CHAPTER 3 Exchange Rate Experiences of Small EMS Countries: Belgium, Denmark, and the Netherlands

International Monetary Fund
Published Date:
September 1990
  • ShareShare
Show Summary Details
Paul De Grauwe and Wim Vanhaverbeke 

In this paper we address issues relating to the exchange rate policies of the small European Monetary System (EMS) countries since 1979. We will consider whether these policies have been forced on these countries by the prevailing economic environment, or whether they were the outcome of discretionary policy choices. In addition, we will address the issue of how these exchange rate movements have contributed to establishing macroeconomic equilibrium. In this connection the interaction of exchange rate and fiscal policies is important.

The analysis will be restricted to the cases of Belgium, the Netherlands, and Denmark. The exchange rate experience of Ireland is sufficiently divergent from these three countries to warrant separate treatment. Such an analysis was performed recently by Dornbusch (1989) and Kremers (1989).

I. Exchange Rate Developments

The EMS agreement has not prevented participating countries from adjusting their parities regularly. As is well known, these parity changes have been more frequent during the first five years of the existence of the EMS than during the recent period.1

The effective exchange rates of EMS members, which measure the average of the exchange rates against all currencies, are influenced not only by the parity adjustments within the system, but also by the movements of the major currencies outside the system relative to the EMS currencies. These movements must, for all practical purposes, be considered as exogenous for the small EMS countries.2

Chart 1 shows the effective exchange rates of the small EMS countries and, for purposes of comparison, the effective exchange rates of the large EMS countries. Chart 1 reveals important differences in the movements of the effective exchange rates of the small EMS countries. In particular, the Netherlands stands out as having experienced quite different exchange rate behavior from that of the other small countries. First, the guilder did not show a depreciation on an effective basis during the first half of the 1980s. Second, since 1985 it has experienced an effective appreciation. As a result, the guilder was on average 12 percent more expensive in 1988 than in 1979.

Chart 1.Nominal Effective Exchange Rates of EMS Countries

(1979 = 100)

Source: IMF, International Financial Statistics.

Chart 1 also makes clear that it is more sensible to distinguish countries using criteria different from large and small. It appears that the Dutch exchange rate trends are similar to those of the Federal Republic of Germany. Both countries have experienced the same movements in their effective exchange rates, and both have seen their currencies appreciate. This contrasts with the experience of all the other EMS countries, whose currencies depreciated from 1979 to 1988. (As will be made clear, this different grouping of countries coincides with the grouping of low- and high-inflation countries.)

An interesting contrast between the small and the large EMS countries is that the former have experienced smaller nominal depreciations or appreciations. The “weak” small EMS currencies depreciated from 1979 to 1988, by 12 to 19 percent. The “weak” large EMS currencies depreciated by more than 20 percent. The same can be said for the “strong” EMS currencies. The Dutch guilder appreciated by 12 percent, and the deutsche mark by close to 20 percent.

Nominal exchange rate movements are important insofar as they affect the competitiveness of a country. The extent to which nominal exchange rate movements influence competitiveness largely depends on how these movements compare with the movements of prices and costs. A composite measure of the changes in competitiveness is provided by the real effective exchange rates. These are shown in Chart 2. We use unit labor costs as our measure of costs.

Chart 2.Real Effective Exchange Rates of EMS Countries

(1979 = 100)

Source: IMF, International Financial Statistics.

The most striking feature of Chart 2 is the strongly divergent exchange rate experience of the small versus the large EMS countries. First, the real exchange rate movements of the small countries exceed those of the large countries. This is rather surprising since we noted the opposite for the nominal exchange rates. Second, all the small EMS countries experienced a real depreciation of their currencies from 1979 to 1988. These real depreciations ranged from 8 percent for the Danish krone to more than 30 percent for the Belgian franc. These real depreciations, however, occurred mostly during the first half of that period, namely from 1979 to 1983. Third, these changes in the real exchange rates of the small EMS countries tended to be larger than the changes in the nominal exchange rates; the phenomenon is illustrated in Table 1.

Table 1.Changes in Nominal and Real Effective Exchange Rates: 1979-83(in percent)
Source: IMF, International Financial Statistics.
Source: IMF, International Financial Statistics.

We also note that although the Dutch guilder experienced a nominal effective appreciation during 1979-88, it depreciated in real terms by close to 20 percent during the same period. This indicates that the Netherlands’ favorable inflation performance more than compensated for the nominal effective appreciation of the guilder. We note here also that the real depreciations occurred at the beginning of the EMS period.

