Making Sense of Subsidiarity

In the Centre's fourth Annual Report on Monitoring European Integration, published on 23 November, a panel of CEPR Research Fellows call for a major reform of the European Community's spending priorities. The case for centralization is strongest in areas of policy in which Community-level expenditure is currently small (such as competition policy) and weakest in those whose budgetary cost is large (including agriculture and regional policy). This year's panel comprises David Begg (Birkbeck College, London), Jacques Crémer (Université des Sciences Sociales, Toulouse), Jean-Pierre Danthine (Université de Lausanne), Jeremy Edwards (St John's College, Cambridge), Vittorio Grilli (Consiglio degli Esperti, Ministero del Tesoro, Roma), Damien Neven (Université de Lausanne), Paul Seabright (Churchill College, Cambridge), Hans-Werner Sinn (Center for Economic Studies, Universität München), Anthony Venables (LSE) and Charles Wyplosz (INSEAD). The German Marshall Fund of the United States provided generous financial assistance essential to the completion of the Report, which was also supported by the UK Department of Trade and Industry. The views expressed are those of the authors, not those of the above organizations nor of CEPR, which takes no institutional policy positions.

The Report calls for a clear set of principles to govern the allocation of powers between the Community's central institutions and its member states. The Single European Act of 1986 represented a decisive step towards federation, but there is no formal mechanism for member states to delegate competences to the Community, apart from those specifically considered in the Treaty of Rome. Maastricht did not clearly establish subsidiarity as an instrument for the allocation or exercise of competences; it therefore remains a general political principle rather than a source of explicit guidance. The European Court of Justice, which must now implement this principle, therefore risks jeopardizing its hard-won credibility by straying into political territory.

Coordination of member states' policies yields benefits when there are significant scale economies or spillovers, and this may sometimes be achieved by collective agreement on rules which are then implemented at national level. Centralization of powers at the EC level is necessary, however, when coordination is desirable but its decentralized implementation is not credible. This also entails costs as the reduction in accountability allows centrally-determined policies to diverge from the best interests of constituent states, regions or localities. The appropriate location of power therefore reflects a trade-off between the costs and benefits of centralization. By laying the burden of proof on advocates of centralization, subsidiarity recognizes both the initial sovereignty of member states and the potentially substantial problems of `government failure' at the centre.

Enhanced mobility of capital and labour following the completion of the single market may increase fiscal competition among EC member states as mobile factors flee taxation. This will impede their ability to raise taxes to fund the welfare state, which is a potentially powerful argument for centralization. Switching to a cash-flow tax on capital taxation for firms may provide an alternative means of limiting fiscal spillovers, however, which would allow policies to be set at national level. Moreover, since intra-EC capital mobility is already high, some of the `doomsday' effects have already taken place: corporate tax rates and the highest marginal rates of personal income tax have already fallen substantially. Labour mobility is also limited by language, cultural and historical barriers, not only within the EC but also in such long-standing federations such as Switzerland; broadly based taxes on workers' income and spending should therefore continue to furnish substantial revenue for the time being. Pressures for fiscal competition are likely to grow as integration progresses, however, and Europe cannot continue indefinitely to exempt those who are best able to escape their national tax authorities from the burden of financing public expenditure. Escape routes must be closed or at least curtailed: accountability and distributive justice as well as economic efficiency require action.

The integration of European goods and capital markets also has implications for labour legislation. Government regulation of employment contracts is necessary to protect workers' interests in health and safety, for example, since employers do not bear the full cost of treatment in case of accident or illness and may therefore provide lower safety standards than are socially efficient. With properly functioning labour markets, however, individual member states will remain able to choose social and health standards that reflect their national preferences. There is no case for centralizing or even coordinating social policies.

