The current development in Europe appears to be extremely contradictory: While the countries in Eastern Europe finally see their chance to free themselves from centralistic controls, mainly the EC-countries march in the opposite direction of increasing the power of central institutions. The “No” of the Danes in the summer of 1992 against the Maastricht Treaty is a rejection of yielding additional decision competency to “Eurocrats” in Brussels, and of centralistic tendencies in the EC. The outcome of the Danish referendum disturbed quite a few policymakers and bureaucrats on the national as well as on the European level. However, it also triggered a long overdue discussion in all of the EC-countries of the question who should be in charge of economic policy.
The critical debate of this question obviously surprised policymakers in Europe. Economists, though, apparently found an adequate answer to this question in the theory of fiscal federalism (Musgrave, 1961; Oates, 1991). According to this theory, allocative, distributional, and macroeconomic stabilization considerations can be reasons to cede competency for economic policy to a central institution (Walsh, 1992). This is the case if the market as coordination mechanism does not lead to satisfactory results in one of these three areas, and if there is a realistic chance that a centralized government institution can reach better results than decentralized government institutions.
Although these considerations hold for all political unions, they nonetheless can be helpful in a discussion of the subsidiary principle in the EC. It remains true that the EC is still far away from a political union, despite or because of vague declarations of intent in the Maastricht Treaty. However, the question of who is to get which competency in economic policy decisions in a European Monetary and Economic Union can no longer be avoided. Allocative and distributional aspects will not be of central concern In the paper (Berthold, 1993), but rather the question whether centralized activities for macroeconomic stabilization are a necessary condition for a successful European economic and monetary integration. The theory of “fiscal federalism” shows conditions under centralized stabilization policies could be useful.
Before addressing this question it must be clarified which objectives should be pursued by stabilization policies. Firstly, such policies should keep the price level stable. There seems to be a consensus in Europe that a European Central Bank which is independent from political pressure is most conducive to this objective. However, whether excluding currency competition in Europe is a successful strategy is yet to be seen. Secondly, stabilization policies should smooth real national income over the business cycle. Therefore, the opinion is widely held that stabilization policies are responsible for keeping employment at a high level. In the following, I will try to give an answer to the very controversial question whose task it is in a European economic and monetary integration to implement such stabilization policies.
In a first step, I will ask the question under which circumstances stabilization policies are appropriate in a European economic and monetary integration committed to maintaining market principles. If the answer is, that under certain conditions macroeconomic activities are appropriate, then in a second step it must be considered which activities should be undertaken. Finally in a third step, the best institutional level for undertaking such activities must be found. Hence, the optimal degree of subsidiary is also a question in the context of stabilization policies. It must first be decided which stabilization tasks should better be fulfilled by the government and which ones should better be left to the market. Then, the appropriate level for pursuing such stabilization policies in the EC must be found - either on a centralized or on a decentralized level, either on the country or on the regional level.
II. Why can there be economic misalignments in a European Economic and Monetary Union?
1. What are the causes for macroeconomic instability of an economy?
Macroeconomic instabilities can be expected in economies when they are hit by shocks, for which they are not fully prepared. All economies are in general constantly hit by shocks: Not only because tastes of economic agents, production technologies and the quantity and quality of production factors permanently change (Eucken, 1990), also because policymakers themselves repeatedly cause such shocks through their monetary and fiscal activities. Adjustment problems are not only caused by the size of these policy and non-policy induced shocks, but also by their quality (i.e., temporary or permanent, symmetric or asymmetric, real or monetary shocks). Temporary and small shocks, or shocks that are stretched out over a longer period of time are easier to absorb than large sudden shocks.
Although these shocks cause imbalances in goods and factor markets, even adverse shocks do not destabilize the overall economy if the private economic agents are willing to bear the burden of adjustment. This is the case if relative prices are flexible and if production factors are mobile. The market mechanism can then satisfactorily fulfill its stabilization task. In reality, though, adjustment capacities of economies are much lower. Relative prices are not flexible enough and production factors are not mobile enough to efficiently remove such imbalances. This causes not only an increase in the mismatch-, but also in the aggregate unemployment rate.
Aggregate unemployment can either be due to Keynesian or to classical reasons, depending on which goods and factor markets are more incomplete (Berthold, 1987; 1988). Incomplete capital and goods markets lead to Keynesian unemployment (Berthold, 1987; Greenwald/Stiglitz, 1988; 1990). If after adverse shocks real interest rates and goods prices relative to nominal wages are not flexible enough downwards, or if aggregate demand does not respond enough to falling interest rates and prices, then the resulting lack of aggregate demand can cause Keynesian unemployment. The reasons for these market failures are not only perceived to be information asymmetries on financial markets and incomplete competition on goods markets, but also government regulations of capital and goods markets.
Classical unemployment can be expected if mainly labor markets are incomplete. If agents on labor markets are not willing to forgo part of their real income, then unemployment is inevitable. It does not matter in principle who gives up real income - employees or employers - as long as somebody does it. Since employers will only invest their capital if they receive an adequate rate of return and cannot be forced to so, employees will apparently have to bite the bullet of accepting a lower real wage. At least temporarily, many employees manage to avoid this in reality. This is due to the market power of “insiders” in wage negotiations, which stems from labor turnover costs. These costs of hiring, screening and firing employees are partly inevitable and to the larger part caused by regulatory government interventions, which reinforce the position of insiders to the disadvantage of unemployed outsiders (Lindbeck/Snower, 1988; Berthold, 1992a).
