Chapter

In the Shadow of the Mark: Exchange Rate and Monetary Policy in Austria and Switzerland

Author(s):
International Monetary Fund
Published Date:
September 1990
Share
  • ShareShare
Show Summary Details
Author(s)
Hans Genberg

I. Introduction

This paper reviews the macroeconomic experience of Austria and Switzerland in light of the different monetary policy strategies adopted by the monetary authorities in these countries since 1973. The Austrian National Bank continued to pursue an exchange-rate-based policy after the breakdown of the Bretton Woods system but changed from a dollar peg to a deutsche mark peg in an attempt to import German price stability. The Swiss National Bank, on the other hand, opted for a freely floating exchange rate and aspired to achieve price stability through a policy of monetary targeting.

The principal question that I seek to answer in this paper is whether this difference in approach to monetary policy had a notable impact on the macroeconomic evolution of the two economies. To this end, the next section briefly reviews the theory of exchange rate targeting versus monetary targeting as optimal strategies. This theory is used to try to explain why Austria and Switzerland have adopted such different approaches to monetary policy. In the subsequent section, I review the macroeconomic evolution of the economies and relate it to the chosen policies. The concluding section provides an answer to the question “Does the choice of strategy matter?”

II. Exchange Rate Targeting versus Monetary Targeting The Strategies of Austria and Switzerland

Throughout the 1960s both Austria and Switzerland adhered to the rules of the Bretton Woods system of fixed exchange rates.1 The U.S. dollar rates of the schilling and the franc were fixed around parities of 26.00 schilling/dollar and 4.37 SFr/dollar, respectively.2 One result of this exchange rate policy was that both countries shared in the worldwide inflation that picked up in the second part of the 1960s and continued into the 1970s.

Having freed themselves in 1973 from the constraint of fixing their exchange rates relative to the dollar, the central banks in both countries were able to pursue their own objectives, the principal, if not the sole, being internal price stability.3 To achieve this objective the monetary authorities in both countries adopted rules governing their policies, rather than relying on discretionary actions. The contents of the rules differed widely, however. The Austrian National Bank opted for a “hard currency policy” consisting of fixing the value of the schilling to the deutsche mark, whereas the Swiss National Bank chose to target a monetary aggregate.4 In the next section the theory of the optimal choice of monetary policy instrument in an open economy is used to explain the reason behind the different choices made by the two central banks.

Theoretical Rationalization5

It is generally accepted that the monetary authorities in an open economy cannot control both the exchange rate and the money supply simultaneously. Which of the two variables to choose as an instrument of policy is a question that has been treated in the literature under headings such as fixed versus flexible exchange rates, the optimum currency area problem, and optimal exchange rate management. In addition, whether the authorities should adopt rules for monetary policy—such as a fixed exchange rate or a pre-announced rate of growth of the money supply—or try to choose optimal settings of their policy instrument period by period has been discussed in the rules versus discretion debate. I will summarize the principal conclusions of these strands of analysis in this section as a background to the presentation of the macroeconomic consequences of the choices of policy strategies made by Austria and Switzerland.

Much of the discussion of the choice between an exchange-rate-based or a money-supply-based policy is conducted in a setting where short-run stabilization of output growth is the main objective of policy. The choice of instrument is then dependent on which of the two possibilities is likely to insulate the economy best from unanticipated disturbances. Following Poole’s (1970) analysis of the choice between money-supply and interest rate control in a closed economy, the answer depends on the sources of the shocks in the economy. Suppose for example, that there is an unanticipated increase in the demand for money. With a constant money supply the domestic currency would appreciate and the interest rate would increase. As a consequence of partially rigid prices—an assumption that is justified by the short-run nature of the analysis—domestic output will contract. On the other hand, if the exchange rate were held fixed, the adjustment to the money demand disturbance would have entailed a corresponding change in the supply of money with no change in output. In this case a fixed exchange rate would be superior to a floating rate. Analogous arguments would show that holding the money supply constant and letting the exchange rate adjust is preferable when the disturbance is such that long-run adjustment requires a change in the real exchange rate. Examples of such disturbances would be a change in domestic fiscal policy or a change in foreign demand for our exports.

Based on the above analysis one would expect two countries that are faced with similar shocks to choose the same monetary policy strategy. A priori, it seems that Austria and Switzerland fall in this category, as far as external shocks are concerned. With respect to internal shocks, it is not clear how the balance falls. On the one hand, it can be argued that Austria is potentially subject to more frequent fiscal shocks than Switzerland in view of the active use of this policy instrument there (see discussion of current account developments). On the other hand, changes in financial regulation have also been more frequent in Austria.6 Since it is the relative size of fiscal and monetary disturbances that is important for the choice of monetary policy instrument, it is again not clear that there is any significant difference between the countries. It appears then that this type of analysis cannot explain why Austria chose an exchange-rate-based monetary policy and Switzerland a money-supply-based policy when they were faced with the choice at the end of the Bretton Woods era.

A different strand of analysis has stressed inflation control, rather than short-run output stabilization, as a basis for the choice between monetary targeting and exchange rate targeting. In the simplest case, where a country wants a different rate of inflation than the rest of the world, the choice is clear. Only a flexible exchange rate regime is possible. This consideration explains why both Austria and Switzerland stopped pegging to the dollar in 1973, but it does not necessarily explain why the Austrians decided to continue an exchange-rate-based policy while the Swiss adopted monetary targeting. As we have seen, price stability was the main objective of the monetary authorities in both countries, and since the German authorities also pursued the same goal, the option of fixing to the deutsche mark was consistent with the policy goals of both countries.

