Towards a Comprehensive Framework for Understanding EU Enforcement Regimes

Abstract

Next to Member States, European Union (EU) level organisations have come to play a larger role in enforcement in the EU during the last two decades. Analysing the roles of Member States, networks of national authorities and EU agencies in this stage of the policy cycle through multiple academic lenses could lead to a more comprehensive understanding and assessment of their design. This article sets elementary steps towards a framework that brings together prominent theoretical insights from the legal and political disciplines, to discuss their combined leverage for understanding the EU legislator's choice for these enforcement regimes.

I. Introduction

Member States of the European Union (EU) have long been primarily responsible for enforcement in the EU. By default, ‘public action with the objective of preventing or responding to the violation of a norm’ by citizens or businesses has been undertaken at the national level.¹ Through case law and legislation, the EU vests a responsibility to enforce EU law in the Member States,² as is the case with the Environmental Crimes Directive (ECD) in the area of environmental policy. However, networks of Member States’ national authorities, EU agencies and their many manifestations also become increasingly involved in enforcement in the EU.³ One example is the European Securities and Markets Authority (ESMA), an agency that (among others) enforces EU law vis-a-vis credit rating agencies.⁴ The Consumer Protection Cooperation (CPC), moreover, is a network of national enforcement authorities that provide mutual assistance and jointly engage in investigations upon cross-border infringements of EU consumer law.

Taking these innovations into account, EU policy-makers can decide to vest enforcement tasks in three different regimes: the Member States, EU networks and EU agencies.⁵ EU policy documents, however, neither capture the varying qualities of these regimes nor do they consistently motivate a choice between them.⁶ While legal scholarship was quick to attribute functional motives to the EU’s involvement with enforcement by the Member States,⁷ researchers of European networks, agencies and multilevel implementation have not yet specified the EU’s motives for delegation of enforcement tasks specifically.⁸ Nonetheless, increased understanding may lay a better foundation for and reasoning of the choice for a particular enforcement regime. Therefore, this article aims to find out why certain regimes are chosen over others. Referring to the examples above: why would tasks for the enforcement of EU consumer policy be conferred upon a network, whereas considerable powers for the enforcement of EU policies for credit rating agencies are given to an EU agency? And what explains the fact that the EU continues to rely on and steer Member State enforcement through secondary legislation?

This article takes stock of insights from legal and political scholarship to understand the EU legislator’s choice for one or more of these three regimes for enforcement, and demonstrates that a combined account of both EU law and politics yields a more comprehensive picture. Taking relevant literature on European integration into account,⁹ I alter the insights of rationalist studies of European regulatory networks and European agencies for enforcement in the EU. Part II first illuminates the object of inquiry (EU involvement with Member State enforcement, EU networks and EU agencies for enforcement). Part III then discusses institutional and policy-specific conditions to demonstrate that room for choice between enforcement regimes remains. Parts IV and V discuss functional and political rationales.¹⁰ Integrating insights from law and politics, I then propose a framework in Part VI which allows for a more comprehensive study of the choice for one or more enforcement regimes. Part VII discusses the CPC and the ECD as empirical illustrations, as regimes for the enforcement of consumer protection and environmental policy respectively. I conclude thereafter.

II. Three regimes for enforcement in the EU

The EU has three enforcement regimes at its disposal, and each regime involves one (or more) public institution(s) on the national or EU level which, pursuant to EU law, jointly or separately perform(s) functions for compliance with EU law by citizens and businesses. As mentioned, the EU conventionally relies on the Member States. This decentralised regime is also referred to as indirect, because the EU institutions do not monitor, investigate or sanction (collectives of) citizens themselves,¹¹ but rather leave that to the Member States.¹² Although the latter were long autonomous in the formal and operational design of their enforcement systems, the EU increasingly controls Member State enforcement through regulatory obligations. Case law of the Court of Justice of the European Union (CJEU) sets general requirements,¹³ while secondary and soft law instruments contain more detailed standards for (among others) the nature of sanctions to be imposed upon violation of an EU rule (criminal, administrative or private), the establishment and architecture of national enforcement authorities and the attribution and use of enforcement powers.¹⁴

However, EU policy enforcement is subject to increased centralisation as two types of EU level organisations have become increasingly involved with the enforcement of EU policies in the past two decades.¹⁵ EU enforcement networks, first, consist primarily of national enforcement authorities. They are (bi- or multilaterally) connected with varying intensity: some merely pool staff or exchange best practices, whereas stronger ones share real-time supervisory information or engage in operational cooperation through mutual assistance. A network’s nodes thus comprise national public (and sometimes private) authorities but may also include EU agencies and the Commission. They exist formally when EU law explicitly provides for their enactment, although the Commission can participate in (informal) enforcement networks outside the scope of EU law.

