Journal Issue

Economic Policymaking in a Federation

International Monetary Fund. External Relations Dept.
Published Date:
January 1992
  • ShareShare
Show Summary Details

Policy-making rules in a federal state must be carefully designed to balance responsibilities between the center and the federating units

The traditional structure of unitary states and centralized federations is being challenged in many parts of the world today. Federations such as the Soviet Union and Yugoslavia have disintegrated, regions in various other countries are vigorously demanding greater autonomy or even independence, while some groups of countries are considering ceding new powers to supranational institutions. These splintering or coalescing movements, which everywhere occupy the newspaper headlines, have a profound effect on the daily lives of the people involved. Any major redistribution of responsibilities between different levels of government affects economic policymaking, and thus economic performance. Beyond economics, such actions have important political and social consequences.

While some decentralization of administration and policymaking is found in almost every country, the powers enjoyed by lower levels of government in a federation are extensive and constitutionally enshrined. A very loose federation is often called a confederation, which may be defined as a lasting alliance of sovereign states that contribute regularly toward the achievement of common objectives. From an economic viewpoint, a federation can perhaps best be distinguished from a confederation by the existence of an important and powerful central government, which can impose its will on lower levels of government within a defined sphere of competence. In order to isolate the issues connected with federalism from those that arise whenever groups are motivated to cooperate, the ensuing discussion assumes the presence of a federal authority that enacts policies for the country as a whole.

A critical issue, therefore, is the balance of economic power between the center and the component regions, be they republics, states, or provinces. The underlying theme of our analysis is that the flexibility and the accommodation of diversity offered by a federal constitution may conflict with the need for stability, decisiveness, and equity. To contain this conflict and preserve a federation, at least some policy instruments, such as monetary and trade policy, must be reserved for the central government. The best allocation of other instruments is contingent on the economic structure of the country, its political traditions, and the degree of trust shown by regions toward one another and toward the center. It is, perhaps surprisingly, hard to find general principles along which to assign responsibilities and privileges such as access to borrowing, contributions by regions to the federal budget, and a system of transfers. A discussion of a sample of such topics will be used to illustrate the types of economic arguments that generally bear on the issue of federalism, given a framework of distinct regions united in a single state. An analysis of the historical process whereby regional and national allegiances are formed and perpetuated is, of course, beyond the scope of this article.

Decentralization: pro and con

Debates over federalism can be highly emotional, involving split loyalties, old grievances, and, sometimes, atavistic prejudices. Often lacking is a cool consideration of strong economic arguments for a greater or lesser degree of federalism.

One rationale for decentralized policymaking rests on the condition that, despite considerable common interests, regions differ significantly in terms of natural resources, industrial specialization, and preferences. A regional authority may be motivated to assist certain sectors that would be ignored under national, uniform policymaking, and to promote certain activities over others. It may also seek control over its education system and other institutions in order to preserve a distinct cultural or national identity. In Canada, for example, the demands of Quebec and the indigenous peoples for cultural autonomy are bound up with conflicts over claims to natural resources.

Decentralization may also improve the functioning of government by making it more responsive to the will of its constituents, and more concerned about the efficiency of its operations than a distant central authority. People tend to feel more involved and influential with regional government than they do with government in a far-off capital, and so regional government is likely to be better informed of people’s wishes and better monitored. The nationalist movement in Scotland, for example, is motivated in part by a feeling that the Parliament in London will always be preoccupied with English concerns.

Further, regions have to compete with one another in attracting mobile factors of production, which may take the form of, say, new investment. Regional governments, therefore, have incentives to offer an appealing mixture of low taxes and high quality public services demanded by their constituents. Otherwise, they risk losing investors and productive families to other parts of the federation. Regions are certainly aware of this issue—even going so far as to take out advertisements in the press lauding their respective transport connections, cultural amenities, and so on.

But regional governments that ignore, exploit, or are simply unaware of the broader effects of their separate spending on public goods, deficit financing, and so on, may find that everyone in the country is made worse off by the lack of cooperation among the constituent units of a federation. Moreover, competition to offer lower taxes that attract mobile factors may leave regional governments with too few resources to fund adequate public services and may distort the tax system so that immobile factors are overly burdened. Conversely, regional governments may feel little need to devote resources to “public goods,” for example, the prevention of pollution that may be borne away to others by air and water. Even though each region is glad of public spending in other regions that benefits the federation as a whole, its own incentives are to “free ride” on such investments by others. On a more technological level, regions may be too small to exploit all economies of scale in the provision of public goods.