A beginning of an explanation to the different choices can be found in the recent literature on rules versus discretion and the time-inconsistency problem. This literature stresses the interaction between the policymaker’s desire to stabilize the economy, given current prices and wages, and the expectations of the private sector about how the adopted policies will affect prices and wages in the future. In certain situations it can be shown that the inflation-output outcome in the economy is superior if the authorities pre-commit to a noninflationary policy rather than if they choose what seems to be an optimal policy, period by period. In other words, time-inconsistency considerations can be used to make a case for policy rules rather than complete discretion. But rules must be credible if they are to be believed by the private sector and have the desired effects. For example, a central bank that is under pressure from the Treasury to finance budget deficits may not easily be able to convince the public that it intends to follow a policy of strict monetary targeting. A strategy that involves limitations on the Treasury’s ability to force the hands of the central bank would be more credible in this situation. A fixed exchange rate is such a policy since it is typically based on a commitment of the Treasury as well as of the central bank. It can be argued that circumstances in Austria are not unlike those just described, which would explain the choice of tying the schilling to the deutsche mark as part of an anti-inflation strategy.7

The Swiss National Bank is quite independent of the fiscal authorities. For this reason the credibility of a strategy of monetary targeting may be more easily established and maintained. A fixed exchange rate would also, presumably, be credible. Why the Swiss authorities chose the former may perhaps be explained by a judgment that their own credibility was easier to establish independently than by relying on the Deutsche Bundesbank, which could be forced by its links with countries of the European Community (EC) to deviate from its preferred anti-inflation policy.

III. Macroeconomic Policies and Their Consequences

Both the Austrian National Bank and the Swiss National Bank adopted price stability as their main policy objective once the constraint of fixing their exchange rates relative to the dollar was relinquished. They pursued this objective in rather different ways, however, but the outcome was largely similar. In this section I review that outcome by discussing and comparing the evolution of the main macroeconomic variables of interest. The discussion is set against the background of the evolution of external influences that are critically important for both countries in view of their high degree of openness.

Inflation

Chart 1 shows the evolution of the rate of consumer price inflation. Three aspects of that evolution are worthy of special attention. First, while both Austria and Switzerland followed the average in the countries of the Organization for Economic Cooperation and Development (OECD) until 1974, the Austrian and Swiss inflation rates have subsequently been below that average quite consistently. Viewed in this international perspective then, both countries have done well in their quest for price stability.

Chart 1.Inflation

Second, the reduction in the rate of inflation was more rapid and went much further in Switzerland than in Austria during the period between the two oil shocks. This difference can be explained partly by the more restrictive monetary policy followed by the Swiss National Bank during this period and partly by the expansionary fiscal stance adopted in Austria (see discussions of monetary policy and current account developments).

The third and final point to take note of is that since 1979 the inflation rates in Austria and Switzerland have been quite similar to each other and to that of Germany.8

Exchange Rates

Austria did not opt for letting the schilling float independently after the end of the Bretton Woods regime. Instead it began the post-1973 period by pegging to a trade-weighted basket of European snake currencies and the Swiss franc. Given the importance of Germany as a trading partner, this arrangement gave the deutsche mark a weight of over 50 percent in the basket, and the policy became known as the hard-currency policy. As most snake currencies weakened relative to the mark during the 1970s, the schilling/ deutsche mark rate increased as well (Chart 2).9 In response, the Austrian National Bank modified the exchange rate policy to make it into a pure deutsche mark peg. Thus from 1980 on, the schilling has stayed within a very narrow band around the deutsche mark.

Chart 2.Exchange Rates

The policy of fixing to the deutsche mark explains the convergence of the Austrian inflation rate toward the German rate, as discussed in the previous section. However, since the convergence has not been perfect, the schilling has appreciated slightly in real terms vis-à-vis the deutsche mark (Chart 2). On a trade-weighted basis against 12 major industrialized countries on the other hand, there is virtually no trend in the real exchange rate (Chart 2), although some short- and medium-term fluctuations are present.

In contrast with the Austrian authorities, the Swiss National Bank has not had an official exchange rate policy since 1973, except for a one-year period in 1978–79.10 That period was preceded by a sharp appreciation of the franc relative to the U.S. dollar that started in mid-1977 and lasted until the fall of 1978. Faced with strong pressure to break the appreciation to avoid serious damage to the export sector, the Swiss National Bank dropped its monetary targeting strategy and adopted instead an exchange-rate-oriented monetary policy that was explained in terms of the authorities’ desire to keep the value of the deutsche mark clearly above 0.80 Swiss francs.11 This exchange rate target was backed up by a switch to a highly expansionary monetary policy, and the foreign exchange market reacted immediately (see Chart 2). As the deutsche mark rose throughout 1979 and the first half of 1980, the Swiss National Bank again felt compelled to express its view on the appropriate foreign exchange value of the franc. It did so by stating that 0.95 Swiss francs/deutsche marks was a defensible upper limit. Although

no explicit exchange rate targets have been set since this episode, the franc has moved in the 0.80-0.95 range nevertheless.12 An interesting question, to which we shall return later, is to what extent the relative stability of the Swiss franc/deutsche mark rate since 1981 is the consequence of the “acceptable” band announced by the authorities in 1978–79.