Yet in other policies areas, enforcement responsibilities and instruments have (also) been attributed to EU agencies.¹⁶ An EU agency is conventionally defined as a permanent body which has its own legal personality on the basis of EU secondary law,¹⁷ though I include formal EU institutions such as the European Central Bank (ECB), Europol or specific Commission directorates – all with some degree of independence from EU political decision-making.¹⁸ An example of an agency is the 2005 established European Fisheries Control Agency, set up to coordinate control and inspection by Member States and later given the power to conduct inspections itself. An example of a stronger agency is ESMA, which can autonomously decide to fine credit rating agencies. EU agencies do not (yet) replace national authorities or their networks in their entirety.¹⁹

However, this distinction between regimes is blurred in practice. First, regimes are not necessarily limited to enforcement tasks, but can simultaneously fulfil advisory or rule-making functions. Moreover, regimes can coexist within one policy (field) without consuming each other. Table 1 therefore demonstrates the minimum variety of potential configurations of increasingly centralised enforcement of EU policies, whereby each configuration displays a different distribution of some enforcement responsibility among one or more of the identified regimes. Only a few configurations included in Table 1 currently exist. For instance, the Member States are given certain circumscribed responsibilities for the enforcement of the mercury regulation (example of policy 1).²⁰ For the enforcement of EU consumer protection policy (example of policy 3), the EU has vested tasks in the Member States and a network (see Part VII.1), while the Member States and an agency (ESMA) enforce (different) aspects of EU law on credit rating agencies (example of policy 5).²¹ Thirdly, there are various ways in which the EU spreads (formal) authority or (operational) enforcement tasks over multiple regimes. For example, an agency can hierarchically steer enforcement operations by national authorities (eg for EU policy on credit rating agencies); an EU agency and the national agencies can both have similar formal and operational tasks for enforcement of the same set of EU rules (eg for EU policy on airworthiness standards); or an agency can have specific operational powers to only support national authorities (eg the European Medicines Agency).²²

Table 1. Configurations of regimes for EU policy enforcement

This typology demonstrates that enforcement in the EU is more nebulous than anticipated by other frameworks. Among others, it follows from the above that the enforcement of one policy cannot well be described along top-down/bottom-up and task-specific/exclusive jurisdiction dimensions.²³ In case of crimes against the EU’s financial interests, for example, the EU firmly requires (top-down) that Member States take enforcement action, but the latter enjoy a great deal of (exclusive) jurisdiction on the way in which they enforce, while their action is complemented by a (task-specific) EU agency that provides investigatory information and supports enforcement by national authorities (the European Anti-Fraud Office). Therefore, in spite of the above qualifications and the variety within each category, I maintain the broad-brush typology of Member State, network and agency enforcement, because they serve well as analytical starting points for further investigation and because these categories underlie the scholarship feeding into the perspectives discussed below.²⁴ Hence, the following question arises: why do agencies have enforcement tasks for some policies, while networks of national authorities and/or the Member States have a role in the enforcement of others?

III. Institutional conditions: actors and policy instruments

Two existing frameworks provide insights in EU policymaking and implementation – the process whereby EU policy is put into practice²⁵ – more broadly. Wallace outlines how EU policies are made through five policy-making modes, whereby each mode differs in “roles and behaviour of the various key actors, in the approaches to policy dilemmas, and in the instruments intended to address them”.²⁶ For implementation specifically, Heidbreder specifies how the convergence of interests over a policy affects the authority of supranational bodies over the implementation, and outlines how ambiguity over the goals and means of policy implementation determines whether an implementing entity has full or only specific jurisdiction.²⁷ Whereas these frameworks are useful to understand actor preferences for implementation in the broader context of EU policy-making, this article builds upon them for an understanding of the dynamics specific to enforcement vis-à-vis (private organisations of) citizens. Enforcement is particularly different from implementation because it presupposes a regulatory policy instrument with unambiguous (behavioural) rules and direct contact with policy target actors in case of non-compliance. This article discusses how these (and other) characteristics necessitate a revised framework for the choice between enforcement regimes.