Thus, incomplete goods and factor markets prevent a quick reduction of aggregate unemployment - be it Keynesian or classical - after adverse shocks. However, prolonged unemployment causes additional factors to contribute to the persistence of unemployment. Part of the real and human capital stock becomes obsolete; therefore even if relative prices react with some delay, the original level of employment is no longer attained (Franz, 1987; Cross, 1988). This gives a new quality to aggregate unemployment: If imbalances after adverse shocks cannot be reduced quickly, it will be very difficult to do so later on. The path- dependency of unemployment therefore reinforces the importance of flexible relative prices. Hence, the degree of macroeconomic stability of economies depends mainly on the completeness of its goods and factor markets.
2. How does a Monetary Union affect macroeconomic instabilities?
The macroeconomic stability of an economy depends on the size and severity of shocks it is confronted with and on its ability to efficiently absorb such shocks. Hence, the effect of a European economic and monetary integration on the adjustment burdens and on the adjustment capacities of the participating countries has repercussions on the probability of economic misalignments. An economic and monetary union only makes economic sense if it contributes to the reduction of policy-induced shocks and if it enhances the ability of the participating economies to absorb shocks.
On the one hand, there exists widespread consensus that an economic union contributes positively in this respect. The reduction of regulations on goods and factor markets not only intensifies competition, and makes relative prices more flexible and production factors more mobile, it also restricts discretionary activities of policymakers and thus reduces the probability of policy-induced shocks. On the other hand, there exists a controversy on the effect of a currency union on these factors. According to widespread consensus, such a monetary arrangement does not destabilize the participating economies only if it is optimal in the sense of the theory of the optimum currency area (Ingram, 1959; Mundell, 1961; Kenen, 1969). Because of the difficulties of applying this theory, a monetary union can already be called an optimum currency area if it does not force the participating countries to make more use of their stabilization instruments because of larger adjustments burdens and lower adjustment capacities. This only seems to be the case if those economies form a currency union whose political and economic markets work similarly well (Berthold, 1992b).
If only those economies form a monetary union in Europe whose political markets work similarly well, then chances are good that adjustments burdens will not increase further and that the already low adjustment capacity will not be further reduced. Political markets work the better the more policymakers are forced to offer these goods which fit the true preferences of voters. Efficient political markets reduce incentives for governments and bureaucrats to pursue economic activities with little efficiency. However, to keep adjustment burdens under control it is not sufficient to restrict a monetary union to countries with similarly well working political markets. In addition, it is also necessary that member countries have similar preferences for collective goods, especially for the collective goods “security” and “equity” (Courchene, 1992; Berthold, 1993).
If the political markets in a monetary union do not work worse than before, then the political pressure on the European Central Bank to implement inflationary policies will not be higher than the average political pressure to implement such inflationary policies before the monetary union. Thus, neither will monetary disturbances become more frequent, nor will the adjustment capacity be reduced because of allocative distortions due to inflation. Relative prices will then also not be stickier and production factors will not be less mobile, because rent-seeking-activities of interest-groups will not increase because of the monetary arrangement. In addition, regulatory government interferences in goods and factor markets will also not increase due to the monetary arrangement. Hence, the adjustment capacity of the economy is not necessarily reduced.
The danger of additional burdens of adjustment and of a lower adjustment capacity is further reduced in a monetary union if it is restricted to countries with similarly well working economic markets which are hit by symmetric shocks. The burdens of adjustment do not necessarily increase significantly if countries with a diversified production and export structure form a monetary union (Kenen, 1969). The probability is high in this case that sector specific shocks will grow into country specific shocks because different shocks will offset each other. However, there will still be a burden of adjustment. Structural change will be necessary and therefore relative prices must react to intersectoral imbalances and production factors must be intersectorally mobile.
Apart from that, especially those countries in Europe are candidates for a monetary union which are mainly hit by symmetric and not asymmetric shocks. In this case, symmetric shocks are easier to absorb with a unified monetary and fiscal policy or exchange rate policy vis-à-vis third countries than country-specific asymmetric shocks. Considering both criteria - degree of diversification and symmetry of shocks - then especially countries which are connected by large intra-industry trade are suited for a monetary union.
It is therefore no surprise that recent empirical investigations reach the result that the European countries are hit by quite different supply and demand shocks (Bayoumi/Eichengreen, 1992). The supply shocks in the core countries of the EC - Germany, France and the Benelux countries - are smaller and more closely correlated than in the peripheral countries - Great Britain, Italy, Spain, Portugal, Ireland and Greece. A similar conclusion holds for demand shocks although the differences between the country groups are not as pronounced. These empirical findings reinforce the notion of a “two speed” Europe.
The critics of the optimum currency area theory object that the common market of 1992 lets the EC-countries become more similar. They argue that the economic union in Europe lets intra-industry trade increasingly dominate inter-industry trade. Hence, future shocks in a monetary union should be symmetric rather than asymmetric (Cohen/Wyplosz, 1989). However, in an area as large and culturally and geographically diversified as the EC it is very doubtful that there will be no significant asymmetric shocks in the future (Goodhart/Smith, 1992). One apparent reason is that even in Europe with completely free goods and factor markets there are still quite a few goods and services left which are non-tradables (Bean, 1992). In this case, even symmetric shocks will be country-specific.