Indeed, unmitigated decentralization may lead to unacceptable inequities. Whereas competition in the marketplace will result in the contraction of inefficient firms and the re-employment of resources elsewhere, fiscal competition among federating units may leave some areas as backwaters of poverty and poor public services. Unlike an enterprise, a region that is not well endowed does not go out of business but declines and loses its tax base, while the needs of its remaining population go umnet. One may cite in this context the fate of many inner cities in the United States, or regions of Europe that depended on declining coal and steel industries. When public education is the issue, equality of opportunity is a precondition for an open and democratic civil society, but unrestrained decentralization cannot guarantee its adequate funding. Therefore, some form of federal support may be needed to sustain the degree of equality in the provision of public goods desired by the public at large. How does one ensure that this will be done efficiently and effectively?

Division of responsibilities

Some areas of competence must of necessity be reserved exclusively for the federal government. In most federations, it is generally accepted that the internal market in goods, services, labor, and capital must be unimpeded, and that the domestic mobility of finance must be maintained, because such a “level playing field” is necessary for the efficient allocation of resources and healthy competition between enterprises, industries, and regions. As a result, external economic and trade policy must be unified; without internal borders, a regionally differentiated trade policy would be vitiated by goods market arbitrage, and the net effect would be the waste of resources on cross-shipping motivated purely by tax considerations.

These are the types of areas where the federal role need not be so much to command resources, as to provide a framework for cooperation between regions and serve as an enforcement mechanism. The setting of technical standards is another such area.

If the members of a federation share a common currency, the conduct of monetary policy also needs to be centralized. Otherwise, competitive money creation may foster explosive inflation. Before that state is reached, regions will want to break the monetary union by introducing their own currencies, as one has seen happening in the former Soviet Union. A common currency implies the centralized management of reserves, the exchange rate, and any restrictions on convertibility. Finally, a federal judiciary is needed to resolve disputes between regions and between regions and the center, and to enforce cooperative agreements between them.

There may be strong arguments for the retention by the center of other powers. Banking supervision and regulation need to be uniform across the currency area, or at least minimum standards need to be imposed. If regions were charged with prudential regulation, and at the same time the central bank of the federation were to act as lender of last resort and insure depositors against bank runs, any one region would be inclined to assist its financial sector by instituting relatively lax regulation on its own banks, safe in the knowledge that the system was underpinned by the national monetary authority. It is noteworthy that even those central banks that are organized on a regional basis, as in the United States and Germany, function as centralized institutions.

Fiscal policies

In most countries, the central government reserves the right to use fiscal policy to influence aggregate demand. But there is no reason to suppose that only the federal deficit matters. In an integrated national economy, the transmission of fiscal impulses between regions will be rapid and powerful. If regional governments attempt to conduct independent fiscal policies, the net result may be far from optimal. It is possible, for example, that no one regional government would be inclined to stimulate flagging demand by incurring a deficit, if the main effect is to draw in imports from neighboring regions.

One solution to this problem is to set up some elaborate mechanism to coordinate the fiscal policies of regional governments; the mechanism is likely to be cumbersome and time-consuming, but it may be unavoidable if fiscal sovereignty is highly valued by regions. The alternative solution is to leave fiscal policy solely in the hands of the central government. In a country without well-developed capital markets, centralized fiscal policy also finesses the issue of the allocation of bank credit to regional governments, which could otherwise be the subject of endless quarrels between the center and the federating units and could undermine monetary policy.

However, a centralized fiscal policy can be detrimental to regions suffering from unusual disturbances relative to the nation as a whole. A case in point is Slovakia, whose industrial structure was built largely around the premise of the Soviet market as a supplier of cheap energy and raw materials and an outlet for Slovak heavy industry. Not surprisingly, Slovak policymakers have put forward demands for direct compensation or the development of alternative policy instruments, such as special financing facilities, as the price of their continued participation in the Czechoslovak federation. Note, though, that as economic development and integration between regions lead to convergence in economic structures, shocks can be expected to have more similar effects across the federating units.