This section discusses the relevant institutional and policy context. Who are the key policy-making actors that delegate enforcement to the Member States, networks or agencies, and what constrains them? On the basis of EU law, I first introduce the (constellations of) key actors that choose between regimes. Thereafter, the section demonstrates how competences in the Treaty on the Functioning of the European Union (the TFEU)²⁸ can constrain policy-makers’ options for enforcement, as competence specificity varies considerably across policy areas. The last part discusses how the degree of ambiguity of EU policy instruments can also limit the range of available enforcement regimes.

1. What does the TFEU determine for enforcement? Key decision-making actors and competence

The Commission, the European Parliament and the Council of Ministers together delegate enforcement tasks to the above-mentioned regimes. Their respective influence differs per decision-making procedure: under co-decision (COD), the Commission has the sole right of proposing legislation, after which the Parliament and the Council can amend, approve or reject. The TFEU can also prescribe the consultation procedure (CNS), in which the Commission sometimes shares its right of proposal with the Council, and whereby the Parliament is merely consulted. Compared to COD, therefore, CNS leaves considerably less space for supranational preferences, while intergovernmental bargaining has a larger effect on the outcome. Part V.2 demonstrates how the latter can be of particular importance for enforcement in the EU.

However, these decision-making actors are bound by the existence of a legal basis (a competence) for the establishment of any enforcement regime. Such regimes are ruled out entirely if a basis is absent, whereas legal bases that do exist might circumscribe a choice beforehand. The necessity of legal bases connects to the doctrines of legality and the rule of law²⁹ – only the law constitutes (and constrains) public authority³⁰ – and to the principle of conferral: competences not attributed to the EU remain with the Member States.³¹ The TFEU provisions thus determine the ambit of EU governance:³² in what policy domains can the EU take action and what are the objectives? The collection of direct taxes for example, is outside the scope of EU authority, while competition policy is within.³³ Additionally, the competences can determine qualitative aspect of EU governance:³⁴ how are policy objectives achieved, what should a policy look like, and what forms of action (eg distributive, regulatory or enforcement) does it prescribe?

The specificity of TFEU competences varies and is moderated by case law. Several policy areas, eg internal market and air transport,³⁵ exhibit openly-worded (“flexibility”) provisions into which many forms of EU action can fit.³⁶ Moreover, CJEU case law in principle allows EU action if it aims to improve the functioning of the internal market – but achieves something else by effect³⁷ – and recognises that EU action, including standards on enforcement,³⁸ can be implied by a TFEU article. In my view, regimes for enforcement are justifiable through these openly-worded provisions,³⁹ especially if such action is a necessary step in effectuating existing policy instruments or their goals.⁴⁰ The TFEU is more specific only by way of exception: Article 127(6) of the TFEU allows powers for financial supervision to be attributed to the ECB, while for competition policy, the TFEU identifies the Commission (in cooperation with the national authorities) as primarily responsible for enforcement and provides for corresponding sanctioning powers (Article 103). One finds more detailed competences for the Area of Freedom, Security and Justice (AFSJ): the TFEU specifies the instruments and organisations that can be enacted.⁴¹ However, more often than not, an available TFEU competence for EU enforcement action does not specify how enforcement should be organised.

2. How do policy rules impact enforcement? Harmonisation of law

In addition, the unity of the rules to be enforced limits the range of available regimes. After all, enforcement presupposes an assessment of behaviour in the light of prohibitions or obligations, and involves a decision on enforcement action. However, a single conclusion about (non)compliance with EU law becomes problematic in the case of (multiple sets of) diverging substantive policy rules. Therefore, whereas conventional principal-agent theory predicts that high ambiguity results in delegation to an independent agent,⁴² this article maintains that a high degree of ambiguity of substantive policy rules hinders delegation of enforcement authority to an independent EU agency or a (strong) network of national enforcement authorities. After all, the parallel existence of multiple and diverging obligations, prohibitions or sanctions for the enforcement of a policy – usually stemming from the EU Member States’ laws – would render a judgment about (non)compliance, by one or a few enforcement authorities jointly, ambiguous. This ambiguity is reduced if the EU legislator decides to harmonise the laws of the Member States. The more harmonised a body of obligations, prohibitions and/or sanctions, the more likely becomes enforcement by an agency or a strong network.

Thus, the EU legal framework denotes and binds key actors in their choice between enforcement regimes. The TFEU provisions determine which actors may devise enforcement and can limit their options with varying degrees of detail. Yet, even if competence is present, only a harmonised body of substantive EU law allows delegation to an agency or strong network for enforcement. However, competence and harmonised substantive policies seem to be conditions only: they leave further actor preferences for regimes undetermined. Therefore, the limited determinacy of EU law necessitates extra-legal perspectives to explain the choice between enforcement regimes.