The danger of unstable macroeconomic developments could be further reduced if only countries with a similar adjustment capacity agreed to form a monetary union. Relative prices will only be sufficiently flexible and production factors sufficiently mobile if there is a high degree of competition on goods and factor markets. Therefore, a monetary union should be restricted to the countries which agree on the immensely important question of competition policy. However, the discussion about industrial policy shows that the core countries in the EC do not agree on this issue. Only the Benelux countries and to a lesser extent Germany seem to reject the idea of industrial policy. But if there is not a clear consensus that competition is desirable, then a monetary union does not make sense. In such a case the already insufficient adjustment capacity, i.e., inflexible real wages (Layard/Nickel/Jackman, 1991) and immobile production factors (Ermish, 1991; Dooley/Frankel/Mathieson, 1987), is not increased in the EC. In contrast, in the USA relative prices on labor markets are more flexible and the production factor labor is more mobile between regions, so that the USA can adjust faster to shocks than the EC. This seems to be necessary to have a monetary union work without major government interference.
It is sometimes objected that a high adjustment capacity is not necessary to limit the size of unstable macroeconomic developments in a monetary union. In a monetary union, the agents on labor markets would be forced to react much more in wage settlements to imbalances, because the anti-inflationary position of a European Central Bank would be more credible than the one of national central banks. Hence, inflationary expectations would be reduced (Horn/Persson, 1988) and the possibility to externalize the negative employment consequences of excessive wage increases via inflation would be reduced. In addition, fiscal policy in the EC would also be more credible with a credible consensus on a “no bail-out” clause. This would contribute to a more moderate wage behavior by employees in regions with economic difficulties (EC-Commission, 1990), because the costs of excessive wage increases in these regions can no longer be externalized on the government. Finally, integrated goods markets reduce the market power of firms, making labor demand more elastic and forcing wage policy to consider to a greater extent market conditions (Marsden, 1989).
In the discussion of the monetary union all these three arguments have been questioned. Even a European Central Bank that is largely independent from political pressure still has to acquire credibility concerning its anti-inflationary position - something the Deutsche Bundesbank has already achieved. It has also been pointed out that fiscal convergence criteria like the ones in the Maastricht Treaty do not reduce the danger of a fiscal bail-out, and that labor demand only becomes more elastic if the EC does not degenerate into a fortress Europe. It is furthermore an important caveat that the less wealthy EC-countries demand a stronger social cohesion to limit the differentiation and flexibility of wages. However, this would probably reduce the pressure on economic agents to adjust with flexible relative prices and with factor mobility because such a welfare state approach would lead to a quasi full employment guarantee, a high degree of state employment, and a socially financed unemployment insurance. At the same time, removing the veil of different national currencies exposes the wage differences across countries. As in the case of the reunification of Germany, this would increase the demand for equal wages for equal work if the trade unions are successful in organizing wage bargaining on the EC-level instead of the national level.
The question if economies become more unstable in a monetary union is therefore easily answered: Only if it is formed by countries with similarly well functioning political and economic markets, will the danger of increased adjustment burdens and reduced adjustment capacity be reduced. In reality however, the fear is justified that such a multi-speed Europe concerning monetary policy is not enforceable in the political process. Hence, economic misalignments in Europe will not be lower, but rather larger in the future. This development would be enhanced if the road towards a political union in Europe is pursued further as planned (Berthold, 1993). In such a case, welfare activities would increase, the achieved progress towards economic union would again be questioned, the politically induced adjustment burden would increase, the flexibility of markets would be further reduced by government regulations, and the sensitivity towards real shocks would be increased.
III. What should be the approach towards economic misalignments in a monetary union?
The process of ratifying the Maastricht Treaty has demonstrated so far that economic and political rationality generally deviate. One must therefore be concerned that the monetary union in Europe will not be “optimal”. As the monetary union will most likely be accompanied by an expansion of European welfare policies, relative prices are bound to be less flexible and production factors are bound to be less mobile on the national level. Thus, shocks are more likely to lead to economic misalignments. However, since in a monetary union neither monetary nor exchange rate policy are available to cope with country specific shocks, it is often argued that the central policymaker should have a strong fiscal policy instrument. This would be the only way to stabilize the economic development. According to this opinion, a monetary union is only successful if countries which are hit by adverse shocks automatically receive financial transfers (De Grauwe, 1992) to stabilize in a typical Keynesian way the economic development.
1. Is fiscal policy helpful in coping with temporary shocks?
Unstable macroeconomic developments are also in a monetary union the result of an excessive adjustment burden in connection with an insufficient adjustment capacity. Hence, a traditional Keynesian demand policy is only successful under very special circumstances, namely if adverse shocks are not permanent, but rather temporary. With such shocks there is no need for a real adjustment of economic agents. It is sufficient to finance the imbalances until the adverse shock is reversed.
The owners of physical capital can insure themselves against such adverse shocks by diversifying their portfolio interregionally and internationally. However, the owners of human capital cannot do this because of a lack of bilocation and because adverse selection and moral hazard problems make efficient contracts between individuals of different regions infeasible (Eichengreen, 1990). These individuals can still diversify the risk of region-specific shocks by buying assets that are not completely correlated with the yield of human and physical capital of their region. In reality though, this way of diversifying risk does probably not work because most employees have apart from their illiquid and undiversifiable human capital only comparatively few financial assets, which are often invested in undiversifiable, region-specific real estate.
This does not necessarily imply that employees cannot insure themselves against temporary shocks. If the shocks are of limited size, and if capital markets are complete, and if individuals live long enough, then individual income can be smoothed over the business cycle by using the capital market to intertemporally reallocate individual income. However, these conditions are in reality often not fulfilled, so that most employees are in general unable to finance imbalances after temporary shocks. This seems to make macroeconomic misalignments inevitable.
