2 Normative theories: Criteria for competence allocation between member states and the EU

2.1 Concepts of federalism

The concept of a federation has implications for the allocation of powers between the European and the national level, depending on whether it points to a federal or confederal polity. Scholars supporting the idea of a federal Europe or regarding it as a legal reality conclude that the EU needs a constitution defining the distribution of powers (Auer 2005). How powers should be actually allocated to the different levels is left open, but the notion of a federal Europe implies a limitation of the EU’s competences as well as representative institutions empowered to make decision-making by qualified majority rule, i.e. a structure that allows for further integration (Trechsel 2005). From a normative point of view, federalists usually tend to plea for a European government with all powers to fulfil regulative, distributive and redistributive functions.

In fact, this idea – which was supported by leading European politicians during the 1950s and 1960s – has shifted to the background. Although it gained new momentum during the discussion on the Constitutional Treaty, scholars widely acknowledged that the EU combines federal and confederal elements (Burgess 2000: 260–265; Elazar 2001: 36–37) and they only disagree on the relative weight of these elements. Giandomenico Majone (2004) explained this mixed structure – which, in his view, leads the EU into serious dilemmas – as the result of an integration dynamic fostered mainly by executives. Following Fritz W. Scharpf (1999), he strongly makes the case for a confederal Europe with powers limited to regulatory policies of “negative integration”, while under the condition of social heterogeneity, redistributive policies should be left to the member states. Hence, competences of the EU should be restricted to those required for a market-preserving federalism (Weingast 1995): “Aside from foreign and security policy, the public agenda would mostly include efficiency-enhancing, market-preserving policies – a combination of liberalization and negative integration measures to remove obstacles to the free movement of people, services, goods, and capital within the territory of the federation. (…) In contrast, redistributive policies can only be legitimated by majority decisions and hence place too heavy a burden on the fragile normative foundations of a transnational policy” (Majone 2004: 191). Reflecting on the plurality of the European “demoi” and the social inequalities between member states, others plea for a polycentric polity resulting from flexible integration (Wind 2003). With this concept the allocation of competence would not only depend on constitutional rules but also on member states’ decisions on “enhanced cooperation” or “opting-out”.

The problem with both the federal and the confederal model is that they cannot determine in detail which competences should be allocated to the EU and which should remain at the national or sub-national level. “Federalists” make the decision a matter of constitutional policy-making, but they cannot provide a convincing normative theory of centralisation and decentralisation, not to speak of proposals for coping with interdependent tasks cutting across levels. In the same manner, “confederalists” ignore that negative and positive integration, efficiency-enhancing regulation and redistributive policies cannot be clearly distinguished. Independent of the issues at stake, most policies have redistributive implications which are obvious if we consider the effects of a free flow of goods, services and capital in the market on the territorial distribution of wealth. Efficiency-enhancement is a positive-sum game, but at the end of the day some may profit more than others, considering the fact that in many cases external effects occur. All this makes it impossible to clearly delimit the EU’s competences according to the proposed categories.

Independent of its federal or confederal character, a multilevel polity requires rules for an allocation of powers between levels and political mechanisms designed to enforce these rules. Therefore, the normative debate on the kind of the European constitution necessarily raised the competence issue. Political scientists mainly contributed to discussions on the second question of how to enforce competence rules. The first question has been dealt with by lawyers and economists. While the former have proposed principles and categories of rules (e.g. exclusive, concurrent, shared competence), including rules for coping with rule conflicts, economic theory of federalism provides substantial criteria for determining how specific competences should be allocated between levels.

2.2 Economic theory of federalism

Economic theory of federalism (Oates 1999, 2005) states that, in principle, powers of governments should be applied at the lowest possible level. This principle of decentralisation (which conforms to the principle of subsidiarity) is supported by at least three arguments: First, decentralisation increases the chance that policy-making follows the preferences of citizens and that powers can effectively be controlled by citizens. Therefore, the more a society in a federal system reveals territorial cleavages, i.e. the more citizens with similar preferences are concentrated in regions, the more policies are to be decentralised. Second, if citizens are mobile, a decentralised polity gives them the opportunity to choose between different units in which governments offer different sets of public policies. Third, competition between decentralised governments induced by mobile tax-payers increases the efficiency of public policies, as governments have to provide for an optimal ratio of services and tax burden. On the other hand, for many policies, decentralised units are too small for effective governance and intergovernmental coordination causes considerable costs. Hence, centralisation of competences is justified if common goods reach beyond the scope of lower level governments, if they produce external effects, or if they cannot exploit economies of scale. Moreover, negative dynamics of competition require central regulation and so do economic disparities in public revenues which violate social norms of distributive justice (summarised in: Blankart 2007; Oates 2005; Persson, Gérard, and Tabellini 1997).