The real exchange rate relative to Germany and the multilateral real rate shown in Chart 2 both display long-term movements similar to their corresponding Austrian variables, namely, a relatively continuous appreciation throughout the period represented in the chart in the first case, and an appreciation until the mid-1970s followed by no particular trend in the second. The two countries differ substantially, however, in terms of short- and medium-term variability in the real rates, which is substantially larger in Switzerland. This is, of course, a reflection of the greater variability of the nominal Swiss franc/deutsche mark rate compared with the nominal schilling/deutsche mark rate.

Monetary Policy13

While the “hard-currency policy” is the proximate reason for the convergence of Austria’s inflation rate to Germany’s, it is clear from the description of Swiss exchange rate policy that the cause for the reduction in Switzerland’s inflation rate has to be sought elsewhere. Rather than pursuing the goal of price stability by tying the value of the franc to the deutsche mark, the Swiss National Bank has adopted what has been described as “pragmatic monetarism.”14 Monetary aggregates have been used as intermediate targets of policy, informally in 1973 and 1974 and formally in the sense of preannouncing target growth rates either for Ml or for the monetary base since 1975, with a temporary discontinuation in 1978.

Chart 3 (top panel) displays the growth rate of Ml in each month measured as the percent change relative to 12 months earlier together with the announced target rates. It shows clearly the sharp reduction in monetary growth that already started in the first part of 1972. This reduction was so strong that the level of the money supply was essentially unchanged during 1973-74. In the beginning of 1975 the authorities announced a target growth rate for Ml of 6 percent.15 The policy of announcing target growth rates for Ml continued in 1976–78, with targets of 6, 5, and 5 percent, respectively. As can be seen, actual growth rates were not far from the targets until the beginning of 1978. The spectacular rise in the value of the Swiss franc during that year induced the authorities to abandon the monetary target in favor of an exchange rate target. Apparently judging that the stability of the demand for money had been damaged owing to the volatility of exchange rate expectations, the Swiss National Bank did not return to monetary targeting again until the end of 1979. At the same time, it also switched from using Ml to the monetary base as the target variable. It was felt that the demand for the monetary base was less subject to shifts because of exchange rate expectations than Ml, and that it was thus a more useful aggregate to target. Since this episode, the policy of monetary targeting has been maintained uninterrupted.

Chart 3.Money Growth

The success in reaching and then maintaining a relatively low inflation rate in Switzerland is no doubt attributable in large part to the monetary targeting strategy adopted by the Swiss National Bank. The Bank’s implicit focus on the medium- to long-run behavior of the monetary aggregates convinced the public to ignore short-term deviations from the targets. As Chart 3 (top panel) illustrates, these deviations were often quite large.16 However, since the errors have not been allowed to accumulate, the level of the money supply has tended to return to a path defined by the cumulation of the target growth rates (see Chart 3, middle panel).17 This in turn has made it unnecessary for the Swiss National Bank to try to prevent or to justify short-term deviations. Furthermore, the convergence toward the target path in the long term has probably reinforced the credibility of the anti-inflationary stance of the Bank, which in turn has made the policy easier to carry out.

Fortuitous external and internal factors, however, may also have contributed to the accomplishments of monetary policy. First, the pace of innovation in domestic financial markets has been very slow (Béguelin, 1984), partly because low inflation and low nominal interest rates have made such innovations less profitable.18 Severe instability in the relationship between the monetary base and the ultimate inflation objective has thus been avoided.19 Second, apart from two or three episodes, there has been little conflict between attainment of the monetary target and exchange rate movements. This is due probably to the fact that Germany has been pursuing roughly similar anti-inflation policies, and the suspicion that a return to exchange rate targeting would take place if the deutsche mark/Swiss franc rate should start to move substantially. Germany has thus performed the role of buffer against volatility in international financial markets.

The behavior of monetary aggregates in Austria is determined by the fixed exchange rate policy adopted by the Austrian National Bank. Theoretical considerations would dictate that German monetary policy would in these circumstances also govern Austrian money growth. This prediction is borne out by the data on annual growth rates of M1 in both countries, as shown in Chart 3 (bottom panel). Apart from an episode in 1979, the paths of the two series are remarkably close.20 It can be concluded that the quantity of money in Austria is demand-determined and unlikely to play a causal role for other variables in the economy.

Interest Rates

International comparisons of interest rate movements can provide valuable information on the degree of financial integration of an economy with external financial markets. For example, the relationship between Euromarket and domestic interest rates has been used to evaluate the extent of controls on capital movements. Applying this criterion, Cosandier (1987) argues that the domestic German money market has for practical purposes been perfectly integrated with the external deutsche mark money market since 1974, which signifies the absence of administrative controls against arbitrage since that date.21 Likewise, he shows that the internal and external market for Swiss franc deposits has been fully integrated since 1975.22

From this evidence we can conclude that both Germany and Switzerland have maintained open financial markets during most of the post-Bretton Woods period. It does not of course follow that German and Swiss interest rates will move together given the floating Swiss franc/deutsche mark exchange rate. Indeed, part of the attraction of a flexible exchange rate is the independence it bestows on monetary policy, part of which presumably is reflected in a decoupling of local interest rates from foreign-currency interest rates.