IV. Functional rationales

Functional rationales help us to understand the choice between regimes.⁴³ They indicate how key actors consider the utility of a policy option: what respective use do the Member States, networks and agencies have for enforcement? And, more generally, what is the function of enforcement within the EU? First, this section connects two distinct types of enforcement (functions) to EU policy-making modes distinguished by Wallace. Thereafter, the section builds upon the literature on European regulatory networks and EU agencies,⁴⁴ and proposes modifications to explain the delegation of enforcement tasks specifically.

1. What are the functions of enforcement in the EU? Two types

Enforcement in the EU has at least two broad functions, and one is enforcement of EU policies proper. Many EU-made policies employ legal rules as a key instrument to steer the behaviour of EU citizens and businesses,⁴⁵ whereby compliance with the rules is an implicit yet crucial part of achieving policy objectives. Rules are required both for combatting fraud with EU funds and typically for the EU’s many regulatory policies. Considering enforcement’s instrumental position to secure and effectuate these distributive and rules-based policies, I expect key actors’ positions to resemble the mode in which they devise those policies more broadly. Within both the regulatory and distributive policy mode, the Commission acts as agenda-setter and defender of policy objectives while the Member States’ governments take some (political or operational) responsibility to implement policies or secure subsidies from the EU purse.⁴⁶ Moreover, given enforcement’s position as the logical tailpiece of regulation,⁴⁷ functional rationales of necessity, effectiveness or efficiency (subsection IV.2) rather than political notions of identity and sovereignty (subsection V.1) are likely to prevail during decision-making.

The second function is EU cooperation for (problems with) national enforcement. Decision-making thereon demonstrates characteristics of intensive transgovernmentalist policy-making, in which national preferences are both more articulate and more diverse, given the politically sensitive issues it concerns.⁴⁸ Intra-EU cooperation for public security, for example, is primarily a Member State response to common pressures: integrated police and judicial cooperation following increased cross-border crime resulting from removed national border controls, as part of the (earlier EU policy of) free movement within common market.⁴⁹ As for other policies devised through an intensive transgovernmentalist mode, decision-making on cooperation for enforcement can be characterised by a dominant Council and a subordinate Commission and Parliament, constrained by detailed legal bases (see Part III above).⁵⁰ Hence, although functional benefits of cooperation for enforcement can equally be present, political rationales could prevail among the Member States. After all, the integration of state capacities for public order, national security or immigration control are amenable to a considerable amount of political saliency and sovereignty-driven electoral resistance.⁵¹

2. Which regime? Functional reasons for delegating enforcement tasks

The previous section indicated that functional reasons may underlie the integration of enforcement in the EU, in particular for the enforcement of the EU’s own policies. Hence, functional rationales may explain why policy-making actors adjudicate between the three regimes to delegate enforcement tasks to. Agencies, their networks and the Member State regimes all have a different utility to perform enforcement functions.⁵² This section discusses three rationalist models of delegation embedded in studies of European agencies and regulatory networks, that are most relevant in the context of enforcement in the EU.⁵³

First, delegation to European agencies – and networks of national authorities to some extent – results from a credible commitment to cooperation. Politicians generally delegate to independent agencies to isolate regulation from electoral or interest group pressure, and thus credibly bind themselves or each other to policy.⁵⁴ The same reasoning applies to enforcement in the EU: Member States may have difficulties in committing themselves to comply with EU law⁵⁵ and to enforce it on citizens and businesses, and/or have a lack of trust in other Member States doing so.⁵⁶ In the face of national (economic) interests, delegation by the EU legislator of enforcement tasks to EU agencies prevents potential lenient enforcement by (weak networks of) Member State authorities towards domestic companies or industries. Purely Member State enforcement, which requires transposition through the (national) political process,⁵⁷ can especially frustrate the enforcement of EU policies, given opposition by domestic interest groups.⁵⁸ EU agencies (and strong networks) therefore, are credible enforcers in the face of national interests which could undermine the effectiveness of EU policy.

