The increased likelihood of significant advances in European monetary integration and even of European monetary union has given a new importance to the coordination of the fiscal policies of member states. The issues raised range from the design of an appropriate institutional framework for the coordination of such policies with both national governments and the European monetary authority to the theoretical exposition of the externalities and welfare economics of fiscal policy coordination.

In Discussion Paper No. 418, Research Fellow Willem Buiter and Kenneth Kletzer focus on the relationship between the `Eurofed' proposed by the Delors Report and the fiscal policies of member states. They note that the recommendations for coordinated and effective limits on national governments' budget deficits and on their recourse to external public borrowing appear to have been dropped, and they note that the assignment of primary responsibility for the EC's external exchange rate to the fiscal authorities that this implies appears to be inconsistent with the independence of Eurofed.

With high capital mobility, Eurofed can only control the external exchange rate if it controls all changes to the money supply. European fiscal authorities will then be able to influence the exchange rate only by influencing money demand, and the relative inflexibility and uncertain effects of the instruments available to them may render this option impractical. A Eurofed that is formally independent, but does not have the substance of independence, may nevertheless be forced to inflate away part of the debt overhang. Eurofed's anti-inflationary credibility may be enhanced, however, if its directing authority has a known extreme even irrational aversion to inflation.

With a single currency and perfect capital mobility, national governments relinquish both seigniorage and the use of national monetary policy for stabilization purposes. Since national public spending can therefore be financed only through taxes or public borrowing, interest rates for a given risk and return will be equalized. Hence, if Country I allows rapid growth of its public debt/GDP ratio while Country UK aims to pay off its national debt quickly, the premium on the rate of return on Country I's debt will reflect differences in risk on the two debts.

Buiter and Kletzer argue that the central imposition of limits on budget deficits are not required to safeguard against the risk of default by Country I, provided that Eurofed plays the central bank's role as lender of last resort. A `spillover', whereby Country I's borrowing raises interest rates not only for itself but also for all other borrowers, would only justify intervention on grounds of efficiency if market prices do not properly measure scarcity.

In Discussion Paper No. 419, Buiter and Kletzer study the international transmission of fiscal policy in a two-country model with perfect international capital mobility, in which the substitution of public borrowing for current taxation tends to boost private consumption and thus either `crowds out' interest- sensitive components of private demand or increases the external current account deficit.

The fact that the fiscal actions of one country may affect others by changing the common world rate of interest is sometimes taken as prima-facie evidence of an efficiency loss due to a fiscal policy externality, which may then be internalized through international fiscal policy coordination. A proper evaluation of the welfare consequences of such spillovers requires a distinction, however, between `technological' and `pecuniary' externalities. Technological externalities do not work through the price mechanism and may justify intervention on efficiency grounds. Pecuniary externalities affect the input or output prices of other agents: under standard assumptions of Pareto optimality, these are a legitimate welfare concern, but only on distributional grounds.

Buiter and Kletzer focus on arguments for coordination when the efficiency conditions of Walrasian equilibria are satisfied. They show that anything the government can achieve with non-zero public debt and unbalanced budgets can also be achieved with zero public debt and balanced budgets, if there is a sufficiently flexible set of intergenerational lump-sum taxes and transfer instruments. Hence the effect of the public debt depends on the extent of intergenerational income redistribution the government achieves through fiscal policies.
Coordination of national fiscal policies is not required to achieve Pareto-optimal allocations, but different national fiscal strategies will result in different interest rates, which will redistribute income between debtors and creditors, both within and between countries. These conclusions are not affected by the introduction of long-dated private or public debt, but introducing distortionary taxes does provide an efficiency argument for policy coordination, since the distortions arising from source-based or residence-based taxation of capital income will otherwise tend to reduce capital formation and saving.

In Discussion Paper No. 420, Buiter and Kletzer assess the rationale for the design of cooperative fiscal policy and evaluate economic performance in terms of both efficiency and distributional effects within and across national economies, again with reference to the role of fiscal policy in intergenerational income distribution. Here too they use a two- country model of the global economy with perfect international mobility of financial capital. Private savings may be influenced by policies designed to affect the rate of return or to redistribute income between the the working population and the retired.

Buiter and Kletzer present national and global social welfare functions, derived as weighted averages of the lifetime welfare of all national residents and of the two national social welfare functions. With no distortionary taxes and transfers and no public spending externalities, each national government can redistribute freely between its own working and retired populations in each period, and fiscal policy coordination is not required for `household' Pareto efficiency. Borrowing by one government will raise interest rates world-wide and this will crowd out capital formation at home and abroad, but this involves no efficiency losses. `National' Pareto optimality nevertheless requires fiscal policy coordination even in this case, and maximizing global social welfare also requires the coordination of international transfer payments.

If international lump-sum redistribution is not possible, two second-best schemes to maximize global social welfare may be considered. First, the two national social security retirement schemes may be used to influence the world interest rate and thus the distribution of income between the creditor and the debtor nation. Second, the international distribution of income may be influenced by changing payments to labour, which is assumed to be internationally immobile, by subsidizing capital formation in one country and taxing it in the other.

Buiter and Kletzer show that the optimal policy for a single large `active' country facing another large `passive' country is to tax international lending (or subsidize international borrowing) to influence the domestic intergenerational distribution of income.

They finally show that, in the absence of international coordination, national provision of `international public goods' (such as national expenditure on crime prevention when crime is global in scope) is Pareto inefficient in all three senses. Coordinating the supply of such goods internationally will suffice to ensure Pareto optimality with respect to preferences of individual households, but the achievement of outcomes that are efficient at the national level also requires the international coordination of their financing.

Reflections on the Fiscal Implications of a Common Currency
Fiscal Policy Interdependence and Efficiency
The Welfare Economics of Cooperative and Non-cooperative Fiscal Policy
Willem H Buiter and Kenneth M Kletzer

Discussion Papers Nos. 418 and 419, May 1990, and 420, June 1990 (IM)