This decoupling is illustrated in Chart 4 (top panel), which displays quarterly averages of three-month Euro-deutsche mark and Euro-Swiss franc deposit rates. The sometimes rather substantial differential would not have been possible if the Swiss National Bank had pursued a fixed exchange rate policy vis-à-vis the deutsche mark. But the chart also illustrates the tendency of the two interest rates to follow broadly similar medium-term developments. This can perhaps be ascribed in part to the similar medium-term monetary policy strategies followed in the two countries, but formal empirical analysis shows that it is a more general phenomenon. Genberg and Salemi (1987) studied the determinants of Swiss short-term interest rates and found that for the 1973–81 period, between 50 and 60 percent of their variance could be accounted for by shocks to the U.S. Treasury bill rate, whereas only about 10 percent was due to shocks to the Swiss money supply. For reasons that we shall return to below, the Swiss National Bank has apparently allowed foreign interest rate movement to be transmitted even during the floating rate period.

Chart 4.Short-Term Interest Rates

In view of the close relationship between the Austrian schilling and the deutsche mark, one would expect Austrian interest rates to follow rather closely the corresponding German rates, provided capital was free to move between the two countries and the fixed exchange rate was expected to be maintained. Chart 4 (lower panel) suggests that one or both of these conditions for complete interest-rate equalization have not been totally fulfilled. Except during certain infrequent and brief episodes, schilling rates have been higher than corresponding deutsche mark rates, indicating either that the markets have consistently expected a depreciation of the schilling despite the record of stability, or that restrictions on capital inflows have been in place limiting the ability of foreign investors to take advantage of the higher Austrian interest rates. Consistent with the former explanation is the fact that the interest rate differential has narrowed since the early 1980s when the exchange rate policy has been expressed explicitly as a deutsche mark peg and may therefore have been regarded as more credible.23 Furthermore, estimated offset coefficients suggest that the National Bank enjoyed rather limited autonomy in the 1980s, indicating that controls could not have been very effective.24 On the other hand, one suspects that limits on inward movements of capital were prominent in the episode of high domestic rates from the end of 1976 to the end of 1978. This suspicion is based on the fact that monetary growth in this period was quite different from that in Germany (see Chart 3),25 and that offset coefficients estimated for this period are substantially smaller.26

Industrial Production

One of the important issues surrounding disinflation policies is the extent to which a reduction in inflation is obtained at the cost of reduced output. In this connection it is relevant to ask whether an exchange-rate-based strategy as adopted in Austria generates a different inflation-output trade-off than a monetary-targeting strategy such as that followed in Switzerland. Chart 5 offers preliminary evidence suggesting that the difference between the strategies, if it exists at all, is not large once the effects of external influences on output growth are taken into account. The chart shows the deviation of the logarithm of an index of industrial production in each country from a country-specific linear trend.

Chart 5.Industrial Production*

*Deviations from trend.

It is clear from the evolution of these deviations that both Austria and Switzerland are highly dependent on the state of the business cycle in Germany.27 For this reason it seems a priori unlikely that domestic policies—whether related to the goal of price stability or otherwise—are responsible for much of the fluctuations in industrial production in these countries.28

Did the strategy of monetary targeting followed by Switzerland lead to a smaller or greater loss of output than the exchange-rate strategy of Austria during the period following the break-up of the Bretton Woods system? A superficial glance at Chart 5 might suggest that the Swiss strategy was costlier, but such a conclusion would be hasty. First of all, it must be kept in mind that the Swiss disinflation was much quicker and went further than in Austria (see Chart 1). Second, it is not obvious that the sharper contraction in Switzerland during 1975–77 is due to domestic policy rather than, for example, to a greater sensitivity to the 1973 oil shock. More discriminating tests must be used to settle this question, which we shall return to later on.

Current Account Developments

It has been suggested29 that a consequence of Austria’s hard-currency exchange rate policy has been a loss of competitiveness and the emergence of an external deficit, particularly in the early-to mid-1970s. According to Chart 2, the schilling did indeed appreciate in real terms, both vis-à-vis the deutsche mark and on an effective basis, by some 10–15 percent between 1970 and 1975, mostly after 1973. Over about the same period, the current account deteriorated from a roughly balanced position in 1970–71 to a deficit of about 4 percent of GDP in 1975–76 (see Chart 6).

Chart 6.Current Account and Fiscal Balance

(as percent of GDP)

But it does not follow that the coincidence of these two phenomena necessarily reflects a case of cause and effect. At least two pieces of evidence suggest that it might not. The first relates to the stance of fiscal policy that was deliberately expansionary during 1973–76 (with the exception of the first part of 1974).30 The plot of the Austrian fiscal balance in Chart 6, which represents the federal fiscal balance as a percent of GDP, shows that this expansion coincided with the deterioration in the external balance—as theory would predict. According to this explanation, fiscal policy brought about both a current account deficit and a deterioration of competitiveness, owing to pressure on domestic prices.31 If this explanation is correct, it is unlikely that a floating exchange rate would have helped. Mainstream, open-economy macroeconomic theory—as well as the experience of the United States in the early 1980s—has taught that expansionary fiscal policy with a flexible exchange rate is likely to bring about a deterioration of the current account and an appreciation of the domestic currency, exactly the problems encountered by Austria during 1973–77.

