Standard applications of game theory demonstrate that the European Monetary System (EMS) can benefit all its members: coordinated policies can help avoid competitive appreciation or depreciation of exchange rates. The cooperative interpretation of the EMS, however, faces two empirical difficulties. In the first place, it cannot explain the dominant position of Germany within the EMS. In the second place, this interpretation fails to provide a convincing explanation for the observed tendency towards disinflation within the System. If the members are bent on disinflation, the cooperative interpretation suggests that the EMS should induce greater ease, not tightness of monetary policy. The monetary discipline hypothesis has the merit of meeting both these objections. It offers a clear explanation of why the other members expect Germany to retain its monetary independence: other members would like to have German monetary policy themselves but are unable to achieve this. The hypothesis should also predict that the non-German members would experience lower inflation once inside the System. The basic drawback of the monetary discipline hypothesis, however, is that it does not provide a simple explanation of why Germany would wish to remain a member of the EMS.
This paper attempts to combine the two interpretations of the EMS. The argument relies on the monetary discipline hypothesis to explain the interest of the inflation-prone members in the System, while relying on analysis of strategic cooperation to furnish a motive for the German presence. The analysis also suggests that the advantages of membership for Germany are secure, while the gains to its other members depend on a variety of factors, whose net effect is uncertain. Germany is bound to benefit from improved competitiveness between EMS realignments as long as the other members continue to inflate. But the monetary discipline which the System provides to other members brings with it costs which need not always outweigh the benefits. The analysis also suggests that the EMS induces tighter monetary policy in Germany. The reason for this is that, between realignments, Germany can increase its competitiveness by disinflating; this may not be possible outside the EMS, when the Deutschmark would appreciate immediately against more inflation- prone currencies.
This last conclusion helps explain a concern often voiced in Europe, that the EMS has intensified German disinflation in the 1980s. This explanation of German gains from membership does, however, rest on the assumption that inflation in other countries is not so high that the Deutschmark appreciates more than German policy-makers consider desirable.
The analysis in this paper is based on a model which essentially follows Barro and Gordon's analysis of monetary discipline. The problem of discipline arises in this framework because output is normally too low and there is a temptation to increase output by introducing surprise inflations. Yielding to this temptation has costs, because it increases anticipated inflation. Policy-makers may resist this temptation if they assess these costs appropriately, but since the costs are suffered only in the future, policy-makers may discount them excessively and yield to the temptation to inflate.
We adopt a two-country framework for our analysis. Germany is assumed to be a country whose monetary authorities assess correctly the future costs of surprise inflation whereas in France the authorities undervalue these costs. As a result Germany settles for low inflation (assumed to be zero), while France fixes upon a significant positive level without any compensating benefits (since the French inflation is fully anticipated). The difference between German and French behaviour, at a deeper level, may be thought to arise from a difference in central bank independence. The German Bundesbank is completely independent and thus perfectly able to discount the future properly, whereas the Banque de France is dominated by elected officials, and is therefore obliged to assign heavy priority to events preceding the next election.
We introduce the EMS in this context, analysing the System in two stages. In the first the EMS is assumed simply to increase the political costs to France of creating more inflation than exists in Germany. These costs arise because of the burden of keeping the exchange rate fixed between realignments, which requires unpopular capital controls, for example. The costs may also arise because of the unpopularity of devaluations, which voters interpret as failures of their political leaders. At this stage of the analysis, movements in the terms of trade are not allowed to affect competitiveness but can only affect inflation: wages adjust to prices in the traded goods industry thus keeping employment in the industry the same.
In the second stage of the analysis, non-traded goods are introduced. We assume that the ratio of traded goods to output tends to be too low in both countries because of barriers to trade (consisting not of tariffs which are prohibited within the European Community, but of non-tariff barriers such as licensing, quality regulations, and preferential government purchases of home goods). In the presence of non-traded goods, employment can be attracted into the traded goods industry from elsewhere in the economy. Movements in the terms of trade can alter the competitiveness of the traded goods sector. Therefore inflation in France can lower the ratio of traded goods to output between realignments, so that France suffers more from inflation than it would outside the EMS and, accordingly, has a greater incentive to keep inflation down. Germany is also even more likely to disinflate within the EMS, as mentioned before, since disinflation affords gains in competitiveness to the Germans that would not arise if the Deutschmark could appreciate immediately to reflect inflation in France.
The analysis also gives some insight into the frequency of realignments. Longer intervals between realignments encourage higher French inflation by reducing the impact of higher producer prices on consumer prices in France: import prices cannot rise because the franc cannot depreciate. On the other hand, such longer intervals promote lower inflation through an incentive to maintain competitiveness. If the second factor prevails, longer intervals stimulate monetary discipline. The view usually expressed in Europe is that this second factor dominates: the desire to avoid frequent realignments works in favour rather than against monetary discipline.