Denmark stands out in this group of small EMS countries. Up to 1982 it experienced a strong real depreciation comparable to Belgium. Since 1984, however, the Danish krone has appreciated by more than 10 percent in real terms.

In the large EMS countries we generally do not observe the same large real depreciations as in the small EMS countries. The real effective exchange rates of the large EMS countries exhibit changes, which remain within a band of approximately plus and minus 10 percent.

The previous evidence on the real exchange rate movements allows us to formulate a number of questions. First, what is the source of the large real depreciations of the currencies of the small EMS countries? Are they the result of policy decisions aimed at improving these countries’ competitive position? Or are these real depreciations the outcome of external disturbances? Second, what have been the effects of the real depreciations? How have they affected the output and employment situations in these countries? The latter question, of course, cannot be answered separately from the former. For example, one may find that the real depreciation of a currency is the result of an exogenous disturbance (say, a decline of world demand for the home product). This same disturbance will affect output and employment in a particular way. The correlation we observe between the real exchange rate and the output and employment levels of the country will be quite different from what we would observe if the real depreciation owed to a decision to devalue the currency.

II. Real Exchange Rates and the Goods Markets: The Theory

The relation between the real exchange rate and the levels of output and employment depends on the nature of the disturbance to which the economy is subjected. In this section we analyze two such disturbances from a theoretical point of view. One is a real depreciation engineered by the authorities who devalue the currency.3 A second shock is a reduction of world demand for the home good. We will see what theory teaches about how the real exchange rate and the levels of output and employment correlate in these two alternative scenarios.

Standard macroeconomic models predict certain effects of a devaluation (see, for example, Marston, 1985; and Dornbusch, 1980). To the extent that there is some price and wage rigidity, an unexpected devaluation will lead to an improvement in competitiveness and a temporary boom in domestic economic activity. During this phase of the cycle we will also observe a real depreciation of the currency. After some time (the length of which depends on such things as the length of labor contracts and the speed with which wages are indexed), domestic prices and wages adjust upward. This reduces competitiveness again and slows down economic activity. In this phase we observe that the currency appreciates in real terms and tends to return to its initial level. As a result, the devaluation has only a temporary effect on the real exchange rate. In addition, the (temporary) real depreciation of the currency typically does not exceed the size of the (nominal) devaluation.

A decline in world demand for the home output produces the following stylized effects in standard macroeconomic models of open economies: domestic output and employment are negatively affected, and the deflationary forces exerted by this shock tend to slow cost and price increases relative to other countries. As a result, the country that maintains a fixed (nominal) exchange rate with its trading partners experiences a real depreciation of its currency (its competitiveness improves). This real depreciation softens the negative output and employment effects of the reduction of world demand. If this reduction is permanent, the real depreciation of the domestic currency will also have a permanent character. In addition, the real depreciation will be larger than the nominal depreciation, if the country involved pegs its currency.

III. Output, Employment, and Inflation

In this section we analyze the trends in output and employment of the small EMS countries. Chart 3 presents the average yearly growth rates of GDP (in constant prices) of the small and large EMS countries. The chart shows that, since the end of the 1970s, GDP has grown at a slower rate in the small EMS countries. Whereas the weighted average growth rate of GDP was 2.1 percent during 1979–88 in the large EMS countries, it reached only 1.6 percent in the small EMS countries. Note also that the difference in growth performance between small and large countries was especially pronounced during the turbulent period 1978–81.

Chart 3.Real Growth of GDP

Source: OECD, Economic Outlook.

The same unfavorable performance of the small EMS countries is evident when we compare the evolution of total employment in both groups of countries (see Chart 4). We observe that employment declines significantly faster from 1979 to 1983 in the small EMS countries relative to the large ones. Since 1983, however, we see a stronger recovery in the small countries.

Chart 4.Employment Indicators

(1979 = 100)

Source: OECD, Economic Outlook.

These developments in output and employment have also had important implications for the level of unemployment. Chart 4 also presents the unemployment rate in the two groups of countries. From 1979 to 1983 the unemployment rate in the small EMS countries increases from 6.4 percent to 13.5 percent. In the large EMS countries the unemployment rate rises from 5.1 percent to 8.6 percent. Since 1984, however, we observe a substantial decline of the unemployment rate in the small countries. This decline is absent in the large countries, where the unemployment rate “creeps up” steadily until 1988.