A second rationale that could impact delegation of enforcement tasks is a regime’s reflection of the (geographical) heterogeneity among a policy’s target actors. Blauberger and Rittberger assert that European regulatory networks are chosen over agencies because the former possess the operational resources and ‘street-level expertise’ for regulation that requires ‘case-by-case implementation’. Those national authorities are thought to be in a better position to reflect the particularities of citizens and organisations in their respective jurisdictions.⁵⁹ This argument is most appropriate for enforcement, which involves more direct contact with target actors than any other stage of the policy cycle. Moreover, the logic can differentiate between enforcement by the Member States or by agencies as well: a large heterogeneity among (especially locally operating) target actors may best be reflected by leaving enforcement to the Member States, allowing adaptation to the national contexts.⁶⁰ The target actors of the Ambient Air Quality Directive, for example, are not specified and can thus include both natural persons and both small and large companies active in any sector.⁶¹ Conversely, only an EU agency has the required proximity to entities in a pan-European sector with a relatively small number of similar companies, such as credit rating agencies.

A third reason for delegation lies in a regime’s capacities to process enforcement information and reduce related transaction costs. Political actors have difficulties in gathering and processing all information relevant for determining a course of action, given similar interests.⁶² Delegation to independent agents can then have a twofold function when it comes to information: EU agencies and networks facilitate the exchange of information⁶³ and – depending on their strength – bring together relevant stakeholders for problem or even solution definition.⁶⁴ In the specific context of enforcement, this could entail that networks and agencies share information on (the harmful consequences of) target actor behaviour to facilitate investigations into non-compliance in cross-border or Europe-wide situations by respective national authorities. Moreover, and provided that substantive law is harmonised sufficiently, agencies or networks can foster a mutual agreement on violations of EU law. For example, national authorities comprising the CPC are not only obliged to share information on infringements of EU law and resulting damage, but also ought to agree on enforcement action in a case of non-compliance. On the other hand, a unitary decision on enforcement is best taken by a single strong agency.

Thus far, I have identified three functional considerations that impact the choice to vest enforcement responsibilities in Member States, an EU network of national authorities and/or an EU agency. These three regimes can reflect varying credible commitments to effective enforcement in the face of distinct national interests; varying needs to share information on non-compliant behaviour and facilitate agreement on sanctioning; and varying natures and sizes of target actor groups. Yet functional rationales alone may also be insufficient to understand the choice between enforcement rationales: if delegation was indeed functionality-driven, the EU legislator could simply delegate tasks to the Commission – an existing institution with considerable expertise that already prevents downwards regulatory competition among Member States.⁶⁵ In other words, functionality alone cannot explain the more nebulous and novel enforcement regimes under consideration (and outlined in section III). Therefore, additional political rationales are needed.

V. Political rationales

A given functional demand to integrate enforcement in the EU thus leaves questions of political supply unanswered. What are the costs and benefits of an enforcement regime in terms of resources and political identity? And how do the preferences of the Member States, the Parliament and the Commission interact in this respect? As discussed in Part IV.1, some political rationales may be particularly articulated when it concerns intra-EU cooperation for national enforcement. This section thus provides for a brief account of political preferences regarding enforcement, and outlines how the interaction among the key actors is crucial in the EU legislative decision-making process. Together with the institutional, policy-specific and functional rationales outlined above, their preferences determine the choice to delegate tasks to the Member States, networks or agencies for enforcement in the EU.

1. What are the costs of enforcement regimes? Resources and sovereignty

Enforcement is expensive. As with other core state powers, such as the military and the collection of taxes, enforcement is an essentially limited resource.⁶⁶ Whereas legal rules can last forever after enactment, their enforcement requires a continuous appropriation of limited public funds. Therefore, actors will try to render enforcement as efficient as possible, while arguing over the location of enforcement costs. When it comes to the choice between regimes, Member State enforcement is rather costly: every Member State has to devise an administrative and coercive apparatus to enforce EU laws, in order to remain in compliance with EU standards themselves.⁶⁷ Agencies, however – and networks to some extent – allow resources to be exchanged or bundled: if these regimes allow the pooling staff, training and expertise, or if they perform (information-gathering) functions that would otherwise be performed by a multitude of Member States’ authorities themselves, these regimes could result in lower total costs for enforcement. Moreover, their enactment may allow the Member States to redirect costs to the EU’s budgets. The more a proposed regime presents efficiencies for the Member States, the more they may be willing to support its establishment.