A second piece of evidence casting doubt on the hypothesis that exchange rate policy was the source of the Austrian external deficit is the evolution of the real exchange rate and the current account in Switzerland. As Charts 2 and 6 (lower panel) make clear, the appreciation of the franc was even sharper than that of the schilling, yet the Swiss current account did not show any sign of deterioration.32 The absence of a fiscal imbalance in Switzerland is quite likely the reason for the clear difference.

IV. Conclusions: Does the Choice of Strategy Matter?

The analysis in the previous section suggests some tentative conclusions concerning the consequences of alternative monetary policy strategies. First, it is clear that the approaches of both the Austrian National Bank and the Swiss National Bank succeeded in achieving their main objective, namely a substantial and durable reduction in inflation. In this respect there is little difference between the two countries apart from the fact that in the initial phase of the adjustment, the Swiss inflation rate was reduced faster and by a greater amount than the Austrian rate. This would illustrate the advantage of independent monetary targeting, provided there were no adverse side effects.

To what extent did the outcome for other variables differ between the countries? It has been argued that the reduction in output during the 1974–75 disinflation was somewhat larger in Switzerland, but that it is unclear to what extent this was associated with differences in monetary policy since Austrian fiscal policy was quite expansionary in that period. Furthermore, if adjustment is made for the sharper reduction in the Swiss inflation rate, it would be difficult to argue that the output cost of disinflation differs substantially between the Swiss monetary-targeting approach and the Austrian hard-currency policy.

As expected, given that the schilling has been fixed to the deutsche mark, the evolution of the main macroeconomic variables in Austria—such as industrial production and inflation—has been closely correlated with developments abroad, especially in Germany. The fact that the Swiss franc has been allowed to float freely (most of the time) has not, however, insulated the Swiss economy from external shocks. Formal empirical evidence, as well as casual inspection of the relevant data series, show that the Swiss economy also has remained highly dependent on external events—during the flexible rate period since 1973. If there were ever a belief that a floating exchange rate would provide insulation for a small open economy, it should be dispelled by the Swiss experience.

One difference that has often been noted in comparisons of fixed and flexible exchange rates is the greater variability of real exchange rates when the nominal rate is floating.33 This is also present in the two countries studied here. It is less clear, however, what the real consequences of the greater variability of exchange rates have been. Genberg (1986) shows that the variability of exchange rates is positively correlated with the dispersion of relative prices. This could be a sign that greater exchange rate variability brings about a greater variance of relative price changes, which in turn might have undesirable effects on resource allocation. But the causality could also be the opposite, in which case no particular welfare implications could be drawn.34

To recapitulate, it does not appear that the different monetary policy strategies chosen by Austria and Switzerland have in and of themselves led to different outcomes as far as the real developments in the respective economies are concerned. This would not surprise those who believe that real outcomes are independent of the nominal exchange rate regime. It does, however, cast some doubt on the applicability of that strand of the literature on fixed versus flexible rates that emphasizes short-run output stabilization.

Generalizations to other countries based on the comparison between Austria and Switzerland may have to be tempered by two considerations. First, both Austria and Switzerland have pursued their respective strategies with a great degree of firmness; this presumably increased the credibility of their policies over time.35 The outcome might easily be different for countries that show more ambiguity in their resolve to achieve their objectives.

A second consideration that has to be taken into account is the fact that both countries have strong economic ties with Germany and that Swiss and German monetary policy objectives have been quite similar. This has reduced the effective difference between the monetary policies in Switzerland and Austria, despite the different strategies followed.36

Comment

by Georg Rich

Hans Genberg offers a comprehensive analysis of monetary policy in Austria and Switzerland. He concludes that the monetary policy performance of the two countries has been similar, despite the fact that they have pursued different policy strategies. Both countries have managed to maintain a high degree of price stability and have thus come close to achieving their principal monetary policy objective. Moreover, the Swiss experience suggests that floating exchange rates have not provided much insulation from foreign shocks. Even so, the Swiss National Bank—in the mid-1970s—was able to reduce the domestic inflation rate more quickly than the Austrian monetary authorities. Movements in output, on the other hand, were similar in the two countries. In Genberg’s view, changes in Austrian and Swiss output were triggered by common foreign—rather than domestic—disturbances, and were thus highly correlated.

Although I agree with the thrust of Genberg’s reasoning, I nonetheless submit that a comparison of Swiss and Austrian policy experience is more complex than might appear on the basis of his paper. I would like to comment on five issues raised by Genberg.

First, I share Genberg’s view that the theory of optimum currency areas and exchange rate management does not seem to answer the question of why Austria chose an exchange-rate-based monetary policy and Switzerland a money-supply-based policy. However, this type of analysis is not completely fruitless. It does show why Austria was led to peg the exchange rate of the schilling to the deutsche mark. This choice was obvious because of Austria’s close trade links with the Federal Republic of Germany. Switzerland’s international trade, by contrast, is much more diversified. If Switzerland were to adopt a fixed exchange rate, the choice of an optimum peg would be more intricate than in the Austrian case. Another difference between the two countries arises from the fact that Switzerland features a major financial center, unencumbered by restrictions on international capital flows.