These arguments played an important role in discussions on the Stability and Growth Pact and the size of the EU’s budget. According to traditional reasoning on economic federalism, macroeconomic policy requires a central government to coordinate fiscal policies of lower level governments, because otherwise competing member states tend to exploit the common good of economic stability and growth in the common market. Redistributive policies at the European level determined to support regions in need are defended in order to countervail territorial disparities of economic development. However, both pleas for competence transfers to the EU have been contested. Economists and political scientists have criticised the ineffectiveness or instability of the current allocation of powers between the EU and its member states in these policy fields or they have pointed out the political costs of centralisation (summarised in: Alves 2007; Hallerberg 2006; McKay 2005).

According to this theory of federalism, inter-jurisdictional competition is a decisive reason for determining the allocation of competences. While until the 1990s the predominating model assumed that jurisdictions compete for mobile tax-payers and regarded this process as driven by the market mechanism (competing decentralised governments provide public goods demanded by private actors with the tax constituting a kind of price that has to be paid for public goods), some literature now has introduced a different model of competition which has different implications for the vertical allocation of competences. According to this concept of a “laboratory federalism” (Oates 1999: 1132–1134), horizontal intergovernmental competition in a decentralised political system should lead to experimentation with new policies and the diffusion of innovation (Salmon 1987; Breton 1996; Kerber 2005; Kerber and Eckardt 2007; Oates 1999). The conclusion again is that decentralisation should be preferred to centralisation because it promises to find best policies. However, this result cannot be expected under all conditions. The main problem which has to be solved concerns adequate incentives for governments to innovate and to mutually learn from each other. If decentralised governments search for mobile resources, they are strongly induced to improve policies in terms of efficiency, but this can lead them to a “race-to-the-bottom” in terms of quality of public goods or to ignoring external effects. Therefore, scholars like Albert Breton and Pierre Salmon argue that incentives should be set by the intrajurisdictional democratic process. However, they depend “on the possibility and willingness of citizens to make assessments of comparative performance”. Only “[i]f these conditions are fulfilled, comparisons will serve as a basis for rewarding politicians in power (re-electing them) or sanctioning them (voting for their competitors)” (Salmon 1987: 32). Leaving aside the question of whether citizens are able or willing to comparatively evaluate the performance of their government, incentive problems can occur if governments make policies without considering the will of their citizens and if accountability mechanisms are deficient (Kerber and Eckardt 2007: 237). These reasons imply that decentralisation may come to its limits with the consequence that interjurisdictional competition fails without central regulation. Therefore, mechanisms like budget constraints, rating or performance standards are recommended.

It is interesting to note that with this focus on the interplay of intergovernmental and intragovernmental politics, the economic theory of federalism approaches political theories of multilevel governance (Blankart 2007; Persson, Gérard, and Tabellini 1997). However, this makes the evaluation of costs and benefits of an allocation of competences to the EU much more difficult than in the basic model. In fact, one can argue that deficits in democratic politics at the EU level speak for decentralisation. However, in the same way the argument can be turned against decentralisation if one puts the emphasis on incentive deficits. On balance, political economy tends to recommend more decentralisation of competences, but also more effective decision-making at the European level by applying majority rule in order to enable the EU to fulfil coordinative and regulative functions (e.g. Blankart 2007; Salmon 2003: 128–129). As a rule, the recommendations for the allocation of competences between nation states and the EU remain rather abstract and rarely refer to particular policies.

In political debates, the economic reasoning on federalism has often been used to support pleas for a separation of powers. In fact, the economic model of a competitive federalism presupposes “fiscal equivalence” (Olson 1969). It requires that a government takes account of all costs and benefits of a public policy. On the other hand, differentiated reasoning based on this theory has to consider costs and benefits of decentralisation and centralisation, and more often than not the balancing leads to a rather complicated allocation of competence varying between regulative, executive and fiscal functions of a policy. Moreover, the degree of centralisation and decentralisation depends on the size of a jurisdiction with the consequence that the smallest unit of a multilevel system decides on the vertical structure. To avoid such a result, the sharing of powers turns out to be inevitable.

In view of this problem, Frey and Eichenberger (1999) proposed a model of functionally overlapping competing jurisdictions (FOCJ). The authors assumed that public services and demand for services by consumers “extend very differently over space and have different degrees of scale economies (or diseconomies)” (Frey and Eichenberger 1999: 5). Therefore, they argued that the traditional organisation of the public sector, where a government provides all services in a territorial unit, should be substituted by functionally specialised jurisdictions. As a consequence, a multilevel polity would constitute overlapping jurisdictions that should compete for the support of citizens in the same way as assumed by the economic theory of federalism. Frey and Eichenberger suggested that at least regional policy in Europe can be organised according to their model. Moreover, they presented the FOCJ-model as a promising concept for European integration under the condition of high diversity of national economies and cultures and refer to ideas like variable geometry or European integration by different speeds (“Europe à la carte”) (Frey and Eichenberger 1999: 79).