Additionally, the integration of enforcement can be costly in non-monetary terms. As the nation-state is traditionally connected to the provision of national security and the use of coercive power against its citizens, domestic electorates easily connect enforcement by public authorities to national community and self-determination.⁶⁸ Conversely, Genschel and Jachtenfuchs maintain that European integration of such enforcement is a salient topic within domestic politics and is receptive to political mobilisation against it,⁶⁹ but I would argue that this is particularly so when it concerns enforcement of policies made within a transgovernmentalist mode (see Part IV.1 above). When it comes to a choice between regimes then, sovereignty-driven electoral resistance against integrated enforcement in the EU may well correspond to the territorial divergence between an enforcement regime and the actors targeted by a policy. Domestic political groups will be less keen to accept enforcement by an EU agency or another Member State’s authority, if the targeted actors (such natural persons and (small) enterprises) behave locally and primarily within Member State boundaries. Conversely, they are more likely to accept a distant EU enforcer against actors that operate across Europe and benefit from cross-border movement.

2. Actor interaction and distributional conflict

Taken together, the choice between enforcement regimes depends on the collective position of the Member States and the preferences of the supranational institutions. First, as for intergovernmental bargaining generally, the preference of the Council depends on the relative division of preferences over the more and less powerful Member States and the relative exposure to external pressures. The supranational preferences on the other hand, can assumed to be stable: the Commission generally seeks an expansion of powers to achieve its objectives and thus opt for a high degree centralisation,⁷⁰ while Parliament concurs for reasons of voter popularity and because EU organisations allow for better accountability relations.⁷¹ The institutions act in their self-interest and prefer the most centralised regime of an agency.

Thus, Kelemen and Tarrant assert that the degree of distributional conflict among Member States is a key factor for the choice between regimes of interest⁷² – and probably particularly so when it concerns a choice between regimes for EU cooperation for (problems with) national enforcement (Part IV.1). In case of low distributional conflict (leaving more room for the Commission and Parliament preferences for centralisation), enforcement tasks are more likely to be attributed to an agency. Delegation to the Commission is unlikely, given its lack of expertise and more general hesitation by the Member States. Conversely, greater distributional conflict – eg where a sufficiently large group of states would incur high costs – decreases the chance of a strong agency and increases the likelihood of a network or weak agency (with little powers and autonomy), as it allows Member States to retain their influence, yet efficiently pool information resources and avert compliance costs.⁷³ The relationship between distributional conflict and Member State enforcement, however, is more ambivalent: although implementation costs increase, it provides more possibilities to wield political influence.

VI. Towards an analytical framework

The central question of this article is: what rationales inform the choice between Member State, network or agency for enforcement? On the basis of the available literature, I have mapped rationales from various institutional, functional and political angles and discussed their applicability to the delegation of enforcement tasks to any of the three regimes. The arguments presented above allow us to construct a framework to determine which rationales could play a role in the choice for enforcement regimes. The framework is visualised in Table 2.

Table 2. Rationales for three enforcement regimes in the EU

* More articulated when cooperation for enforcement is functional for policies resulting from transgovernmentalist policymaking mode

It follows from this framework that a Member State regime seems more likely with a low degree of substantive law harmonisation. Given its capabilities, it is also more likely to be chosen when the policy involves large and/or heterogenous group of actors that operate locally and when cross-border information-sharing on non-compliance is not (or is less) relevant. In addition, enforcement can be left to the Member States when political actors do not perceive distinctly domestic interests to be present or less likely to hamper effective enforcement in the long term, and/or when there is high electoral opposition against cooperation or high distributional conflict. The network regime, secondly, occupies middle ground. The delegation of enforcement functions to networks is more likely when the applicable obligations, prohibitions and sanctions are harmonised to some extent. A network is suited to target a mixed group of actors operating cross-border but within a limited geographic range, and has the ability to share information on (potential) violations, damage and corresponding enforcement action involving several Member States. Distinctly domestic interests could exist, but probably do not impede effective enforcement. Networks are more likely if there is some electoral opposition against or distributional conflict over enforcement cooperation. Agencies, lastly, can only be delegated strong enforcement powers if substantive EU law is harmonised to a large extent. Agencies are only capable of targeting a relatively small and/or homogenous group of actors that operate across the continent, and/or whose violations and damage require a single problem and solution definition. Also, agencies are likely to be preferred if distinct national interests can impede effective enforcement, and if electoral opposition and distributional concerns are limited. The latter two regimes facilitate cost reductions.