Compensatory fiscal policy - discretionary or bound by rules - is only welfare enhancing if the policymaker has better and cheaper access to capital markets than employees. Capital markets are integrated by the “Common Market 1992” in the EC, making it easier for private economic agents to insure themselves against temporary swings of their real income. Nonetheless, an efficient insurance for private individuals against the negative real effects of adverse shocks will probably not be possible soon. This strengthens the importance of fiscal policy in a monetary union. This does not necessarily mean the widely rejected discretionary fiscal policy, but could rather also consist of automatic stabilizers like a socially financed unemployment insurance.
This positive judgment concerning fiscal policy must possibly be revised considering the difficulty of distinguishing temporary from permanent shocks, and that the planning horizon of economic agents has lengthened, and that a successful implementation of fiscal policy is difficult in a democratic environment.
If government institutions are not successful in distinguishing temporary from permanent adverse shocks, then with a permanent shock automatic stabilizers are welfare reducing. It is counterproductive to finance such permanent shocks. It is rather necessary that private agents adjust to the new economic situation with flexible relative prices and with regionally and sectorally mobile production factors. The adjustment capacity is generally lowered by automatic stabilizers because it enables private agents to avoid at least temporarily the necessary adjustment process. This results in less flexible relative prices and less mobile production factors.
It is possible that even with temporary shocks credit financed fiscal policy is not successful. This is always the case if private agents are ultrarational (Barro, 1974). If tax payers realize that current public deficits must be financed by future tax surpluses, they anticipate the higher taxes of tomorrow and already reduce their expenditure today, except if they do not care at all about the welfare of subsequent generations. In such a case, higher public demand would be offset by lower private demand leaving aggregate demand unchanged. Neglecting the general critique of the neoricardian equivalence theorem, it is difficult to conceive a world in which relative prices are inflexible like in a European Monetary Union presumably because agents have only imperfect foresight, and in which this theorem still holds.
Even if the use of automatic stabilizers is not always doomed to fail in the presence of temporary adverse shock, its possibilities to succeed in a democracy are still very limited. Although public deficits can increase aggregate demand, the peculiarities of the democratic process can make it impossible to use fiscal policy for short term stabilization purposes (Berthold/Külp, 1989; Persson, 1992). A series of well known time-lags (Friedman, 1953) contribute to prevent policymakers from reacting immediately to shocks, and to prevent fiscal policy from having an immediate effect upon aggregate demand. Automatic stabilizers can reduce the inside lags, but not the outside lags. Therefore, there is always the danger that fiscal policy is procyclical instead of countercyclical.
2. Does fiscal policy prevent hysteretic developments from occurring?
Even though all these arguments shed considerable doubt on the usefulness of fiscal policy in the presence of temporary shocks, it is often argued that such a policy is helpful in preventing hysteretic developments. If economies only react very sluggishly to permanent shocks, there is the danger that the increase in unemployment - be it Keynesian or classical -becomes persistent or even hysteretic. Thus, expansionary fiscal policy might be useful to avoid such a path dependency of unemployment, and it might help in the transition towards a new long run equilibrium.
This is also the reason why giving up the exchange rate instrument in a monetary union is a considerable loss (Berthold, 1990). It is clear that an adjustment of the exchange rate does not free private agents from the need to adjust to the new real economic situation. In the short run, though, nominal exchange rate changes can help to avoid the path dependency of unemployment. This is firstly the case if nominal prices are not sufficiently flexible and if the central bank has an information advantage and is thus quicker in reacting to such shocks. It is secondly the case, if not only the real wage, but also the relative position in the wage hierarchy matters to employees (Keynes, 1936; Gylfason/Lindbeck, 1984). Hence, at least in the short run flexible exchange rates increase the adjustment capacity of an economy and contribute to prevent the economy from ending up in an unfavorable, new equilibrium with lower real income and higher unemployment.
In contrast to the adjustment of exchange rates, traditional fiscal policy gives private economic agents an incentive to constantly delay the unavoidable real adjustment. If the government reacts with discretionary or rule-bound fiscal policy, then the labor demand curve becomes less elastic (Calmfors/Horn, 1985; Begg, 1991; Calmfors, 1992). This reduces the willingness of employees - insiders - to make wage concessions, and thus reduces the chance of the unemployed outsiders to find a new job. Insiders are obviously successful in appropriating a large part of the macroeconomic benefits of an expansionary fiscal policy, and thus in avoiding a real adjustment by themselves.
The negative secondary effects of fiscal policy would be smaller, if it contributed to increasing the adjustment capacity of an economy. The traditional Keynesian policy of stabilizing nominal income is inefficient because it makes markets less flexible. Fiscal policy in the presence of permanent shocks only makes sense if it helps to make relative prices more flexible (Agell/Dillén, 1991) and production factors more mobile. It is therefore necessary to think about new fiscal policy instruments to increase the adjustment capacity, e.g., “tax based income policy” proposals, which were suggested to avoid inflationary developments.
However, in general fiscal policy in such situations is not only of the traditional Keynesian type, but rather the government often degenerates in situations of lasting imbalances on the labor market into the employer of last resort. The increased employment in the public sector does not really reduce unemployment though, but rather only hides it and contributes to a slower real adjustment of the economy. This necessarily leads to a lower growth rate of the economy and eventually to a higher rate of unemployment (Lindbeck, 1990; Calmfors, 1992). There is in general increasing pressure in such an unsatisfying situation on the central bank to loosen monetary policy. This pressure would be higher if fiscal policy were centralized in the EC because this would increase the importance of real relative to monetary macroeconomic activities (Buiter, 1992).