It should be noted here that the output and employment trends in the small EMS countries are dominated by the experience of Belgium and the Netherlands. These two countries’ GDP represents more than 75 percent of the total. Chart 5, therefore, shows the employment trends of the small countries separately. One can see that, since 1982, Denmark provides a striking exception to the unfavorable employment performance of the other small EMS countries, experiencing an increase in total employment of close to 10 percent. We will return to the Danish case in a later section.

Chart 5.Total Employment

(1979 = 100)

Source: European Economy.

We can summarize the evidence as follows. We have observed that the small EMS countries (and in particular Belgium and the Netherlands) have experienced a substantial real depreciation during 1979–83. This real depreciation coincided with a considerable slowdown in output and employment, which tended to be more pronounced than in the large EMS countries. (And, as noted earlier, these real depreciations of the small EMS countries were larger than the nominal exchange rate changes.) All this seems to suggest that external disturbances, such as the oil shock of 1979–80 and the world recession of 1981–82, tended to have stonger adverse effects on output and employment in the smaller EMS countries than in the larger ones. This necessitated higher real currency depreciations in the small EMS countries.

The question that arises immediately, of course, is why these small countries were hit more severely than the others. Does it have something to do with being small? Or should one look at such variables as the industrial structure and labor market institutions? The size factor can be dismissed quickly. Table 2 presents evidence on the increase in the unemployment rates from the 1970s to the 1980s in the small OECD countries relative to the large ones. We have also split the small OECD countries into two groups, those belonging to the EMS and those that do not.

Table 2.Increase in Unemployment Rate from 1974–79 to 1980–88(in percent)
Large OECD countries+2.9
Small OECD countries+2.6
of which:
Source: OECD, Economic Outlook.
Source: OECD, Economic Outlook.

There does not seem to be any evidence that, on average, the small OECD countries (which include Austria, Sweden, and Switzerland) have experienced more difficulties in adjusting to the worldwide shocks of the 1979–82 period than the large OECD countries. The increase in unemployment of the large and the small OECD countries is approximately the same, that is, 2.9 percent and 2.6 percent, respectively. A more interesting observation is that the small OECD countries outside the EMS have seen their unemployment increase during the 1980s by only 1.2 percent, whereas the small EMS countries experienced a substantial increase of more than 5 percent.

A more promising explanation is provided by a look at structural considerations. As was indicated earlier, the experience of the small EMS countries has been dominated by the economic situation of Belgium and the Netherlands. These two countries belong to the industrial region of the Northern European continent. This region is characterized by of a relatively old industrial structure, with well-known implications: a concentration of production in mature markets with less growth potential; relatively rigid labor market institutions that have been built up over many years of adversarial industrial relations; and favorable social security systems that tend to increase labor costs.

All these elements have been analyzed in much detail in the literature. There is now a consensus that these unfavorable supply characteristics made these economies vulnerable to the supply shocks of 1979–80 and the worldwide recession of 1981–82. This vulnerability shows up in a stronger decline in output and employment than in regions with more recent industrialization.

The hypothesis that the unfavorable macroeconomic performance of the small EMS countries—such as Belgium and the Netherlands—during the early 1980s has something to do with economic and industrial structures can be given some indirect confirmation by analyzing the performance of comparable regions during the same period. We chose to study the Northern part of Germany (Nordrhein-Westphalian, Nieder-Sachsen, Hamburg, Bremen, Schleswig-Holstein), which in many ways has economic structures similar to Belgium and the Netherlands. The hypothesis we wish to formulate is that this whole region suffered more than other regions from the worldwide disturbances of the early 1980s.

To test for this hypothesis we collected output and unemployment data for Northern Germany, which we compare with data for Belgium and the Netherlands, and which we also contrast with Southern Germany. The results for output are given in Table 3, and for unemployment in Chart 6.

Table 3.Average Yearly Growth of Output During 1980-87(in percent)
North Germany1.23
South Germany2.5
Source: Statistisches Jahresbuch, 1988; and OECD, Economic Outlook.Note: North Germany comprises Nordrhein-Westphalian, Nieder-Sachsen, Hamburg, Bremen, and Schleswig-Holstein; South Germany consists of Hessen, Rheinland-Pfalz, Saarland, Baden-Wiirtenberg, and Bayern.
Source: Statistisches Jahresbuch, 1988; and OECD, Economic Outlook.Note: North Germany comprises Nordrhein-Westphalian, Nieder-Sachsen, Hamburg, Bremen, and Schleswig-Holstein; South Germany consists of Hessen, Rheinland-Pfalz, Saarland, Baden-Wiirtenberg, and Bayern.

