values of the loss functions are lower in the former than in the latter case for three of the four European authorities (only the central bank of the outs prefers a noncooperative arrangement). Second, given its extreme complexity, the model cannot be solved analytically. The results, therefore, amount to simulations based on specific numerical assumptions for the key parameters of the model. Hence, in the absence of sensitivity analysis, it is impossible to judge the robustness of these results.

But my main criticism of the paper lies with its basic approach. In particular, I do not share the view of the authors that the most important characterization of the relationship between the ins and the outs is the relative size of the two groupings. To my mind, the crucial differences between the two groupings are in terms of economic structure and macroeconomic conditions. By sweeping these differences under the rug, the authors abstract from the key issue of membership of the ins and the outs. Besides, by focusing on size rather than composition, Ghironi and Giavazzi make the strong assumption of a uniform behavior within the ins and within the outs. While this may be realistic for the presumed ins, it seems a poor assumption for the presumed outs.

Finally, I wish the authors had not treated Europe and the United States symmetrically. What comes to mind, in particular, is the difference in the functioning of labor markets. The paper assumes that, at the start of monetary union, the three countries of the model share equal initial conditions, their economies being in equilibrium. Unfortunately, the reality is far from such ideal. Whereas it can reasonably be argued that the United States is close to full employment, the situation in Europe is one of high unemployment, much of it of a structural nature. This, I believe, is an important potential source of conflict not only between the United States and Europe, but also between central banks and governments in Europe.

Discussion

Mr. Szápary said he would like to take up two broader issues that had not yet been discussed at the conference but which were worthwhile to consider. First, how would fiscal policy fit into the monetary union? As far as he knew, EMU would be the first large monetary union without a federal budget. There had been earlier examples of similar situations, but not good ones. The Austro-Hungarian monarchy had a monetary union with two separate budgets but the monetary arrangement worked as a quasi-gold standard. So that monetary union was not a good example to illustrate the problems that could result from divergent fiscal policies under EMU.

Having given up monetary and exchange rate policies to treat asymmetric shocks, there remained only fiscal policy for EMU, Mr. Szapáry continued. But depending on the structure of the fiscal policy, asymmetric shocks would affect fiscal policies differently. How could that be dealt with? Should one have a federal budget of some sort? He was not suggesting that one should put on top of the existing budget a federal structure that would actually increase the size of the government, but perhaps some of the responsibilities could be transferred to the federal budget. Otherwise, what kind of coordination in the structure or convergence in the structure of taxes and expenditures would be needed?

Mr. Mussa, in response to a comment by Mr. Alogoskoufis that the potential negative shock to money demand as a result of EMU should be addressed, indicated that there would probably be some effect. The sensible thing to do would be to accommodate the change in the demand for money, and there was no way to do that except by facing up to the difficulty of diagnosing whether and to what extent such a change in the demand for money had taken place.

However, Mr. Mussa continued, there might also be an effect in the other direction. If people had a particular preference to hold on to the old form of bank notes because of a mistrust of the new currency, they might not want to transact them quite as actively. The demand for the old currencies might therefore go up in the short term, and the monetary authorities will need to be alert to whether the demand for high-powered money at given prices, interest rates, and output had changed so that their policy would not become either unduly expansionary or contractionary. The Swiss had faced precisely that type of problem when they changed their reserve requirement procedures for banks and misjudged for a while the impact on the demand for high-powered money.

Paul Kragman, “Lessons of Massachusetts for EMU,” in Adjustment and Growth in the European Monetary Union, ed, by Francisco Torres and Francesco Giavazzi (Oxford: Cambridge University Press, 1993), pp. 241–61.

Lorenzo Bini-Smaghi and Silvia Vori, “Rating the EC as an Optimal Currency Area,” Banca d’;Italia, Temi di Disaissione (Rome: Bank of Italy, January 1993).

    Other Resources Citing This Publication