After all, stabilization policy of the Keynesian type is also in a monetary union only second-best. On the one hand, it reduces the costs of persistent unemployment after negative shocks. On the other hand, it also reduces the adjustment capacity of the economy because it contributes to making relative prices less flexible and production factors less mobile. A first-best therapy would directly reduce the incompleteness of goods and factor markets. Hence, there is less need for stabilization policy and more need for “Ordnungspolitik” to improve economic markets. The incompleteness of economic markets in Europe would surely be reduced by setting the appropriate rules in the Common Market Project of 1992, i.e., implementing a high degree of competition on all markets and avoiding all industrial policy activities. However, the fear is justified that increasing political integration contributes to setting up new implicit barriers (Purvis, 1992). The planned monetary union will probably destabilize the macroeconomic development in Europe from the start because it will be suboptimal for political reasons, leading to an expansion of welfare state activities with its negative effects on the mobility of production factors and on the flexibility of relative prices. This increases the necessity for stabilization policy because the pressure on policymakers will rise to offset negative shocks. If fiscal policy is indeed unavoidable, the question on which institutional level it is best located must be answered.
IV. Which is the best institutional level for stabilization policy in a monetary union?
The discussion if fiscal policy in a monetary union should be centralized or decentralized is not new. The defenders of a fiscal policy on the EC-level argue that external effects create the need for a central budget (Meade, 1957; Scitovsky, 1957; Lundberg, 1972; McDougall-Report, 1977) to cope adequately with unstable macroeconomic developments. The opponents of such a central solution have for quite some time been of the opinion that national economies are by themselves able to master stabilization problems by resorting to the national and international capital markets. A new argument has been added lately: Unstable macroeconomic developments could be prevented by offering an inter-regional government insurance (Sala-i-Martin/Sachs, 1992).
It is often argued that national policymakers do not use sufficiently expansionary fiscal policies in the presence of adverse symmetric shocks. They have an incentive to behave like a free rider on the stabilization issue. While the costs of an expansionary fiscal policy - higher real interest rates and higher taxes in the future - are mainly borne by the country that applies the policy instrument, the benefits accrue also to the other countries in open economies like in the EC. Hence, there is the danger that the fiscal policy instrument is not used sufficiently. It therefore seems to make sense to demand a centralization or at least a coordination of fiscal policy on the union level in the EC (Giovannini/Spaventa, 1990; Wyplosz, 1991; Masson/Mélitz, 1991).
This view is not convincing for different reasons: The lengthy and intense discussion of the Mundell-Flemming-model yielded the result that an expansionary fiscal policy can have positive as well as negative effects on other countries. The form of macroeconomic external effects depends in reality in a monetary union on the relative size of a country, of the tradable sector, and more importantly on the adjustment capacity -flexibility of relative prices and mobility of production factors - of the participating economies (Boughton/Haas/Masson, 1989). It is therefore not a surprise that it has repeatedly been found that expansionary fiscal policies in a monetary union had negative effects on other member countries (Roubini, 1989; Masson/Meredith, 1990). The reason is simple: Higher interest rates have negative effects on other countries, and the temporary appreciation of the common currency vis-à-vis third countries reduces the net exports of the other member countries (an example for this is the negative effect of German unification upon aggregate demand in other EC-countries). Several simulation studies have found that the effects of an expansionary fiscal policy are largely confined to the country that applies the policy (Masson/Taylor, 1992). The external effects of fiscal policy seem to be not only small, but also irrelevant for economic policy. Since these external effects are not technological but rather pecuniary (Buiter, 1992), there is no need for fiscal activities.
The defenders of an active fiscal policy argue that incomplete capital markets and incomplete or even the absence of private insurance create the need for an expansionary fiscal policy after an adverse temporary and permanent shock to stabilize real income. In addition, they demand a centralization of the stabilization function not only for symmetric, but also for asymmetric shocks. They are for different reasons of the opinion that national governments will not be successful in smoothing real income by using countercyclical debt policy.
It is generally possible that the neo-ricardian equivalence theorem holds on the regional (national) but not on the EC-level (Sala-i-Martin/Sachs, 1992). With a fiscal arrangement on the EC-level, the tax and transfer payments are distributed between the regions (countries) in such a way that the ones which are mostly hit by region-specific shocks are most favored financially. The lower tax payments and the higher transfer payments that regions (countries) receive in times in which they have to cope with an unstable macroeconomic development are not offset by higher tax payments of these regions in the future to pay for this help. Rather, the higher tax payments must be financed by all member countries jointly. The main problem with this argument is - as discussed above - that in a world with inflexible relative prices the neo-ricardian equivalence theorem is unlikely to hold on any level. In such a case, it does not make sense to centralize fiscal policy for this reason.
In addition, it is argued that all regional (national) attempts to smooth real income over the business cycle in the presence of asymmetric shocks must fail if the production factors labor and capital are very mobile (Eichengreen, 1990; Sala-i-Martin/Sachs, 1992). If regions (countries) which are hit by adverse asymmetric shocks incur higher debt on the capital market, then the future tax burden of the private economic agents is increased. Hence, the incentive for the production factors labor and capital (enterprises) to migrate to regions with a lower tax burden rises. The economic decline of the respective region (country) can be accelerated, and some even fear that a complete depopulation might occur. Since regional (national) governments are afraid of such a development, they will incur less debt than necessary. Thus, they will hardly be able to stabilize the economic development. The ability to smooth real income is lower on the regional (national) than on the EC-level because production factors are surely more mobile on the former than on the latter level. Fiscal centralization seems to be the logical consequence.