However, this (primarily rationalist) framework requires some nuancing in light of several contextual factors.⁷⁴ First, enforcement regimes can inherit structures of organisational predecessors, which may account for incremental change rather than a sudden establishment or overhaul of enforcement regimes. Regimes can inherit pre-existing structures vertically – whereby an enforcement network may well turn into an enforcement agency over time⁷⁵ – or horizontally, whereby the EU legislator stipulates enforcement tasks for a regime when that regime previously fulfilled little or no such functions at all.⁷⁶ As a second group of contextualising factors, crises and other contingent events may influence the timing of a (shift in) regime choice: the BSE or financial crises for example, correspond to the enactment of the European Food Safety Authority and the delegation of enforcement tasks to the ECB respectively.⁷⁷ Such events expose deficiencies in a current system, and may provide the impetus for reform.⁷⁸ Lastly, one should account for model or example regimes that are appreciated within the broader institutional environment, which may inform the EU legislator’s delegation preference. One mechanism through which the copying of regimes may occur, is where decision-makers are faced with uncertainty, and hence mimic another policy area’s regime, which they consider successful or legitimate.⁷⁹ I take these nuances into account throughout the illustration of the theoretical framework in the section hereafter.

VII. Empirical illustrations

To illustrate the utility of the theoretical framework developed above, this section provides for two examples. How can the EU legislator’s choice for a network and for the Member States (the CPC and the ECD for the enforcement of EU consumer protection law and EU environmental law respectively) be understood, using the insights above?

1. Consumer protection enforcement through a network

What regimes did the EU choose for the enforcement of its consumer protection policies? The legislator initially chose the Member States through a Directive on injunctions,⁸⁰ and later also established a network of national authorities. Since 2004, Regulation (EC) No 2006/2004 (the CPC Regulation) requires that national liaison offices coordinate enforcement with offices in other Member States.⁸¹ Coordination involves a mutual assistance system, allowing an enforcement body to request its counterpart in another Member State to enforce on its behalf.⁸² The latter can refuse, eg if it finds that no intra-community infringement has taken place.⁸³ In addition, the CPC Regulation required public authorities to have certain inspection powers and the ability to impose sanctions, either directly or through court proceedings. A real-time information database was set up as well.⁸⁴

Institutional conditions did little to structure the choice between enforcement regimes for consumer protection. First, Article 114 TFEU – the current legal basis for the CPC Regulation – is a widely-formulated provision for the harmonisation of national laws for the functioning of the internal market⁸⁵ and is indeed not predisposed towards any enforcement regime. The Commission chose Article 114 because it previously served as the legal basis for consumer protection rules in general,⁸⁶ for the earlier Directive on injunctions and for the enforcement of other internal market policies.⁸⁷ Moreover, substantive consumer protection law was harmonised considerably prior to the establishment of the CPC network. At the time of the Commission’s proposal, at least 14 instruments specified obligations and prohibitions for various aspects of business-to-consumer relations, including distant marketing⁸⁸ and prices indication.⁸⁹ Moreover, the Commission found the proposed increase in harmonisation of substantive rules logically induced revision of the rules on enforcement.⁹⁰

Secondly, actors indeed articulated the functional capabilities of the regime for the enforcement of consumer law as an EU policy proper. The ability to share information for the definition of enforcement problems and the proximity to target actors clearly informed the choice for a network. The Commission expected consumers and fraudulent traders to increase cross borders activity, due to the growth of online shopping, the single currency and English as a more common language.⁹¹ Additionally, it considered mutual agreement between the respective national authorities required for problem and solution formulation: the requested authority – in the state of the fraudulent trader – was best suited within its home jurisdiction and national culture to decide on enforcement action, while the requesting authority in the Member State of the harmed consumer to be best placed to understand and judge the harm suffered.⁹²

Consequently, political rationales were less prevalent during the decision-making process. The Member States did not voice concerns in terms of a loss of power,⁹³ although the network clearly facilitated the Commission’s preference for extension of its own influence. After all, the network would facilitate stronger Commission involvement and increased feedback from the Member States, and allow it to participate when an infringement affects more than two Member States.⁹⁴ The Council was divided only over the costs for the adaptation of national enforcement structures: the Netherlands, Luxembourg and Germany would have to substantially adapt their national laws.⁹⁵ Luxembourg and Germany thus emphasised the successful functioning of the existing systems of informal and judicial corporation, resulting in their abstention from the final vote.⁹⁶ Apart from national law adaptations, additional costs for national authorities arising from the network would be covered by the multi-annual EU budget for consumer policy.

In addition, the Commission occasionally referred to similar cooperation tools in other policy areas as a way to legitimise its preference for a network.⁹⁷ Isomorphism therefore seems to have played a role in the decision-making process, but it must be noted that EU agencies were a fashionable idea within that period of time as well⁹⁸ – and therefore, isomorphism is a problematic explanation for a preference for a network over an agency. In addition, the pre-existing structure of an informal network was present, but the inheritance of core attributes was fairly limited. The CPC involves a considerably stronger cooperation mechanism than the bi-annual informal conventions of its predecessor.

