The European Commission's Ability to Act under European Competition Law: The Example of Telecommunications and Electricity Policy

Abstract

European competition law allocates far-reaching competences to the European Commission. The paper asks for the conditions under which the Commission may use these rights against the member states, focusing on the most powerful provision - the right of the Commission under Article 90 to issue directives by itself in those cases where member-state governments have allocated specific rights to undertakings that conflict with the Treaty's rules. In addition the Commission may pursue Treaty violations on a case-by-case basis. In European telecommunications policy the Commission has used its powers rather successfully, with all liberalization decisions being based on Article 90. But for European electricity policy the Commission has shrunk away from using these powers in favor of initiating council legislation. The paper analyzes the conditions of the Commission's ability to act under European competition law in a multi-level framework, drawing among others on a principal-agent approach.

Susanne K. Schmidt, Max-Planck-Institut für Gesellschaftsforschung, Köln

Abstract: European competition law allocates far-reaching competences to the European Commission. The paper asks for the conditions under which the Commission may use these rights against the member states, focusing on the most poewrful provision - the right of the Commission under Article 90 to issue directives by itself in those case where member-state governments have allocated specific rights to undertakings that conflict with the Treaty's rules. In addition the Commission may pursue Treaty violations on a case-by-case basis. In European telecommunications policy the Commission has used its powers rather successfully, with all liberalization decisions being based on Article 90. But for European electricity policy the Commission has shrunk away from using these powers in favour of initiating Council legislation. The paper analyzes the conditions of the Commission's ability to act under European competition law in a multi-level framework, drawing among others on a principal-agent approach.

Introduction

European competition law allocates far-reaching rights to the European Commission. In particular state measures of the member-state governments, in the form of financial aids or the granting of special rights, can be overseen and regulated by the Commission quite independently. These rights are a major reason for the often criticized bias of European integration in favour of negative integration at the expense of positive measures. Commission action under these rights is blamed to lead to serious constraints on member states' ability to act that add to the already declining possibilities of state action under the pressures of economic interdependence and globalization (Scharpf 1996).

In this paper I aim to analyze the conditions under which the Commission may apply its rights under competition law. For this I focus on a particular aspect of these rights - the Commission's possibility to intervene against the granting of special rights by member states to "undertakings", which gives the Commission the unususal possibility under Art. 90.3 of the Treaty to issue its own directives aimed at all member states, without any formal decision rights of other bodies, including the Council of Ministers. This prerogative is complemented by case-specific possibilities of the Commission to act, for instance via decisions under Art. 90.3 or through infringement procedures under Art. 169, which will also be analyzed. European telecommunications policy is the sector which has so far been shaped most by these competences of the Commission. In fact, all the liberalization decisions on the European level, which have been transforming European telecommunications systems from integrated monopolies in the late 1980s to full liberalization in 1998 have been taken unilaterally by the Commission. Of course, this impressive record of Commission action may be very case-specific to telecommunications and with little impact on other areas where governments have allocated special rights. But it may also be that the beneficial conditions in telecommunications could be used by the Commission to set a precedent. European electricity policy will be taken as an example where the Commission has encountered much more difficulties. Taken together the two cases allow to draw up some conclusions on the Commission's ability to act under European competition law.

In the following, I expand shortly on the logic of delegation and summarize major rights entrusted to the Commission. Secondly, I present an overview of the policy developments in the two sectors. Having developed a framework for analysis I will subsequently examine the conditions under which the Commission is able to act under European competition law.

The Delegation of Competences to the Commission

For an analysis of the Commission's ability to act it is useful to begin with the underlying rationale of member-state governments to delegate competences to the European Commission. For this a principal-agent perspective has been advocated by several scholars (Majone 1996: Moravcsik 1995; Pollack 1995a; Pierson/ Leibfried 1995: 434). "The principal-agent model is an analytic expression of the agency relationship, in which one party, the principal, considers entering into a contractual agreement with another, the agent, in the expectation that the agent will subsequently choose actions that produce outcomes desired by the principal." (Moe 1984: 756).