This hypothesis, that a successful fiscal policy is impossible on the regional (national) level, depends crucially on the mobility of production factors. In spite of regional differences in income and employment possibilities in Europe, labor is still quite Immobile not only on the EC-level but also on the national level. It is therefore quite difficult to imagine that the production factors should react strongly and quickly to national differences in taxation. Especially the low mobility of the production factor labor enables countries to pursue independent national fiscal policies, which are not limited by the migration of labor. Following this line of thought thoroughly means also that national fiscal policies should not be restricted by institutional rules, like the fiscal convergence criteria of the Maastricht Treaty (Buiter, 1992). However, it is often asserted that fiscal policy is hardly able to stabilize unstable macroeconomic developments, and that because of the peculiarities of the voting process public debt is generally too high in democracies (v. Weizsäcker, 1992). The fiscal convergence criteria could then be interpreted as a way to limit the negative effects of incomplete political markets (Issing, 1992; Giovannini, 1992; Kenen, 1992).
A further argument in favor of a centralization of fiscal policy in a monetary union are certain moral hazard problems on the national level, which could prevent a sufficient degree of fiscal stabilization after country-specific adverse shocks (Eichengreen, 1990). Without a credible no bail-out clause, member countries have an incentive not to pay back public debt, either partially or at all. Thus, the public debt to finance not only stabilization, but also allocative and distributive public tasks is generally too high. Hence, countries that have already accumulated a large public debt face a steep supply curve of credits. The danger exists that they will be rationed on capital markets. Such countries are then in a recession probably no longer able to smooth income and will probably ask for fiscal transfers from other countries for that purpose. To conclude that fiscal policy in Europe should be centralized in the EC seems unwarranted however. The more reasonable, i.e., simpler and cheaper, solution would be to create a really credible no-bail-out clause in a monetary union, thus preventing excessive fiscal policies of individual member countries.
Although there are only few arguments in favor of fiscal stabilization policy and practically none in favor of centralizing fiscal policy in a European Monetary Union, it is repeatedly emphasized that the monetary union in the USA only works well because of a system of fiscal federalism. The federal system of taxes and transfers is interpreted to be an automatic stabilizer and shock absorber. This would compensate more than one third of the negative effects of region-specific shocks on disposable income. If income is one dollar lower in one state, it is increased by 35 to 44 cents by fiscal policy. The larger part - 33 to 37 cents - are lower tax payments to the federal government, the smaller part - 1 to 8 cents - are financial transfers from the federal government (Sala-i-Martin/Sachs, 1992).
The hypothesis that the American system of fiscal federalism smoothes unstable regional income has been questioned. Taxes and government expenditures change with national income for stabilization policy purposes, while for redistribution policies the size of the tax burden and of government expenditures depends on national income. Hence, the fiscal system in the USA is only to a very limited extent able to stabilize the regional economic development. It is rather a massive redistribution program from the rich to the poor states (v. Hagen, 1991). If the income of a state is lowered by 1 dollar, only 10 cents are compensated by fiscal policy, 8 cents are lower tax payments and 2 cents are financial transfers. Since the stabilization part of centralized fiscal policy in the USA is therefore only very small, the USA cannot be used as an example to show that central fiscal stabilization is necessary in a monetary union.
Apart from this fundamental critique of the system of fiscal federalism in the USA, it is possible that the system of automatic stabilization works better in Europe than in the USA (Giovannini, 1992). Although there is no central fiscal institution on the EC-level, there exists already a system in Europe that works like an automatic stabilizer. While the system in the USA consists mainly of taxes and transfers via the federal government, Europe relies to a larger extent on the market mechanism. If EC-countries are hit by an adverse shock, they will demand credits on the European capital market. The means of finance are provided by the surplus countries, which are hit by positive or at least not so large negative shocks. It is therefore no surprise that fiscal deficits in Europe provide a similar protection against asymmetric shocks as the fiscal deficit in the USA (Atkeson/Bayoumi, 1991).
The European system is possibly superior to the American system because it is disciplined more by market forces, while the financial relations between states in the USA are not very transparent and only under the control of the federal government. The current system in Europe forces countries which are hit by adverse shocks to consider how they can pay back the public debt. Hence they are always forced to choose between real adjustment and merely financing adverse shocks. This decision can be avoided in a system of fiscal federalism. Such a system has a bias in favor of financing and thus has the danger of making single states dependent on transfers (Courchene, 1992).
Centralization of fiscal policy does not seem to be a necessary condition for a successful European Monetary Union. If incomplete economic and political markets, which not only reduce the adjustment capacity but also increase the adjustment burden of an economy, are the main reasons for unstable macroeconomic developments, then reacting to shocks with central fiscal policy does not make sense. An appropriate therapy would rather apply efficient measures in the field of “Ordnungspolitik” thus increasing the capacity of adjusting to shocks and reducing the probability and severity of policy-induced shocks.
It is therefore not only necessary to improve the Common Market Project 1992 in its rules concerning “Ordnungspolitik”. It is also necessary to avoid grave mistakes in the creation of the monetary union. It does not make sense to tie countries together in a monetary union when they already have great difficulty to cope with shocks adequately. These are also the countries which have difficulty accepting market principles and a strong preference for redistributional policies. Countries with differing preferences concerning these issues and concerning the short run trade-off between inflation and unemployment should not be tied together in a monetary union. This would only lead to increasing political tensions, endangering also the progress made so far in the economic integration process. Hence, at a later stage only countries with similarly well working economic and political markets should form a monetary union.