Second, as suggested by Genberg, an important reason why the Swiss National Bank chose a money-supply rule was its desire to establish credibility independently of the Deutsche Bundesbank. But the Swiss National Bank was also convinced that a floating exchange rate would widen its scope for conducting an autonomous monetary policy. The exchange rate problems of 1978—which Genberg discusses in some detail—made the Swiss National Bank aware of the constraints on its policy autonomy. Although Switzerland’s policy autonomy is far from complete, it appears to be larger than Genberg’s analysis indicates. I will return to this issue below.

Third, Genberg provides an interesting analysis of exchange rate developments in Austria and Switzerland. He shows that the policy of pegging the schilling to the deutsche mark, as expected, caused the Austrian inflation rate to converge with that of Germany, but convergence was not complete. In other words, the schilling appreciated slightly in real terms vis-à-vis the deutsche mark. In the case of the Swiss franc, the real appreciation was even more pronounced. As far as I know, the reasons for this real appreciation are not clearly understood and should be carefully investigated. In the Swiss case, the real appreciation of the domestic currency raises further doubts about the wisdom of pegging the Swiss franc to the deutsche mark. Considering exchange rate developments in the past, with a fixed Swiss franc/deutsche mark exchange rate, the Swiss National Bank would have imported some inflation from abroad even if the Bundesbank had managed to keep the German inflation rate near zero. For this and other reasons, the Swiss National Bank—though prepared to counteract extreme exchange rate movements—has been loath to announce target zones for the exchange rate.

Fourth, the least convincing part of Genberg’s paper is that on interest rates. In my opinion, the greatest difference between the Austrian and Swiss policy regimes lies in the behavior of interest rates. While short-term Austrian interest rates are closely tied to their German counterparts over both the short and long runs, Swiss and German interest rates, at least over the short run, frequently do not move in a parallel manner. The Swiss National Bank’s own research (for example, Hermann and Rich, 1988) suggests that movements in Swiss short-term interest rates are largely explained by domestic disturbances such as changes in the money supply, output, and prices. Foreign interest rates, though statistically significant, exert only a minor influence on their domestic counterparts. Thus, floating exchange rates do drive a wedge between domestic and foreign money markets. This result stands in sharp contrast to Genberg’s conclusions. The difference between Genberg’s results and our own, I believe, is largely explained by this choice of sample period. His research is based on the period 1973–81. This period includes 1978–79, during which the Swiss National Bank followed an exchange-rate, rather than a money-supply, rule. If Genberg had also examined the period 1982–87, during which the Swiss National Bank paid little attention to the exchange rate, he would likely have obtained different results. Unfortunately, Genberg’s paper, though written in 1989, covers only the Austrian and Swiss policy experience up to the mid-1980s. I fail to understand why this policy survey is not completely up to date.

Fifth, a study of the second half of the 1980s would have been worthwhile in another respect. Since 1984, output movements in Switzerland have been somewhat out of phase with those in Germany, while Austrian and German output have been closely correlated. The output patterns recorded in the second half of the 1980s raise two questions. First, to what extent do they reflect the differing exchange rate regimes prevailing in Austria and Switzerland? Second, to what extent does monetary policy account for these patterns? In my view, a study of these questions would be worthwhile. I would not be surprised if—after studying the policy experience of the second half of the 1980s—Genberg were compelled to modify some of his conclusions.

To sum up, although Austrian and Swiss policy experiences have been similar, Genberg understates the room for maneuver afforded to the Swiss National Bank by a floating exchange rate. Even so, I admit that the case for a floating Swiss franc may become less compelling as time wears on. Should the major member countries of the European Communities manage to maintain price stability, Swiss monetary authorities would have to consider seriously the alternative of fixing the exchange rate. However, for the time being, Switzerland intends to stick to its traditional policy approach centered on a money-supply rule.

Appendix
Appendix Determinants of the Real Effective Exchange Rate and the Current Account in Austria
Dependent Variable
Current account/GDPEffective exchange rate
Constant.73.78
(.81)(5.61)
Industrial production−35.59−.65
(deviation from
trend)(−2.65)(−2.44)
Budget surplus/GDP.45.03
(2.04)(8.11)
Growth of M1−18.67
(−2.03)
Effective exchange.34
rate for Germany(2.43)
Lagged current.42
account(1.67)
R2-bar.33
D-W2.841.00
Note: Annual data for 1971–84 were used to estimate equations for the real effective exchange rate and the current account.
Note: Annual data for 1971–84 were used to estimate equations for the real effective exchange rate and the current account.

According to these results, a cyclical expansion in Austria (measured by a positive deviation of industrial production from trend) would lead to a current account deficit and a real appreciation consistent with standard theory if cyclical movements in industrial production were demand-driven.37 An increase in the growth rate of money would also lead to a current account deficit, whereas the effect on the real effective exchange rate was not significantly different from zero. (The point estimate indicated a depreciation.)

Turning to the effect of fiscal policy, 1 notice again that the predictions of standard theoretical models are borne out. A budget surplus of 1 percent of GDP would—according to the estimates-lead to a current account surplus of 0.45 percent of GDP and a real appreciation of about 3 percent. With a fixed nominal exchange rate, the real appreciation would result from a higher inflation rate in Austria than in Germany.

Although I would not claim that these results represent the last word on the determination of the Austrian current account and real exchange rate, they do provide a theoretically defensible explanation of the evolution of these variables. This explanation emphasizes the importance of fiscal policy and suggests that the hard-currency policy was not necessarily the reason for the current account deficit in Austria and the real appreciation of the schilling in the second half of the 1970s.