Chart 6.Unemployment Rate: North and South Germany

Source: Bundesanstalt für Arbeit.

Table 3 confirms that Northern Germany experienced the same kind of low growth rates as the Benelux countries during the 1980s. Chart 6, which shows the unemployment rates in the North and the South of Germany, illustrates the same phenomenon for North Germany that we observed for the small EMS countries. Unemployment in these regions increases rapidly during the early 1980s. In this connection, it is interesting to note the contrast between Northern and Southern Germany.

This evidence then suggests that the unfavorable economic performance of small EMS countries, such as Belgium and the Netherlands, is part of a more general problem of inappropriate economic structures that we also find in the Northern part of Germany. Note, however, that contrary to North Germany, Belgium and the Netherlands seem to have been able to improve their employment situation since the mid-1980s. This does not seem to be the case in North Germany. This is made clear from Chart 7, which shows the unemployment rates of Belgium, the Netherlands, and North Germany. The unemployment rate in North Germany remains at its high level since 1985, whereas it tends to decline in Belgium and the Netherlands. This phenomenon may be explained by the large real depreciations of the Belgian franc and the Dutch guilder, which boosted the competitiveness of Belgian and Dutch products. North Germany could not rely in the same way on this real depreciation to improve its competitive position. As a subsitute for real exchange rate flexibility, North Germany could count on a higher mobility of labor to reduce unemployment. The evidence, however, seems to suggest that this mechanism worked imperfectly.

Chart 7.Unemployment Rate: Belgium, Netherlands, and North Germany

Source: European Economy.

IV. Fiscal Policies

In the previous section we have argued that the sizable real depreciations observed in the Benelux countries during the early 1980s can be explained by the large and worldwide disturbances occuring during that period. The question also arises about the role of fiscal policies in this process.

Chart 8 shows the average government budget deficits in the small and large EMS countries. The most striking feature is that the budget deficits in the small countries have exhibited greater fluctuation in the 1980s than the large countries. During 1978–82, the budget deficits in the small countries more than doubled, whereas in the large countries they increased only marginally (from 4 percent of GDP in 1978 to 5.5 percent in 1982). From 1982–83 on, however, the deficits in the small countries decline substantially.

Chart 8.Budget Deficit

(as percent of GDP)

Source: OECD Economic Outlook.

Chart 9 shows the individual countries’ budget deficits and confirms the contrast between the large fluctuations in the budget deficits of the small countries with the relatively steady movements in the large countries. The contrast in fiscal policies between the two groups of countries is striking. It can be explained along the lines developed in the previous section. The worldwide disturbances of 1979–82 had particularly strong effects on the small EMS countries. The automatic stabilizers in the government budget (such as unemployment benefits and progressive taxation) then led to increases in the budget deficits, which were more pronounced in the small EMS countries. This fiscal policy effect may have been reinforced by the fact that in the small countries—especially in the Netherlands, Belgium, and Denmark—social security coverage and payments are larger than in other countries. This structural difference in the level of social spending is illustrated in Table 4. It also explains the greater sensitivity of the budgets in the small countries to a strong recession, like the one that occurred in 1981–82.

Table 4.Spending for Social Security and Social Assistance in 1983(as a percent of GDP)
Germany, Fed. Rep. of15.3
Source: OECD, Economic Surveys: Netherlands, 1988/89, p. 41.
Source: OECD, Economic Surveys: Netherlands, 1988/89, p. 41.

Chart 9.Budget Deficit of Small and Large EMS Countries (as percent of GDP)

Source: European Economy.

This automatic accommodation of the worldwide shocks through increasing budget deficits led to an explosion of government debt. Around 1982 it became clear in Belgium, Denmark, and the Netherlands that these increasing levels of government debt were becoming unsustainable. These countries then launched a policy of fiscal retrenchment, which has continued to this day.

The combination of fiscal restriction and the maintenance of a “real” depreciated currency appears to have helped such small countries as Belgium and the Netherlands improve their macroeconomic performance. This is most clearly shown in the evolution of total employment, which from 1984 on, begins to increase faster than in the large EMS countries (see Chart 4); we observe the same improvement with the unemployment data (Chart 4).4

V. Groping Toward External Equilibrium

This combination of real depreciation and fiscal restriction was also instrumental in leading to a turnaround in the current accounts of Belgium and the Netherlands (see Chart 10). These two countries’ have managed to eliminate their current account deficits during the early 1980s and to transform them into surpluses.