AgellJ. and M.Dillén“Macroeconomic Externalities: Are Pigovian Taxes the Answer,”Institute for International Economic Studies Seminar Paper #487 (1991).
AtkesonA. and T. A.Bayoumi“Do Private Capital Markets Insure Against Risk in a Common Currency Area,”Evidence from the United Statesmimeo.University of Chicago and IMF (1991).
BarroR. J. “Are Government Bonds Net Wealth” Journal of Political Economy No. 82 (1974) pp. 1095–1117.
BayoumiT. A. and B.Eichengreen “Shocking Aspects of European Monetary Unification” NBER Working Paper 3949 (1992).
BeanC. R. “Economic and Monetary Union in Europe” Journal of Economic Perspectives No. 6 (1992) pp. 31–52.
BeggD.“Alternative Exchange Rate Regimes: The Role of the Exchange Rate and the Implications for Wage-Price Adjustment,”The Economics of EMU European Economy 1 Ch. 10. (1991).
BertholdN. “Keynesianische versus Klassische Arbeitslosigkeit,” WISt 17 (1988) pp. 485–493.
BertholdN. “Arbeitslosigkeit in Europa - Ein schwer lösbares Rätsel” E.Kantzenbach and O. G.Mayer (eds.) Beschaftigungsentwicklung und ArbeitsmarktpolitikBerlin1992 pp. 51–87.
BertholdN. “Der Weg zur Europäischen Währungsunion - Überwindbare Hindernisse,” Aussenwirtschaft 47 (1992) pp. 175–205.
BertholdN.“Sozialunion in Europa - Integrationsfaktor oder Sprengsatz einer Wirtschafts - und Währungsunion”Tübingen (1993).
BertholdN.B.Külp “Kann eine Stabilisierungspolitik in Demokratien überhaupt Erfolg haben,” Jahrbuch für Neue Politische Ökonomie 8 (1989) pp. 45–61.
BoughtonJ. M.R. D.Haas and P. R.Masson “The Role of Exchange Rate Movements in Transmitting International Disturbances,” Macroeconomic Policy and Economic Interdependence (1989) pp. 180–213.
BuiterW. H. “Should We Worry About the Fiscal Numerology of Maastricht” CEPR Discussion Paper #668 (1992).
CalmforsL. “Lessons from the Macroeconomic Experience of Sweden,” Institute for International Economic Studies. Seminar Paper. No. 522 (1992).
CalmforsL. and H.Horn “Classical Unemployment, Accommodation Policies and the Adjustment of Real Wages,” Scandinavian Journal of Economics No. 87 (1985) pp. 234–261.
CohenD. and C.Wyplosz “The European Monetary Union: An Agnostic Evaluation,” Macroeconomic Policies in an Interdependent World (1989) pp. 311–337.
CourcheneT. J.“Reflections an Canadian Federalism: Are There Implications for European Economic and Monetary Union?”Gutachten fur die EG-Kommission (1992).
CrossR. B.“Unemployment, Hysteresis and the Natural Rate Hypothesis” (1988).
De GrauweP. “German monetary unification,” European Economic Review No. 36 (1992) pp. 445–453.
DooleyM. P.J. A.Frankel and D. J.Mathieson “International Capital Mobility: What Do Saving-Investment Correlations Tell Us?” IMF-Staff Papers No. 34 (1987) pp. 503–530.
EC-Commission “One Market, One Money: An Evaluation of the Potential Benefits and Costs of Forming an Economic and Monetary Union,” European Economy No. 44 (1990).
EichengreenB. “Is Europe an Optimum Currency Area?” CEPR-Discussion Paper No. 478 (1990)
ErmishJohn (1991) European Integration and External Constraints on Social Policy: Is a Social Charter Necessary?in:National Institute Economic Review Nr. 136 (May1991) pp. 93–108.
EuckenWalter (1990) Grundsätze der Wirtschaftspolitik. 6. Aufl., Tübingen 1990 Franz, Wolfgang (1987), Hysteresis, Persistence and the NAIRU: An Empirical Analysis for the Federal Republic of Germanyin:RichardLayard and LarsCalmfors (eds.) The Fight against Unemployment. Cambridge, MA1987 pp. 91–122.
FriedmanMilton (1953) The effects of a full-employment policy on economic stability: a formal analysisin:Milton Friedman Essays in Positive EconomicsChicago1953 pp. 117–132.
GiovanniniAlberto (1992) Discussionin:Matthew B.Canzoneri et al (eds.) Establishing a central bank: issues in Europe and lessons from the US. Cambridge1992 pp. 222–227.
GiovanniniAlberto and LuigiSpaventa (1991) Fiscal Rules in the European Monetary Union: A No-Entry ClauseCEPR Discussion Paper #516London1991.
GoodhartCharles A. and S.Smith (1992) Stabilization. Gutachten fur die EG-KommissionBrüssel1992.
GreenwaldBruce L. and Joseph E.Stiglitz (1988) Imperfect Competition, Finance Constraints, and Business Fluctuationsin:Meir G.Kohn and Sho-chiehTiang (eds.) Finance Constraints Expectations and MacroeconomicsOxford1988 pp. 103–140.
GreenwaldBruce L. and Joseph E.Stiglitz (1990) Macroeconomic Models with Equity and Credit Rationing. NBER Working Paper #3573. Cambridge, MA1990.