References

    ArtusJacques“Commentary,” in SvenArndted.,The Political Economy of Austria (Washington: American Enterprise Institute tor Public Policy Research1985) pp. 3541.

    Austrian National BankAnnual Report for 1985 (Vienna: Austrian National Bank1986).

    BéguelinJean-Pierre“Financial Innovation and Monetary Policy: The Swiss Non-Case,”Financial Innovation and Monetary Policy (Basel: Bank for International Settlements1984).

    CamenUlrichHansGenberg andMichaelSalemi“Using Inverse Control Methodology to Reconci lé Actual and Optimal Monetary Policy in Switzerland” (mimeographedGeneva: The Graduate Institute of International Studies1988).

    CosandierPierre-AlexisSwitzerland and International Financial Integration : Recent History and New Theoretical DevelopmentsPh.D. dissertation (Geneva: The Graduate Institute of international Studies1987).

    DooleyMichael andPeterIsard“Capital Controls, Political Risk, and Deviations from Interest-Rate Parity,”Journal of Political EconomyVol. 88 (Chicago, Illinois: 1980) pp. 37084.

    EglinMichaela“Offset Coefficients as an Indicator of the Effectiveness of Domestic Monetary Policies” (mimeographedGeneva: The Graduate Institute of International Studies1989).

    FrankelJeffrey A. andA.T.MacArthur“Political vs. Currency Premia in International Real Interest Differentials: A Study of Forward Rates for 24 Countries,”European Economic ReviewVol. 32 (Amsterdam: 1988) pp. 10831121.

    GenbergHansCauses and Consequences of Relative Price Variability: Evidence from Switzerland ami Sweden (Bern: Fonds National Suisse de la recherche scientifique1986).

    GenbergHans“Exchange Rate Management and Macroeconomic Policy: A National Perspective,”The Scandinavian Journal of EconomicsVol. 91 (Stockholm: 1988) pp. 43969.

    GenbergHansMichaelSalemi andAlexanderSwoboda“Foreign and Domestic Disturbances as Sources of Aggregate Economic Fluctuations: Switzerland 1964–1981,”Journal of Monetary EconomicsVol. 19No. 1 (Amsterdam: North-Holland1987) pp. 4567.

    GenbergHans andMichaelSalemi“The Effects of Foreign Shocks on Prices of Swiss Goods and Credit: An Analysis Based on VAR Methods,”Annales d’Economie et de StatistiqueNo. 6/7 (Paris: 1987) pp. 10124.

    LanguetinPierre“The Exchange Rate Relationship between the Deutschmark and the Swiss Franc,”Journal of Foreign Exchange and International FinanceVol. 1No. 2 (1988) pp. 16468.

    MussaMichael“Nominal Exchange Rate Regimes and the Behavior of Real Exchange Rates, Evidence and Implications,” in KarlBrunner andAllanMeltzereds.,Carnegie-Rochester Series on Public PolicyVol. 25 (Amsterdam: North-Holland1987) pp. 117214.

    Organization for Economic Cooperation and Development OECDEconomic Surveys: Austria (Paris: OECD1986).

    PooleWilliam“Optimal Choice of Monetary Policy Instruments in a Simple Stochastic Macro Model,”Quarterly Journal of EconomicsVol. 84 (Cambridge, Massachusetts: 1970) pp. 197216.

    RichGeorg“Exchange Rate Management Under Floating Exchange Rates: A Sceptical Swiss View” (mimeographedZürieh: Swiss National BankSeptember1989).

    RichGeorg andJean-PierreBéguelin“Swiss Monetary Policy in the 1970s and 1980s, an Experiment in Pragmatic Monetarism,” in KarlBrunneret al., eds.,Monetary Policy ami Monetary Regimes. A Symposium dedicated to Roy E. Weintraub Center Symposium Series Np. CS-17Center for Research in Government Policy and Business, University of Rochester(1985).

    HermannWerner andGeorgRich“Zinssätze and Wechselkurse in der Schweiz,”Wechselkursstabilisierung EWS und Weltwährungssystemed, byOtmarIssing (Hamburg: Verlag Weltarchiv1988).

While Austria did so as a member of the International Monetary Fund, Switzerland pursued the fixed rate policy on its own accord as a nonmember.

Neither country followed Germany when it revalued the deutsche mark relative to the dollar in 1961 and in 1969.

See, for example, Rich and Hégueiin (1985) and Camen, Genberg, and Salemi (1988) concerning the objectives of the Swiss National Hank and the Annual Reports of the Austrian National Bank, concerning Austria.

For a description of the details of the chosen strategies, see the discussion under the heading, “Macroeconomic Policies and Their Consequences.”

For a more comprehensive analysis, including a detailed list of references, see Genberg (1988).

According to the Austrian National Bank’s 1985 Annual Report, one reason for adopting their hard-currency policy was to stabilize expectations in order to make “…economic policy as a whole … less subject to pressures to take discretionary measures…” (Austrian National Hank. 1986, p 13).

This similarity in inflation rates was also apparent during the 1960s. The surge in prices in Austria during 1984 can be explained in part by the increase in the value-added tax of 2 percent in October 1983.

The scales in the figure have been designed so that a movement of a given size represents the same percent change for both exchange rates. This choice of scales was made in order to illustrate the much larger fluctuations in the Swiss franc/deutsche mark rate than in the schilling/deutsche mark rate.