Chart 10.Current Account of Small EMS Countries

(as percent of GDP)

Source: European Economy.

Denmark appears to be a special case again. Its current account deficits have remained relatively large throughout the 1980s. Why is this?

Since 1982, the fiscal correction in Denmark went further than in the other small EMS countries. As can be seen from Chart 9, Denmark has achieved surpluses on its government budget since 1986. If anything, this should have helped reduce the current account deficits. At the same time, however, the Danish krone started a process of real appreciation. Compared with 1984, the krone had appreciated by close to 20 percent in real terms in 1988. The loss of competitiveness induced by this real appreciation, however, contributed to maintaining relatively large current account deficits. We return to the Danish case later on.

VI. The Real Exchange Rate: Endogenous or Exogenous?

We have observed the occurrence of large changes in the real exchange rates of the small EMS countries. To what extent are these changes the endogenous result of the worldwide disturbances of the early 1980s? To what extent are they the result of policy decisions of the countries involved?

It appears that the real depreciation of the Belgian franc was to a certain degree policy-induced. In particular, the Belgian authorities decided in February 1982 to devalue the franc by 8.5 percent. This devaluation was imposed on the authorities by the growing domestic and external imbalances that had accumulated by the end of the 1970s and the early 1980s. Nevertheless, it took a policy decision to devalue the currency.

The Belgian devaluation of 8.5 percent in 1982, however, only explains a small fraction of the total real depreciation of close to 30 percent that occurred during 1979–83. The same is true for the Netherlands, whose currency appreciated nominally but experienced a real depreciation of close to 20 percent. Thus, it is fair to conclude that, at least for these two countries, the real depreciations were largely the results of economic forces that changed relative prices, and in so doing, helped these countries adjust to worldwide disturbances.

VII. The Danish Stabilization Program of 1982

We have observed several times that Denmark has experienced macroeconomic trends considerably different from the Benelux countries since 1982. During the early part of the 1980s (until 1982), the krone depreciated in real terms very much as in Belgium and the Netherlands. Since that date, however, a major turnaround has occurred in Denmark. In 1982, the Danish Government decided to change drastically its monetary and fiscal policies. Up to 1982, the Government had systematically taken the opportunities offered by the frequent intra-EMS realignments to devalue the krone, in order to maintain or to improve competitiveness. This, of course, contributed to the real depreciation of the krone. It also led to a substantially higher inflation rate than in the Benelux countries and in Germany.

The stabilization program of 1982 consisted of three components. First, a decision was made to stop the policy of creeping devaluations and to peg the krone to the deutsche mark. Second, a dramatic change in fiscal policies was instituted with spending cuts and tax increases. Third, capital movements were liberalized.

The effects of this policy package on the Danish economy were quite surprising. Most economists would have predicted that such a policy package, especially the first two components, would be highly deflationary. The opposite occurred. The economy began to boom, leading to a strong increase in total employment, as noted earlier. With the benefit of hindsight, it is now possible to diagnose the elements of this success. The expectations of successive devaluations had led to high interest rates in Denmark prior to 1982. Between 1980 and 1982, the government bond yield hovered around 19 percent, reaching 20.4 percent in 1982. Real interest rates were close to 10 percent. The new exchange rate pegging policy of 1982, which was instituted by a new conservative government, was credible enough to lead to a swift decline in long-term interest rates to 15 percent. In addition, large capital inflows occurred,5 made possible by the capital liberalization program and by the belief of foreign investors that with a 15 percent interest rate (and the credible pegging commitment of the Danish authorities), the krone appeared to be a good investment. The result of lower interest rates and of the added liquidity of the capital inflows was an investment and consumption boom.

The domestic Danish boom came as a surprise to many economists. There was a lot of talk of a “Danish miracle,” especially in Belgium and in the Netherlands. Economists in these two countries were urging their governments to emulate the Danish example. These advocates, however, did not sufficiently recognize that an essential part of the Danish policy package was the liberalization of capital movements, which made this policy unique. Contrary to Belgium and the Netherlands, Denmark had an elaborate system of capital and exchange controls prior to 1982. The abolishment of these controls made the large capital inflows possible and helped generate the investment and consumption boom in Denmark. The fiscal and exchange rate stabilization programs applied by Belgium and the Netherlands from 1982 on could not count on this once-and-for-all stock adjustment effect because the capital flows were mostly free in these two countries.