GrönerHelmut (1967) Devisenausgleich - Ein wirksames Instrument der Zahlungsbilanzpolitik?in: Wirtschafts-politische Chronik H. 2/3 (1967) pp. 209–240.
GylfasonThorvaldur and AssarLindbeck (1984) Union Rivalry and Wages: An Oligopolistic Approachin:Economica 51 (1984) pp. 129–139.
HornHenrik and TorstenPersson (1988) Exchange Rate Policy, Wage Formation and Credibilityin: European Economic Review 32 (1988) pp. 1621–1636.
IngramJames C. (1959) State and Regional Payments Mechanismsin: Quarterly Journal of Economics 73 (1959) pp. 619–632.
IssingOtmar (1992) Disziplinierung der Finanzpolitik in der Europäischen Währungsunion?in: Deutsche Bundesbank Auszüge aus Presseartikeln Nr. 21v. 18März1992 pp. 1–6.
KenenPeter (1969) The Theory of Optimum Currency Areas: An Eclectic Viewin:Robert A.Mundell et al (eds.) Monetary Problems of the International EconomyChicago1969 pp. 41–60.
KenenPeter (1992) EMU After Maastricht. Group of Thirty. Washington, D.C.1992.
KeynesJohn M. (1936) The General Theory of UnemploymentInterest and MoneyLondon1936.
LayardRichardStevenNickell and RichardJackman (1991) Unemployment. Macroeconomic Performance and the Labour MarketOxford1991.
LindbeckAssar (1990) The Swedish Experience. Institute for International Economic Studies Seminar Paper #482Stockholm1990.
LindbeckAssar and Dennis J.Snower (1988) The Insider-Outsider Theory of Employment and UnemploymentCambridge, MA1988.
LundbergErik (1972) Discussants on Professor Lutz’ Paperin:FritzMachlup et al (eds.) International Monetary Problems American Enterprise Institute for Public Policy Research. Washington, D.C.1972 pp. 127–131.
LutzFriedrich (1972) Foreign Exchange Policy and European Economic Integrationin:FritzMachlup et al (eds.) International Monetary Problems American Enterprise Institute for Public Policy ResearchWashington, D. C.1972 pp. 107–123.
MarsdenDavid W. (1989) Occupations: The Influence of the Unemployment Situationin:Willem T.Molle and Aad v.Mourik (eds.) Wage Differentials in the European CommunityAvebury1989 pp. 105–139.
MassonPaul R. and JacquesMélitz (1991) Fiscal Policy Independence in a European Monetary Unionin: Open Economies Review 2 (1991) pp. 113–136.
MassonPaul R. and GuyMeredith (1990) Domestic and International Macroeconomic Consequences of German Unificationin:LeslieLipschitz (ed.) German Unification: Economic Issues IMF Occasional Paper No. 75Washington, D. C.1990 pp. 93–114.
MassonPaul R. and Mark P.Taylor (1992) Common Currency Areas and Currency Unions; An Analysis of the IssuesCEPR Discussion Paper #617London1992.
McDougall Report (1977) Report of the Study Group on the Role of Public Finance in European IntegrationBrüssel1977.
MeadeJames E. (1957) The Balance-of-Payments Problems of a European Free-Trade Areain:Economic Journal 67 (1957) pp. 379–396.
MundellRichard A. (1961) The Theory of Optimum Currency Areain: American Economic Review 51 (1961) pp. 756–764.
MusgraveRichard A. (1961) Approaches to a Fiscal Theory of Political Federalismin:NBER Public Finances: Needs Sources and UtilizationPrinceton, N. J.1961 pp. 97–133.
OatesWallace E. (1991) Studies in Fiscal Federalism Worcester1991.
PerssonTorsten (1992) Politics and Economic Policy, Institute for International Economic Studies. Seminar Paper #518Stockholm1992.
PurvisDouglas D. (1991) The Federal Proposals and Economic Unionin:R.Broadway (ed.) Economic Aspects of the Federal Government’s Constitutional Proposals Policy Forum Series Kingston 1991 Roubini Nouriel (1989) Leadership and Cooperation in the European Monetary System: A Simulation Approach NBER Working Paper # 3044Cambridge, MA1989.
Sala-i-MartinXavier and Jeffrey D.Sachs (1992) Fiscal federalism and optimum currency areas: evidence for Europe from the United Statesin:Matthew B.Canzoneri et al (eds.) Establishing a central bank: issues in Europe and lessons from the U.S.Cambridge1992S. pp. 195–219.
ScitovskyTibor (1957) The Theory of the Balance of Payments and the Problem of a Common European Currencyin:Kyklos10 (1957) pp. 18–44.
TriffinRobert (1972) Discussants on Prof. Lutz’ Paperin:FritzMachlup et al (eds.) International Monetary Problems. American Enterprise Institute for Public Policy ResearchWashington, D.C. 1972 pp. 125–127v. HagenJürgen (1991) Fiscal arrangements in a monetary union. Evidence from the U.S. Indiana Center for Global Business The School of Business Indiana University Discussion Paper #58Indiana1991. v.WeizsäckerRobertK. (1992) Staatsverschuldung und Demokratie in: Kyklos 45 (1992) Nr.1 pp. 51–67.
WalshCare E. (1992) Fiscal Federalism: An Overview of Issues and a Discussion of their Relevance to the European CommunityGutachten für die EG-KommissionBrüssel1992.
This paper was written during my stay at the Research Department of the IMF. I want to thank Peter Clark and Thomas Kruger, and especially my assistants Andreas Brandt, Rainer Fehn, Michael Gromling and Wolfgang Modery for helpful comments, without implicating them.