Several reasons make the delegation of rights to the Commission for the national governments desirable (Pollack 1995a). Roughly, they can be seen to root in advantages for the solution of prisoners' dilemmas and in the need to neutralize the possible influence of agenda-seeting rights. More specifically, four advantages of delegation can be singled out. Firstly, the commitment to cooperate among the actors can be strengthened. Through the delegation of monitoring rights, incentives to free-ride can be tamed. A neutral agency can improve the assurance of all participants that everyone conforms to the agreements, which may be particularly relevant for the implementation of regulatory policies with its intransparent character and scope for discretion (Majone 1992: 150). Depending on the degree of trust among the member states, as Majone (1995) has stressed, further rights may be delegated: rights to sanction or to take regulatory decisions can ensure equal treatment among the partners.

Secondly, the delegation of regulatory rights strengthens the commitment vis-à-vis third parties, and avoids, for instance, the danger to fall prey to "regulatory capture" (Stigler 1971). Efficiency is also raised because the political costs of regulatory decisions can be alleviated. "By charging an agency with the implementation of a general regulatory mandate, legislators not only avoid the time and trouble of making specific decisions, they avoid or at least disguise their responsibility for the consequences of the decisions ultimately made." (Fiorina 1982: 47).

Thirdly, delegation may ease cooperation and commitment among the partners as well as with regard to third parties over time. This concerns, on the one hand, the problem of incomplete contracts (Williamson 1985: 178), which may be eased through the definition of appropriate procedures to reach agreement, and through the delegation of decision-making competencies to an independent third-party. On the other hand the problem of "time inconsistency" arises (Kydland/ Prescott 1977), which refers to the anticipation of preference changes "Without a binding commitment holding them to the original plan, governments will user their discretion to switch to what now appears to be a better policy. The problem is that if people anticipate such a policy change, they will behave in ways which prevent policy makers achieving their original objectives. "(Majone 1996: 6). Finally and fourthly, the delegation of agenda-setting rights may ease future cooperation. Independent expertise may be used and the potential to influence decisions through agenda setting may be neutralized, which is particularly important under forms of majority decision rules (Garrett/ Tsebelis 1996).

The advantages of delegation, it is quite apparent, thereby come at a cost: "Once an agency is created, the political world becomes a different place. Agency bureaucrats are now political actors in their own right: they have career and institutional interests that may not be entirely congruent with their formal missions, and they have powerful resources - expertise and delegated authority - that might be employed toward these "selfish" ends. They are new players whose interests and resources alter the political game." (Moe 1990: 143). The danger of "agency loss" may be countered through different means of control placed on the agent by the principals, which, however, are not costless either. Benefits of delegation thus have to be weighed against their cost. "Hence, a given control mechanism will be adopted only if its cost is less than the sum of the agency losses that it reduces. By the same token, principals will delegate authorities to an agent only if the benefits of such a delegation are greater than the sum of the agency losses and agency costs." (Pollack 1995a: 16). From a principal-agent perspective it can be concluded that the Commission's scope for independent action partly corresponds to the longer term interests of the member-state governments in cooperation - and, therefore, to their interests in delegation - but partly it is used to follow self-interests.

Many rights held by the Commission, as well as by the European Court of Justice (ECJ), can be explained on the basis of a principal-agent perspective. Thus, the Commission has the sole right to initiate legislation. The delegation of this right to an independent body may save transaction costs, allow for specialization, and neutralize the bias inherent in the right to make proposals. At the same time, however, the Commission can act as a gatekeeper to Community legislation and influence the proposals (Héritier 1996). With the increasing relevance of majority voting in the Council, the Commission's agenda-setting powers - including its ability to realize its own interests - have been strengthened, due to another procedural provision. Changes to the Commission's proposals require a unanimous decision of the Council, making it then easier for the Council to accept the Commission's proposal than to agree on another option. The Commission, on its part, can pick among different available proposals that are backed by a majority of member-state governments the one it likes best. Next to these formal agenda-setting rights, the Commission can use its central position and resources to strengthen its informal agenda-setting role, for instance by initiating discussions on relevant topics, possibly based on the publication of a Green Paper.