Frey and Eichenberger constructed a theoretical model for the allocation of competences in a territorially differentiated federal system. Their proposition that jurisdictions should be constituted by referenda goes beyond what appears to be practicable in real politics. Nevertheless, this particular model of a federation is not utterly idealistic and it was taken on by scholars interested in comparative research. Aiming at an analytical concept for EU multilevel governance, Hooghe and Marks (2003) suggested that in contrast to territorial federalism, functional federalism should be considered as particularly relevant in governance beyond the nation state.

However, these models of federalism, to which more could be added, leave us with the unresolved problem of how to allocate competences between the constituent units, be they territorial or functional jurisdictions. Beyond this, it is interesting to note that normative theories inspired by the economic reasoning on federalism increasingly tend to deviate from the original premise of a clear separation of powers and now react to the multidimensional and multilevel character of tasks governments have to fulfil. This leads them either to accept an allocation of powers where competences of individual levels concern particular functions of a policy or to think about a functionally differentiated organisation of jurisdictions. In either case the allocation of competence leads to interdependent policies at the different levels. Aside from managing coordination and accountability, this causes the challenge to maintain the stability of the multilevel polity.

2.3 Subsidiarity principle and constitutional delimitation of powers

Problems of clearly sorting out competences increase the risk of “authority migration” (Bednar 2004) beyond the scope that is accepted by citizens or necessary for an efficient policy. This has led scholars to inquire into safeguards against this trend. On the one hand, they proposed rules for a delimitation of competences in the Treaties; on the other hand, the discussion concerned procedures designed to prevent uncontrolled shifts of powers and to solve conflicts on their application.

One aspect of this discussion concerned the legal statement of competences in the Treaties. In an influential article, de Witte and de Búrca argued against an extensive catalogue of competences. Instead, they proposed to define categories of EU powers and to distinguish exclusive, shared and complementary competences in order to bring more clarity into the existing system (de Witte and de Búrca 2002). Moreover, they suggested more precise definitions of legal measures the EU can take. The authors warned against too strict regulations which may reduce the flexibility and adaptability of the EU. Their work has apparently influenced the relevant articles of the failed Constitutional Treaty and the Reform Treaty. With the treaty reform, exclusive and shared competences are defined, whereas the third category including coordinating, supportive and supplementary competences lacks the clarity required by constitutional lawyers.

Following the development of European law after Maastricht, the principles of subsidiarity and proportionality have acquired a prominent role in this research area. While lawyers tried to define the substance of these principles and to assess the potential consequences of their application in cases of dispute, political scientists have engaged in comparative research in order to evaluate the prospects and limits of such a constitutional rule. This research has been guided by the premise that it is not the content or the formulation but the procedures of enforcing these principles that allow for controlling a drift of competences to the EU. In view of the findings from comparative research on jurisprudence on competence issues in federal states (Berman 1994; Thorlakson 2006) as well as from research on the European Court of Justice (de Búrca 1998; Estella de Noriega 2002), which reveals mixed results as to the role of courts in reinforcing subsidiarity, political scientists rely more on participation of lower level institutions. Therefore, the procedure of subsidiarity control by national parliaments, invented during the elaboration of the Constitutional Treaty, is mostly supported although scholars are well aware of the limits of this device (Cooper 2006).

Based on an elaborated theory and comparative research, Filippov, Ordeshook, and Shvetsova (2004) raised doubts on the stabilizing effects of constitutional rules and constitutional courts. Instead, they suggest creating an integrated party system linking actors from all levels. In such a political structure, where all parties have the opportunity to gain power at each level, incentives to concentrate competences at one level are reduced. Whether the evolution of such a party system is a realistic option in the EU is debatable. Some scholars provided evidence for the emergence or the existence of a European party system. But empirical research provides no indication that its vertical integration is rather strong, even if some institutional factors like the extended powers of the EP and the structures of joint decision-making might work in this direction (Thorlakson 2005).

To sum up: In view of the integration dynamics after the Single European Act, the issue of how to delimit the competences of the EU became an important subject for research. It has been stimulated by the Convention working on a constitution for Europe which was triggered by the debate on the democratic deficit, but also by the end of the “permissive consensus” on European integration. So far, normative theories on federalism have shown that the problem can hardly be solved by constitutional rules defining competences for each level. Apart from general guidelines for the allocation of competences, they point out that the real issue is to find mechanisms for stabilising the endogenous dynamics of competence migration between levels. The procedure of subsidiarity control by national parliaments makes sense from this point of view. However, beyond dealing with proposals that are on the political agenda, scholars should analyse the causes of competence shifts in EU multilevel governance.

"The EU’s competences: The ‘vertical’ perspective on the multilevel system" Arthur Benz and Christina Zimmer http://www.livingreviews.org/lreg-2008-3

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 Austria License. Problems/comments to