2. Environmental protection through the Member States

What regime did the EU choose for the enforcement of EU environmental policies? With the ECD, the EU legislator explicitly chose the Member States. An EU agency and various networks existed already, but without responsibilities for enforcement;⁹⁹ and although various EU regulatory instruments already contained clauses that vested some responsibility to enforce in the Member States,¹⁰⁰ the EU firmly reinforced the choice for Member State enforcement of environmental policy. The ECD enumerates various EU regulatory policies, the violation of which is to qualify as a criminal offence within the Member States – given that violation has been committed with a certain severity of conduct.¹⁰¹

Although key actors articulated functional arguments for increased enforcement and although they could have delegated enforcement tasks to the pre-existing network or agency structures, institutional and policy conditions seem to have limited the range of available regimes. The Commission indeed sought to increase compliance for more effective environmental protection,¹⁰² and the Member States’ governments likewise recognised the necessity and appropriateness of increased enforcement action.¹⁰³ However, Article 175 TFEU stipulated that “[w]ithout prejudice to certain measures of a Community nature, the Member States shall finance and implement the policy devised” (emphasis added). Moreover, in terms of policy ambiguity, the sanctioning rules chosen for the ECD were too divergent to facilitate a more centralised enforcement regime. Although some EU policy instruments, eg on the shipment of waste¹⁰⁴ or the protection of species of wild fauna,¹⁰⁵ specify prohibitions and obligations for protection of the environment, the diverging criminal law rules envisaged for sanctioning would hamper a single or joint decision on appropriate enforcement action. As the Commission recognised, the existence of discrepancies in the definition of environmental crimes and sanctions would cause problems when it came to cooperation. Criminal law in Europe was (and remains) hardly harmonised, despite some effort by the ECD itself.¹⁰⁶

Both empirical illustrations thus indicate that institutional and policy-specific considerations can – but need not – limit the EU legislator’s functionally informed choice between enforcement regimes. Given that decision-making revolved around the enforcement of an EU policy proper, functional and not political rationales informed the preferences of the Commission and the Council. Although these preferences resulted in a network for the enforcement of EU consumer policy, the enforcement of EU environmental policy was left to the Member States, for the lack of a substantially harmonised body of rules on sanctioning. I have, therefore, illustrated the potential empirical utility of understanding the EU legislator’s choices for vesting enforcement tasks in the Member States and a network of national enforcement authorities.

VIII. Conclusion

The central question formulated at the outset of this article was: what rationales determine the EU legislator’s choice for the Member States, EU networks of national authorities and/or an EU agency for the enforcement of EU policies? Building upon earlier frameworks of EU policymaking and implementation, this article made a start by connecting various insights from law and politics to understand enforcement in the EU. I outlined how EU law establishes the Commission, the Parliament and the Member States as decision-making actors on the one hand, and how it can readily influence their range of available enforcement regimes on the other. In turn, the article specified the functional and political logics that allow us to explain why the EU is increasingly involved in the enforcement of its policies and which regime is most likely to perform a particular enforcement function under particular circumstances. Lastly, I illustrated the framework’s potential usefulness by means of an example of the CPC network and ECD.

The combination of (legal and political) insights provides us with a richer understanding of the choice to delegate enforcement tasks to the Member States, networks of national authorities or EU agencies. It also has potential for a normative assessment of a regime in light of rationales that preceded its enactment: should the EU enact a network or agency to bring stakeholders together for unitary problem definition? And is the risk of downwards competition among Member State authorities high enough to warrant an enforcement agency? This very knowledge about when to choose which regime for the enforcement of EU policies can then feed into the legislative process itself: it provides policy-makers with the opportunity to adapt their regime choice to the relevant contextual factors of a policy and its environment, as provided for by this framework.

Nonetheless, the above needs refinement and expansion through application. To what extent can the above rationales indeed by identified in the legislative processes that led to (the delegation of enforcement tasks to) regimes? And when comparing multiple regimes, can we identify which rationales are necessary and sufficient conditions for a particular enforcement regime? The empirical illustrations of consumer and environmental protection alone do not teach us about the establishment of an agency or the attribution of enforcement tasks thereto. The EU legislator did not establish an agency, though distributional conflict between the Member States over the centralisation of enforcement was low. Despite these further questions, the framework developed here is a useful starting point for explaining various regimes for enforcement in the EU.