Cross-border mergers in the EU: the Commission versus the Member States
by
Michael HARKER
Michael Harker, ESRC Centre for Competition Policy, NorwichLawSchool, University of East Anglia, Norwich. NR1 7TJ. UK.
Paper presented to the ESRC Centre for Competition Policy Summer Conference:
Comparative Perspectives on Multi-Jurisdictional Antitrust Enforcement
Norwich, UEA, 14 - 15 June 2007
[Preliminary Draft – please do not cite without permission]
ABSTRACT: This paper reviews and analyses cases where the Commission has taken action against Member States where the latter have either acted contrary to the exclusivity principle under the EC merger control regime, or have maintained in place ex ante measures with the purpose of seeking to control or deter cross-border investment in key strategic industries. As has been confirmed on several occasions by the European Court Justice (ECJ), subject to very limited circumstances, such measures are contrary to the free movement of capital provisions of the EC Treaty. Nevertheless, on several notable occasions, Member States have been willing to use such powers, seemingly judging the political costs of standing by and permitting the merger to outweigh the (future) costs associated with infringement proceedings before the ECJ. In so doing, Member States have been able to modify transactions significantly and even frustrate them. Such cases pose serious problems for the stability of the EC merger regime, and have implications for the possibilities of expanding Community competence in this area.
1. Introduction
Jurisdiction under the EC Merger Regulation (ECMR)[1] is allocated as between MemberStates and the Commission by reference to quantitative criteria, based on both the world wide and Community aggregate turnover thresholds of the merging parties.[2] Where these thresholds are met, the concentration has a “Community dimension” and consequently falls within the exclusive competence of the Commission. To this end, Article 21 ECMR provides that for such concentrations, the Commission has sole competenceand that Member States shall not apply their national competition laws in respect of the same (the “exclusivity principle”).[3] Conversely, the Commission has no competence under the ECMR where the concentration does not have a Community dimension.[4]
This “bright line” allocation of jurisdiction is subject to some notable qualification and exceptions: even where the aggregate turnover thresholds are met, the concentration will not have a Community dimension where two thirds of the aggregate Community turnover of each of the undertakings concerned occurs in “one and the same Member State” (the “two thirds rule”).[5] There also exists mechanism for the commutation of cases as between the Commission and the MemberStates, although these will not be discussed here.[6] The two thirds rule is one corrective mechanism for the crudeness of turnover thresholds. Subject to certain limitations, Member States may also “take appropriate measures” to protect their legitimate interests, notwithstanding the Commission’s exclusive competence (the “legitimate interests clause”).[7]
At first sight, these somewhat mechanical rules appear uncontroversial. This is not so. The exclusivity principle and the legitimate interest clause have proven fertile ground for disputes and litigation between the Commission and the MemberStates. These cases are discussed fully in this paper. Furthermore, the Commission has consistently taken action against Member States who maintain in force ex ante measures under which they seek to control trans-border investment in key strategic industries (the so-called “golden share” cases). The link between these cases and the merger regime is clear: to the extent that such rules deter trans-border investment or acquisitions within the Community, their existence undermines one of central objectives underpinning the ECMR and the Treaty: the emergence of European scale firms operating on a single market, undistorted by protectionist measures at national level. These cases, which have fallen to be decided under the free movement of capital principles of the Treaty, are discussed in full in this paper. It is clear from the case law that the European Court of Justice (ECJ) has taken an expansionist approach in defining the extent of the right of free movement of capital and what amounts to a restriction, while at the same time taking a very narrow view to both what can amount to a justified restriction (in terms of the objective pursued and the permissible extent of the restriction). In this regard, it is possible to draw parallels between this and the role of the Court in securing the adoption of the ECMR by Member States, in the face of consistent opposition by some of the latter. This is addressed in the next section, which sketches briefly the historical background to the ECMR.
There is a central riddle posed by all of this. Drawing upon some of the insights provided by agency theory, which broadly speaking assumes that where delegation takes place as between the principals (in this context, the Member States) and the agent (the Commission), it is because it is seen by the former to be “functional” (i.e., in their interests). In conceptualising the relationship between the Commission and the MemberStates in this way, emphasis would normally be given to control and oversight mechanisms put in place by the principal(s) to prevent shirking or slippage on the part of the agent. Yet as is demonstrated by the case studies in this paper, the direction of control tends to be in the opposite direction. This point is retuned to in conclusion.