The domestic Danish boom was not without effects on the real exchange rate. The boom put upward pressure on domestic wages. Since the Danish authorities, as they had announced, maintained the krone/deutsche mark rate unchanged, the result was that the krone started to appreciate in real terms. As mentioned earlier, this real appreciation was instrumental in maintaining relatively large current account deficits. The question is whether this real appreciation of the krone is sustainable. The persistent current account deficits of Denmark raise the spectre of future devaluations of the krone. The market seems to believe that this may happen. In 1989 the government bond yield in Denmark was 10.5 percent, versus 6.8 percent in Germany. This was at a time when the Danish inflation rate had dropped to the German level.

VIII. Real Exchange Rates and Fiscal Policies

One of the more interesting theoretical issues that arises from the analysis of the small EMS countries has to do with the interdependence between fiscal policies and the real exchange rate. We observed that, at least during the 1980s, there was a positive correlation between the government budget deficit and the real exchange rate; that is, when the government budget deficit increased (declined), the currency tended to depreciate (appreciate) in real terms. This is somewhat surprising. As frequent users of the standard Mundell-Fleming model, we are accustomed to associating growing budget deficits with real currency appreciations.

The puzzle is only superficial, however. As was argued earlier, the real depreciation of the small EMS currencies during 1979–83 was the result of external disturbances that led to an endogenous increase of the budget deficits. The rising budget deficits cannot be interpreted as discretionary fiscal expansions (like U.S. fiscal policy during 1981–83), which, in the Mundell-Fleming model, produce a real appreciation.

The more recent fiscal policy developments, however, cannot easily be explained in the framework of the Mundell-Fleming model. We have observed that since 1982, Belgium and the Netherlands introduced policies of fiscal contraction. This coincided with a stabilization of the real exchange rates of these countries. In Denmark, the restrictive fiscal policies—which even led to a surplus of the government budget—were associated with a real appreciation of the Danish krone.

In order to understand the Belgian and Dutch experience one would have to invoke other external disturbances. The experience of these countries seems to indicate that the relation between fiscal policies and the real exchange rate is considerably more complex than standard theory predicts. In particular, the role of expectations about future policies and the sustainability of present policy initiatives seem to be important. They are also difficult to model and complicate the prediction of the effects of fiscal policies on the real exchange rate.

IX. Conclusion

Real exchange rates of the small EMS countries have exhibited large fluctuations since the institution of the EMS. Countries such as Belgium and the Netherlands experienced real effective depreciations of 20 to 30 percent during the last decade.

In this paper we have analyzed the sources of these real exchange rate movements. We have argued that in the case of Belgium and the Netherlands the real depreciations that occurred during 1979–83 were mostly an internal response to worldwide disturbances. These two countries were particularly sensitive to these disturbances and allowed their real exchange rates to adjust to these shocks. Although it took a considerable amount of time, these real exchange rate movements were instrumental in moving these economies toward a path of economic recovery.

We also compared the Benelux experience with that of Northern Germany, which in many ways exhibits similar economic and industrial structures. We found that the worldwide disturbances of the early 1980s affected Northern Germany in ways similar to their effects on the Benelux region.

In contrast to the Benelux region, however, the employment recovery in Northern Germany has been rather slow. This suggests that the lack of real exchange rate flexibility of Northern Germany may have made the adjustment in that region slower to come about, despite the fact that, presumably, the degree of labor mobility between Northern and Southern Germany is larger than between the Benelux countries and the rest of the world. Further research will be necessary to substantiate this hypothesis.

The Danish exchange rate experience has diverged from the Benelux experience in notable ways. During 1979–82, the Danish krone depreciated in real terms in much the same way as the currencies of the Benelux countries. From 1983 on, however, a remarkable economic recovery in Denmark was made possible by a policy package of fiscal consolidation, exchange rate pegging, and liberalization of capital movements. This contributed to a domestic boom, but also to a substantial real appreciation of the Danish krone.

The most surprising aspect of these real exchange rate changes is their size. Despite the fact that these countries are part of an exchange rate union, their real exchange rates have exhibited large and sustained movements. Many of these movements, as in Belgium and the Netherlands, have helped these countries achieve macroeconomic equilibrium. Sometimes, as in the case of Denmark during 1982–88, these movements were the result of domestic policies and succeeded in redrawing the domestic economic landscape.