Concern about the appreciation of the franc in 1974 did lead to the adoption of measures designed to discourage capital inflows and to sterilize intervention purchases of dollars in the foreign exchange market. However, it was soon realized that both types of measures were largely ineffective, and they were therefore phased out. For a comprehensive description of these and other measures taken in Switzerland to influence capital movements, see Cosandier (1987).

See Languetin (1988) for an account of this episode as well as for an analysis of the relationship between the Swiss franc and the deutsche mark more generally.

The depreciation of the franc in the spring of 1989 elicited similar remarks from Hank officials.

Rich (1989) provides an informative analysis of monetary policy in Switzerland, stressing the difficulty of combining monetary targeting with a concern about exchange rate movements.

The target rates are included in the chart as single observations tor the last quarter of the year. The actual growth rate measured over the corresponding period is recorded at the same quarter.

In the chart (top panel), the target rates for the monetary base have been included since 1980, whereas the actual growth rate continues to refer to M1. While there is some difference between the actual growth rates of the monetary base and Ml, the conclusion that year-on-year targets have frequently been missed still remains valid.

The target path in the chart has been constructed by using a fictitious value of 5 percent for 1979. Also, contrary to how the announced targets are defined, the base period for each year is always assumed to lie the target value for the previous year, implying the absence of “base drift.”

The introduction of an electronic interbank payments system in 1987, and changes in banks’ liquidity requirements in 1988, may require a modification of this assessment. I owe this point to Georg Rich.

Developments in 1989–90 indicate that this statement also has to be reassessed.

The sharp reduction in Ml growth in 1979 was due to a massive switch from demand deposits (included in Ml) into time deposits (included in M2) in the first half of 1979, This change in the composition of deposits was presumably a consequence of a change in the banking law and a new agreement of commercial banks on deposit interest rates introduced in March 1979.

Similar conclusions have been reached by Frankel and MacArthur (1988).

The average absolute value of the deutsche mark/schilling interest rate differential was 2.61 percent in 1974–79 and declined to 0.94 percent in 1980–85, based on the data underlying Chart 4 (lower panel).

See Eglin (1989), who reports an offset coefficient of -0.79 for 1980–88.

As we shall sec later on fiscal policy was highly expansionary in Austria in 1975 and 1976, This led, as theory’ would predict, to a substantial current account deficit in 1976 and 1977. The standard Mundell-Fleming model for a fixed exchange rate economy with perfect capital mobility would also predict that a capital inflow would have occurred thereby financing the budget deficit and preventing the rise in domestic interest rates. Although some capital did flow in (showing up mainly as a sizable increase in the errors and omissions component of the statistics), it was not enough to prevent an overall balance of payments deficit The implication is that capital controls were to some extent binding during this period.

Eglin, op. cit., reports an estimate of -0.48 for 1977–80.

German industrial production is used to represent external impulses in general. The underlying causes of these impulses may be German in origin, but could also come from Europe-wide, or global, shocks.

Formal evidence supporting this statement can he found in the study by Genberg, Salerai, and Swoboda (1987). They concluded that external influences were significantly more important than internal factors in accounting for fluctuations in Swiss industrial production. For the flexible exchange rate period spanning 197,1 to 1981, they estimated that about 65 percent of the 48-month forecast error variance of industrial production could he accounted for by foreign shocks, whereas only 25 percent could be ascribed to domestic shocks. The close relationship between Austrian and German industrial production shown in Chart 5 suggests that similar results would hold for Austria.

See, for example, Artus (1985).

See annual OECD surveys of Austria.

Empirical results presented in the appendix support this interpretation.

Indeed, the current account improved substantially during 1975–78. presumably because of cyclical factors.

It is relevant to note that (le p berg (t 1986) did not find any relationship between the dispersion of relative prices and growth of industrial production.

Ail illustration of the increased credibility of the Austrian peg to the deutsche mark can be based on the effects of fluctuations in the deutsche mark/U.S. dollar rate 011 the schilling/deutsche mark rate, if there is some doubt about the peg, then a depreciation of the dollar relative to the deutsche mark would be expected to lead to a depreciation also of the schilling relative to the mark. When the peg is perfectly credible, on the other hand, there is no reason to expect such a relationship since expectations of a parity change would he absent. Regression results for 1973–78 indicate that there was indeed some doubt about the resolve of the hard-currency policy in its early phase since a change in the deutsche mark/dollar rate bad a significant effect on the schilling/deutsche mark rate in this period. This effect disappeared in the 1979–89 period, indicating an increase in the credibility of the peg. The credibility of the Swiss policy may be deduced from the Swiss authorities’ adherence to their monetary target over the medium term, as seen in the discussion of monetary policy.

A further similarity between the countries was introduced by the temporary exchange rate target adopted by the Swiss National Bank in 1978–79, when the notion was introduced that an “appropriate” value of the franc is between 0.80 and 0.95 deutsche mark. The National Bank may have abandoned this view, but the foreign exchange markets seem to have incorporated it into their pricing. Evidence again comes from the effect of a change in the deutsche mark/dollar rate on the Swiss franc/dollar rate. Before the 1978–79 episode this effect was statistically significant, whereas it has not been in the 1980s.

When the difference in cyclical position relative to Germany was introduced in the current account equation, its coefficient was not significantly different from zero.

    Other Resources Citing This Publication