Thus the Social Chapter of the Maastricht Treaty directly contradicts its principle of subsidiarity; indeed, its proposals for common EC standards for minimum wages risk a perverse redistribution of income away from poor workers in poor countries to poor workers in rich countries. Labour market imperfections may lead to social dumping if governments lure mobile capital via fiscal competition. Relaxing safety standards to help employers or ignoring the externalities this imposes on national health-care systems will shift the burden of taxation to less mobile factors, but governments have little incentive to engage in such competition. Since capital is already highly mobile, it confers little social profit: the public goods required to entice it cost almost as much as it brings in the tax revenue it confers.

Monetary union as proposed by the Maastricht Treaty enhances both the need for and effectiveness of national fiscal policies for macroeconomic stabilization. Arguments for greater centralization are unconvincing, since EC-level insurance against macroeconomic shocks would generate perverse behaviour and be attractive only to the higher-risk countries; adopting the practices of existing federations would necessarily entail a substantial increase in centralization by requiring compulsory participation and the harmonization of relevant institutions. Arguments for setting fiscal policies nationally on grounds of subsidiarity assume that each government can borrow subject only to its long-run budget constraint, however, while the Maastricht Treaty imposes ceilings on their deficits and debts. This has imposed enormous costs during the current recession as most governments have refrained from fiscal expansion. Pressures for EC central institutions to undertake active fiscal policy stabilization are consequently set to increase, thus violating the principle of subsidiarity.

The Community's clearest attempt to allocate powers of jurisdiction between its central institutions and its member states in accordance with subsidiarity concerns EC competition policy. The 1990 Merger Regulation requires that mergers and acquisitions between parties whose combined aggregate world-wide annual turnover exceeds ECU 5 billion be notified to the European Commission, unless they conduct two-thirds or more of their business in the same member state. The rider clearly relates centralization to the extent of spillovers by exempting cases whose effects will be primarily concentrated in one member state from EC jurisdiction. A review of EC merger policy to date illustrates that the gains from centralization may be substantial, but the design of central institutions must ensure that these are not dissipated through increased rent-seeking or regulatory capture.

The extent of central intervention in environmental policy that is consistent with subsidiarity varies substantially from case to case. For example, no government would seek to attract mobile factors of production by allowing its drinking water to be polluted, so any differences in the standards they choose must presumably reflect national preferences. The Community should therefore abandon its regulation of such standards and focus instead on cases in which international spillovers are significant, such as pollution of the Rhine, as well as taking measures to promote research and to develop standardized methods of measuring and reporting environmental standards.

Agricultural support accounted for almost 60% of the Community's budget and provides a striking example of the costs of inappropriate centralization of regulatory power. The wide disparities in the Common Agricultural Policy's net impacts on member countries provide its net beneficiaries with powerful incentives to resist reform. For a sector in which spillovers were initially small, centralization has paradoxically created large artificial spillovers which are seriously distorting negotiations on the future of regulation.

Successive reforms of the EC Structural Funds have shifted policy-makers' attention away from the simple distribution of regional assistance to lower-level jurisdictions towards taking direct responsibility for the selection and implementation of projects. They may do better to focus on providing technical assistance to regions in the selection of projects and authenticating their value to potentially sceptical net contributors, in which the centre can clearly play a useful coordinating role. The case for centralizing the power to choose projects assumes either that the Community systematically knows better than its regions what the best projects are (which runs against subsidiarity) or that spillovers are too complex to permit coordination (for which there is too little evidence). Centralization also provides member states incentives to bid for projects in order to manipulate the allocation of funds and not because they are the best projects.

The authors conclude by cautioning that their scepticism about centralization should give no comfort to those taking an `anti-federalist' rather than a `federalist' position in terms of the current political debate. Indeed, this is in many ways a misleading distinction, since the European Community is already essentially a federal state in all but name; its key difference with existing federations is that it has not yet come to terms with the fact. A clear statement of the principles of subsidiarity and an open attempt to ensure that the Community has the procedures and institutions required to put it into practice are all the more necessary now that the key steps towards federation have already been taken.