This large real exchange rate flexibility of EMS member countries may also explain the resilience and the success of the system. It has allowed its members to weather the storms produced by large and worldwide economic disturbances. Since such external shocks can repeat themselves, it is likely that EMS member countries may continue to require relatively large real exchange rate flexibility.


Daniel Gros

The main thrust of the paper by De Grauwe and Vanhaverbeke is, in my reading, that the real exchange rates of the small member countries of the European Monetary System (EMS) are endogenous and were driven in the 1980s mainly by external shocks. Fiscal policy does not appear to have been an important independent determinant of the real exchange rate. Rather, it seems to have been “passive,” in the sense that governments allowed the automatic stabilizers to work, leading to large deficits when the shock hit in the early 1980s.

Since I basically agree with this scenario I will comment on five specific aspects of the paper that could be clarified further:

  • (1) the appropriate time horizon for exchange rate policy;

  • (2) the description of the exchange rate policy followed by the small EMS countries;

  • (3) the role and actual behavior of fiscal policy;

  • (4) the evidence for the shock that is the basis of the story; and

  • (5) the implications of the variability of real exchange rates for European monetary integration.

Time Horizon for Exchange Rate Policy

The paper analyzes exchange rate “experiences” looking at a ten-year (and sometimes four-year) period. I find it hardly surprising that over such a medium- to long-run period one would find that the nominal exchange rate does not matter much for any real variable (the real exchange rate or the current account, for example). I also prefer to look at a medium- to long-run horizon, but is this the appropriate time horizon for considerations about necessarily nominal exchange rate policy?

The appropriate time horizon over which to look at exchange rate policy is rather the one where the Phillips curve is not vertical so that the standard Mundell-Fleming approach (both associated with the International Monetary Fund framework) can be used. This is the time horizon over which the exchange rate can be a useful shock absorber, along the lines discussed in the survey paper by Victor Argy.

If the main issue in exchange rate policy is the use of the nominal exchange rate to neutralize—at least partially—the effects of various short-run shocks on income and inflation (or any other policy targets), the authors have missed it by looking at an excessively long-term horizon.

Exchange Rate Policy of the Small EMS Countries

Given the theme of the conference, one would expect a paper discussing the “experience” of the EMS countries to provide a picture of the exchange rate policy actually followed. The paper does this only to a limited extent since it provides data only about effective exchange rates, whereas the policy of these countries, as recognized by the authors, was and is formulated vis-à-vis the deutsche mark. (This was true already at the time of the snake, that is, before the EMS.) Since this description of the actual exchange rate policy of these countries is missing, let me provide a brief description of the deutsche mark exchange rate of these countries. Charts 1-3 below depict the nominal bilateral deutsche mark exchange rate and the real effective exchange rate of the three small EMS countries considered. These figures reveal that we have really three quite distinct cases:

Chart 1.Nominal Deutsche Mark/Belgian Franc Exchange Rate and Real Effective Rate of Belgium, Denmark, and Netherlands

(1979 = 100)

Source: OECD, Economic Outlook; and IMF, International Financial Statistics.

Chart 2.Nominal Deutsche Mark/Netherlands Guilder Exchange Rate and Real Effective Rate of Belgium, Denmark, and Netherlands

(1979 = 100)

Source: OECD, Economic Outlook; and IMF, International Financial Statistics.

Chart 3.Nominal Deutsche Mark/Danish Kroner Exchange Rate and Real Effective Rate of Belgium, Denmark, and Netherlands

(1979 = 100)

Source: OECD, Economic Outlook; and IMF, International Finanical Statistics.

Belgium. From 1975 to 1981 the franc exhibited a slow, rather constant rate of crawl; a 25 percent depreciation (against the deutsche mark obviously) in two years, 1981–83; and from 1983 on a slower, constant rate of crawl.

Denmark. The krone showed a more variable rate of depreciation throughout the entire period under discussion. It is difficult to detect any clearly defined subperiod, except that the rate of crawl has slowed since 1983.

The Netherlands. Between 1975 and 1979, the guilder exhibited a constant rate of slow crawl (about 10 percent in four years); it stayed constant after 1979, except for a small jump in 1983.