This paper is in four sections. The following section – inspired by agency theory – briefly sketches the possible functional logic(s) of delegation in the context of trans-border merger control. In order to more fully understand the coinciding (and sometimes competing) interests of the MemberStates and the Commission, the historical background to the ECMR is also outlined. Section 3 contains an analysis of the case law: the legitimate interest cases, the exclusivity cases, and the golden share cases. Section 4 concludes.
2 Why delegate in the context of trans-border merger control?
In this section, we first identify the possible functional logic(s) of delegation to a supra-national institution in the context of trans-border merger control. In order to contextualise further these possible explanations for delegation under the ECMR, the section then goes on to briefly outline the historical background to the ECMR.
2.1 The functional logic(s) of delegation[8]
Principal-agent models are used to identify and illuminate the phenomena of hidden information and hidden action with the hope that, in so doing, it will become possible to design mechanisms of control which will better align the actions of the agent with the preferences of the principal. Originally an emanation of organisational economics, one of the claimed benefits of the principal-agent approach is its normative neutrality: it can operate in competing normative contexts because it is merely a tool of positive analysis.[9] For that reason, among others, the principal-agent model has been adapted and applied by political scientists to the problems of delegation within and as between public bureaucracies, as between states and supranational institutions, and as between the population generally and their democratic institutions.[10] In the political science context, most attention has been focused on the problems of agency shirking and slippage which are broadly speaking the equivalents of the moral hazard problem.[11]
There are a number of assumptions underlying some principal-agent analyses which may be less straightforward than appearances would suggest. The first assumption is that in delegating power to an agent the principal must consider it to be in her interests to do so, i.e., the potential benefits of delegation outweigh the anticipated agency costs (the costs of control and agency “drift”).[12] However, while a functionalist logic of delegation may seem attractive, it is too simplistic to say that there is always an automatic link between this and the decision to delegate by a principal or principals. Other contextual factors may be at play, for example, policy learning and institutional isomorphism may perform as important a role in a decision to delegate.[13] The second common underlying assumption behind some principal-agent analyses is that it is necessarily desirable, from the point of view of the relevant principal(s), for there always to be a coincidence between the actions of the agent and the policy preferences of the principal. As Majone points out, a functionalist logic of delegation needs to take full account of the importance from the principal’s point of view of delegation being a means of enhancing the credibility policy commitments.[14] The level of independence or autonomy ceded to an agency will, according to Majone, increase according to the seriousness of the credibility problem. Concerns over credibility of commitment arise from the problem of time-inconsistency where there is the potential that the legislature’s and/or the executive’s optimal long-run policy may conflict with its short-term policy and, to that extent, it has an incentive to renege on its long-term commitments.[15]
Turning to the specific context of the EC merger control, there are several possible (and well rehearsed) justifications – from the point of view of the Member States – for both the decision to delegate and for the method of delegation. First, there are the firm-centric and efficiency based arguments. The ECMR adopts a “bright-line” approach whereby jurisdiction is allocated as between the Commission and National Competition Authorities (NCAs) according to turnover thresholds – the “Community dimension”. This is an expression of the principle of legal certainty, whereby merging firms can predict under whose jurisdiction their merger may fall to be scrutinised. The “one-stop shop” principle is also realised by delegation from multiple principals: mergers with a Community dimension, which might otherwise involve multiple filings within the EU, can be notified to one supranational authority having sole competence. Furthermore, such delegation facilitates the application of a unified substantive norm to mergers having a Community dimension, resulting in a “level playing field” across the single market. The Community Courts and the Commission develop a unified jurisprudence on mergers, clarifying the principles underlying the substantive analyses of mergers, with the consequence that firms will not be deterred from notifying efficiency enhancing mergers. Secondly, there are commitment justifications where delegation is a means of achieving a long-term policy goal where the short-term policy preferences of Member States may be misaligned. Member States who might otherwise engage in strategic behaviour (for example, by promoting the interests of domestic firms at the expense of foreign firms) can, by delegation to a supranational institution, commit to a strictly effects based approach to mergers. Thirdly, by delegating to a central authority, it is possible to apply welfare standards in a dynamic setting, therebyincreasing the long-term welfare of Member States (both individually and collectively) by reducing the potential for the blocking of efficiency enhancing mergers. To illustrate this point, consider the counter-factual to delegation, with a merger which engages the merger laws in several Member States, resulting in increased welfare in all but one. Absent delegation, each MemberState has a veto,[16] even though the merger would result in aggregate welfare increases. Fourthly, there are institutional arguments, such as the need to develop capacity (expertise)which arguably can be done more quickly and efficiently a supranational institution.