Whatever the role of regional budget balances in short-term stabilization policy, the temptation for regions to run unsustainable deficits may be dangerous. Forces in competitive financial markets, and especially international financial markets, may conceivably discourage national governments from accumulating very large amounts of debt, but these forces will not trouble a regional government, for usually it can be almost certain that if need be, it will be saved from bankruptcy. Hence, there are grounds for enforcing budgetary discipline on regional governments; such reasoning motivated the clauses limiting deficit financing included in the recent Maastricht accord between European countries on the move toward Economic and Monetary Union.

If it is accepted that the aggregate fiscal position is to remain a purely federal responsibility, the size and composition of resources channeled through the federal budget must be such that fiscal imbalances can be prevented in a timely and effective manner. Traditionally, it has been argued that such flexibility requires the careful matching of expenditure responsibilities ‘with the sources of revenue at all levels of government. The problem is that efficiency considerations—to collect revenue at minimum cost in terms of utility and resources, and to provide public goods in the mix and quantity desired by the population—may dictate an asymmetric distribution of taxes and public spending across different levels of government, so that close matching or revenues with expenditures is impossible.

Receipts from customs duties and value- added tax, for example, seem to belong in the “natural” tax base of the federal government. Their tax bases are very mobile, and decentralized administration of them, while feasible, would involve large opportunity losses and could impede free circulation of goods. These receipts, however, may far exceed what is needed to finance the few pure public goods, such as defense, which are clearly most efficiently supplied at the federal rather than regional or local levels. In addition, political and constitutional considerations frequently dictate tax and expenditure assignment patterns that deviate significantly from those advocated by economists.

An extreme case was Yugoslavia, where the federal budget never exceeded 10 percent of national income, and most expenditure and revenue was legally mandated; no scope was allowed for active fiscal policy for either macroeconomic or microeconomic purposes.

The assignment of taxes and expenditures, then, is inevitably a second-best exercise, and no general assignment rules can be designed without knowledge of the constraints and distortions in the system. At the most practical level, it is quite feasible in this age of computerization for many parts of the tax base to be assigned to either the federal or regional governments, or to be shared.

The absence of close matching between revenues and expenditures implies that different levels of government need to cooperate and coordinate their activities, and that vertical transfers are an essential element in the design of every federal fiscal system.

Reverse transfers

Some very loose federations have relied on contributions from regions (known as “reverse transfers”) to complete federal financing. In historical episodes where reverse transfers were important, as in the United States in the 1780s and Germany after 1871, they prompted lengthy and acrimonious disputes between regions, and between regions and the center, over their level and distribution. Whereas taxes on economic activities are seen as the responsibility of individual households and enterprises, reverse transfers are easily interpreted as the immediate cost of belonging to the federation. Thus, regions are likely to demand an equally immediate fair return on their contributions. If reverse transfers are renegotiated on a regular basis, the budgetary process may get bogged down in perennial bargaining; if they are determined according to some rule, they may become a very inflexible source of revenue, and regions that feel themselves even temporarily disadvantaged will clamor for revision of the rule. It may be a rational negotiating strategy for a region to delay all federal decisionmaking until it is satisfied with the distribution of reverse transfers. The federal government meanwhile will have little incentive to restrain its spending, if responsibility for its financing rests on regional governments. During the break up of the Soviet Union, the republics certainly expressed their feeling that the (un- elected) central government was spending their contributions profligately and compensating for any shortfall in receipts by money creation.

Reverse transfers can be defended on two grounds. First, they allow regional governments more freedom in organizing their own tax system. Second, when regional governments do not entirely trust each other or the federal government, reverse transfers may be the only way to adequately finance a desirable range of federal activities. Their very impracticality may usefully serve to limit the extension of central government power. This negative argument is most persuasive when a federation is newly established or has undergone a period of internal dissent and confidence in the system has yet to be earned.

Inter-regional transfers

The federal government spends not only on its own consumption and investments but it also provides subsidies and grants to lower levels of government and to individual households and enterprises. Such government transfers can prompt destructive inter-regional conflict because they affect the distribution of income among federating units.

The least controversial aspect of a transfer mechanism is its role in supplementing private markets in the provision of a form of economic insurance for all members of a federation. A federal system of taxation and social insurance helps stabilize incomes across the federating units: regions enjoying a boom will tend to pay more taxes, and regions in recession will tend to receive more unemployment benefits, income support payments, and so on.