For Denmark and Belgium a comparison of the behavior of the real effective exchange rate and the nominal deutsche mark exchange rate suggests that a depreciation against the deutsche mark was generally equivalent to a depreciation of the real effective exchange rate. This should somewhat soften the “surprising” finding of the authors that there have been large real exchange rate changes despite the fact that these countries are part of an exchange rate union. A large part of the observed real exchange rate changes is due to the fact that the exchange rate union was not that tight, especially until 1983 when rather large changes in nominal exchange rates against the deutsche mark were still permitted.

The authors have rightly focused on the effective real exchange rate since that is what matters for trade. They should, however, have also taken into account the behavior of the real effective deutsche mark exchange rate. It is easy to achieve a real effective depreciation if you peg the nominal exchange rate to a currency that is also depreciating in real terms. This effect does not operate over the entire 1979–88 period, since Chart 2 in the paper shows that the real effective exchange rate of the deutsche mark returned in 1988 exactly to its 1979 level. In the meantime, however, the deutsche mark first depreciated in real effective terms by about 10 percent and then had a real appreciation of about the same order of magnitude. The finding of the authors that the real exchange rate varied so much inside the EMS should therefore be qualified, at least for their Table 1, which shows the real depreciation achieved by these three countries in 1979–83.

The Role and Behavior of Fiscal Policy

With respect to the role and behavior of fiscal policy two remarks are in order:

  • (1) From a theoretical point of view the relationship between fiscal policy and the real exchange rate is not as clear as maintained by the authors. Frenkel and Razin (1987) show that this relationship can go either way and can vary in strength depending on the nature of the fiscal policy; for example, its bias toward nontradables or whether it is permanent or temporary. More information on this would have been useful.

  • (2) From an empirical point of view it is well known that it is difficult to measure the stance of fiscal policy accurately. The budget deficits used by the authors are the standard measure, but I would argue that by looking at a measure that takes into account inflation one obtains a somewhat different picture. Charts 4-6, which show an inflation-adjusted deficit for the large and small EMS countries, suggest that within the group of small EMS members Belgium and the Netherlands follow a similar path. Denmark, on the other hand, shows a dramatic improvement, with an adjusted surplus of about 4 percent of GDP in 1988. This is in contrast to Chart 9 of the paper, which suggests that the Netherlands is closer to Denmark than to Belgium.

Chart 4.Inflation-Adjusted Budget Deficit

(as percent of GDP; at market prices)

Source: European Economy.

Chart 5.Inflation-Adjusted Budget Deficit: Large EMS Countries

(as percent of GDP)

Source: European Economy.

Chart 6.Inflation-Adjusted Budget Deficit: Small EMS Countries

(as percent of GDP)

Source: European Economy.

Evidence for the Shock

The main message of the paper, in my view, is that the small EMS countries were hit in the early 1980s by an adverse real shock to which they had to adjust whatever the nominal exchange rate policy. I sympathize with this view, but instead of just invoking a real shock to explain “the residual,” the authors should have provided us with more evidence about this shock.

The obvious candidate is, of course, the second oil shock. (The timing of the real depreciation in Belgium is then somewhat awkward since it comes only in 1981–83.) But even so, the authors might have used some data about industrial structures, for example, to show that there was indeed reason to believe a priori that these countries were so different from the rest of Europe (or just from the large EMS countries) that they needed a special adjustment.

Implications of Real Exchange Rate Variability for European Monetary Integration

An important and undisputable result of the paper is that in the EMS, real exchange rates have been more variable for the small countries and have diverged at times considerably from those of the large countries. This is important since it suggests that even in an area of fixed exchange rates—a monetary union—real exchange rates can move and thus provide a shock absorber.

What I also find extremely interesting is the comparison between Belgium and Northern Germany, since it could give us some idea of the “value” of the nominal exchange rate as an adjustment instrument. However, the authors should not have discussed only similarities in unemployment rates. They should have taken into account the other factors that differentiate Northern Germany from a country like Belgium; for example, the fact that wages are not permitted to differ much across regions in Germany, and the importance of inter-regional fiscal transfers within Germany.


The label “small EMS countries” refers to Belgium, the Netherlands, and Denmark, but excludes Ireland.

Such a policy-induced real depreciation could also be the result of a nominal revaluation, which is lower than the (favorable) inflation differential of the home country with the rest of the world. The example could be the Netherlands, or Germany during the 1980s.

There are, of course, other reasons for the improvement in employment. In particular, “supply-side-friendly” policies may have helped.

Despite a current account deficit of more than $1 billion in 1983, the foreign exchange reserves of the Danish National Bank increased from $2.2 billion to $3.7 billion.

    Other Resources Citing This Publication