If these arguments are so convincing, then why is there a bifurcation of authority at all? There are some justification. Some mergers do not have trans-border effects and NCAs may have informational advantages over a supranational institution.[17] Ultimately, Member States may well be unwilling to give up completely an important “lever” over domestic industrial policy.[18]
2.2 The historical background to the ECMR[19]
This section reviews briefly the difficult and tortuous journey which resulted in the adoption of the ECMR. As is demonstrated, the jurisdictional rules proved to be some of the most controversial and difficult aspects of the ECMR. In this context, the role of the ECJ should not be overlooked, as it proved pivotal in bringing the Member States to the negotiating table and eventually securing their agreement.
2.2.1 The negotiations
The ECMR was eventually adopted in 1989, coming into force in 1990. However, this was the result of a long negotiation process. The first proposal came from the Commission in 1973, but received little support from Member States. There followed further proposals in 1981, 1984, 1986 and 1988.[20] The divergence of views, as between both Member States and Member States and the Commission, centred on three main areas: the level of the jurisdictional thresholds; the substantive criteria, and the ability of Member States to derogate from the merger regime in the public interests; and the strength of the mechanisms by which Member States could oversee and influence the Commission’s decisions.
In respect of the level of the thresholds, the Commission was eventually able to secure agreement, but with the world wide threshold set at some 25 times higher than had originally been proposed in 1973, and five times higher than was proposed in 1988. The centre of gravity principle was first given expression in the 1988 proposal, but set at a higher level, requiring that the Community dimension would not exist where 75 per cent of the Community turnover of the parties occurred in one and the same MemberState. This was the forerunner to the two thirds rule eventually adopted. Both of these changes were secured at the insistence of Germany and the UK.
The substantive test – and the extent to which public interests should be taken into account by the Commission or asserted by Member States –proved highly controversial. On the one hand, France and Italy (later joined by Spain and Portugal), argued strongly in favour of public interest derogations, while Germany in particular resisted strongly any deviation from “pure competition” criteria (reflecting its domestic merger legislation). The UK initially strongly supported the former position, but then joined Germany, fearful that the watering down of an effects-based test might result in the Commission pursuing industrial policy goals at a Community level. Agreement came in 1989 when one of the main protagonists – France – became President of the Council, with one its stated policy objectives being the reaching of an agreement on the merger regulation. In order to reach agreement, it softened its requirement for public interest derogations, accepting the more limited “legitimate interests” clause now contained in Article 21(4) ECMR.
While the Commission compromised on a number of measures, it remained steadfast in its insistence that Member States would not have the ability to shape, veto or delay its decisions beyond the weakest form of comitology, the Advisory Committee.[21]
2.2.2 Merger control by stealth: the role of the ECJ
There were two key judgments of the ECJ which were vital in bringing about the adoption of the ECMR. The first case, decided in 1973 was Continental Can v Commission.[22] Although the Commission lost, the ECJ confirmed the Commission’s view that Article 82 can be used to prohibit a firm from acquiring a competitor thereby strengthening its dominant position. Had the Court adopted the then prevailing (narrower) view of the application of Article 82 – which implied a causal link between abuse and dominance – then it would have prevented the Commission from applying Article 82 to merger situations. Article 82 did not, however, offer the basis for a comprehensive merger regime.[23] In the next case, decided in 1987, the ECJ held that Article 81 was capable of applying to mergers, again despite the prevailing view – based on a literal interpretation of the prohibition – that it could not.[24] Uncertainty over the scope of the Commission’s competence over mergers – and in the face of an acquiescent Court – led to increasing pressure from business on governments to agree a deal at Community level.
3. Facilitating cross-border mergers: the Commission’s enforcement strategies
In this section we consider the case law where the Member States have asserted their interests and/or implemented measures in respect of mergers with a Community dimension.
3.1 Legitimate Interest cases
As was discussed in section 1, Article 21 contains the exclusivity principle, which is subject to the exception that Member States may intervene in order to protect their legitimate interests. Article 21(4) provides:[25]
“…Member States may take appropriate measures to protect legitimate interests other than those taken into consideration by this Regulation and compatible with the general principles and other provisions of Community law. Public security, plurality of the media and prudential rules shall be regarded as legitimate interests… . Any other public interest must be communicated to the Commission by the MemberState concerned and shall be recognised by the Commission after an assessment of its compatibility with the general principles and other provisions of Community law before the measures referred to above may be taken. The Commission shall inform the MemberState concerned of its decision within 25 working days of that communication.”