Transfers may also be used by the federal or state government as inducements for lower-level governments to provide the quantity and type of public good desired by the grantor. A matching grant from the federation, whereby the federal government assumes a share of the cost of a program, in effect reduces the marginal cost to the region of providing a service, and so counteracts the tendency to undersupply public goods. (Non- matching grants from the center are fungible, that is, they can be offset by reductions in state spending; they will have only an income but no substitution effect.) Regions may become dependent on matching grants so that the federation can impose its will on regional governments by making the release of specific matching funds conditional on acceptance of a host of federal policies. For example, in the United States, the federal government instituted a national speed limit by threatening to withhold highway funds from states that did not conform. Transfers and loans to regions to promote economic development contain the same danger of creeping centralization and politicization.

The most problematic form of transfers are those designed to equalize the provision of public goods across regions, or to equalize income generally. Sustained transfers to those who are less well off may sometimes be motivated by fear of dissent, or by the feeling among the fortunate that their fate might change, but in the end each country must make a normative choice about how concerned it is about inequality. In federations of distinct regions, there may be very little sympathy for the plight of the less developed regions, and corresponding resistance to any large transfers to them. A notorious example of such indifference is provided by the passivity of the British Government when faced with the Irish potato famine in the 19th century. Indeed, richer regions in a federation may even be able to negotiate net transfers from the poorer regions, because they could best withstand the dissolution of the union.


Some general principles can be discerned in the design of a practical federal system. Any rules defining responsibilities and intergovernmental relations need to be based on relatively simple criteria that are open to prompt verification. The closest possible link between decisions on the level and composition of spending, and those concerning its funding, is important in promoting accountability. Provided that the federal government can define the framework within which policy is conducted, so that the incentives for “beggar-thy-neighbor” policies are minimized, considerable scope can be allowed to regional governments to pursue their separate agendas and to compete to provide the best public services in the most efficient manner.

Where the centre has responsibility, it must control enough resources and instruments to act in a decisive and effective manner. The worst combination is likely to arise in a system half way between a very loose confederation and a federation with a powerful central government; then regions will have strong incentives to compete to seize control of the center and to obtain benefits at the expense of others.

This danger points to the importance of organizational procedures and the representation of regional interests in federal institutions—subjects that have not been discussed here because of lack of space. Regions will consent to the delegation of extensive powers to the central government only if each is confident that its voice will be heard in the formulation and development of policy. At the same time, regions will have to recognize the necessity of granting the center operational flexibility in its area of competence.

Ultimately a federation will survive only if every region is willing to compromise on occasion to promote the common good, in anticipation that others will do likewise. Without a modicum of mutual respect, no group can be sure that its vital interests will not be ignored or that it will not be the victim of the “tyranny of the majority.” If this level of trust cannot be achieved, it may be better to dissolve the union and minimize the scope for conflict.

Available May 18th in three languages


This 15th edition explores the relationships between development and the environment and notes the sometimes high costs of environmental damage for human health and productivity in developing countries. It identifies the need to embrace policies that are both economically rational and environmentally responsible.


  • Nine chapters on specific subjects such as ecological problems, energy and industry, and international environmental concerns

  • A special statistical section – in the English edition only – presents detailed information on water supply, air pollution, land use, and other environmental data.

  • World Development Indicators – comprehensive, current, and reliable data on social and economic development.

Development and the Environment

_ English hardback edition / 304 pages / ISBN 0-19-520S77-3 / US$32.95 j Order Stock #60877

Paperback editions – all US$16.95

_ English / ISBN 0-19-520877-3 / Order Stock #60876

_ French / ISBN 0-8213-2070-X / Order Stock #12070

_ Spanish j ISBN 0-8213-2071-8 / Order Stock #12071

Coupon orders accepted only from the U.S.A. and France. To order from other countries, request a list of World Bank Publications Distributors. Prices vary by country.

In the U.S. return the coupon to: World Bank Publications P.O. Box 7247-8619 Philadelphia, PA 19170-8619

In France return the coupon to: World Bank Publications 66, avenue d’léna 75116 Paris, France

Daniel Hardy

Dubravko Mihaljek

Other Resources Citing This Publication