The Commission's rights under European competition law eases the different problems of its principals to make binding commitments to cooperation. The Commission, on the one hand, has rights with view to private enterprises. Art. 85 prohibits cartels and Art. 86 the abuse of a dominant position. Complementing these two articles is Council Regulation No. 17, agreed on in 1962, which gives the Commission quite far-reaching executive powers for the implementation of Art. 85 and 86 (Graupner 1973). Also the Merger Regulation has to be mentioned in this context, which was agreed by the Council in 1989 and broadens the Commission's competence vis-àa-vis private enterprises to regulate mergers above a certain threshold that impact on the Community market (Bulmer 1994). The independent competences of the Commission in the enactment of Art. 85 and Art. 86 - the member-state governments only participate in the decision-making in an advisory committee - find a rationale in the interest in transaction-cost savings, in avoiding inconsistent decisions, and in having a competition authority with a mandate for the whole internal market, which lies outside the reach of national authorities. Beyond this functionalist rationale significant German preferences from the conclusion of the Rome Treaty up until today have helped a strong European competition regime (Bayliss/ El-Agraa 1990).

On the other hand, the competition rules of the Treaty take issue with the actions of the member-state governments. Art. 92 and Art. 93 concern the control of state aids. Art. 90 deals with the granting of special rights to undertakings which conflict with the rules of the Treaty. In these cases the Commission has the right to independently direct decisions or even generally binding directives to the member-state governments. Given that these are exceptional rights the latter will be my focus in this paper. With regard to state action, the rationale of delegation is particularly apparent. While for the overview of private activities transaction-cost savings may dominate interests in delegation next to concerns with equality, in the overview of state action delegation may be already required to establish credible commitments. Otherwise, the concerned governments could simply evade sanctions via promises of "turn-taking". For the EEC Treaty, historically also considerations of fairness and trust have played a role. The traditionally very different sizes of the public sector among the founding members were a reason to define Art. 90 and to allocate specific powers to the Commission. Without a Treaty provision there was the fear of the Benelux countries and of Germany, that Italy and France could gain unfair competitive advantages by granting special rights to public firms.

Delegated rights give scope to the agent to pursue his own interests rather than those of his principals. In how far the Commission's ability to act is strictly or only more loosely constrained by its principals will be examined on the basis of an analysis of European telecommunications and electricity policy. Before engaging in the empirical details of these two cases, the principal-agent perspective will serve as a background to assess the scope of the Commission's ability to act further.

The Commission's Ability to Act - A Framework of Analysis

On the basis of the principal-agent perspective, three major structural reasons can be singled out for agency loss. First of all, there is a structural differences between bureaucracies and politicians in their capacity for long-term planning (Moe 1990: 124; Pierson 1996). The fact that politicans may discount future disadvantages highly in order to obtain momentary benefits can give the Commission scope to realize its own interests. Secondly, the neutral agent may profit from information asymmetries (McCubbins et al. 1987: 247). For instance, the analysis of Héritier and collaborators of European air quality policy revealed that the Commission could use its privileged position based on its oversight on the situation in all member states to influence decision-making in its favour (Héritier 1993: 441). Thirdly, the agent can exploit differences in interests among multiple principals. "Assuming some preference heterogeneity, an agency in a multiple-veto-point system has a number of possible protectors: [...] for any change in agency policy that one veto-point might demand, there is always some other veto-point which prefers the status quo (or even prefers a move in the opposite direction)." (Hammond 1996: 143). There are two sides to this possibility. Either the agent can use a lack of "policy clearness" in documents to further its own interests based on the inability of principals to draw up a clear agreement in view of their divergent interests. Or, the agent may diverge from clearly delineated political compromises, making sure that the favours the interests of at least one principal (Torenvlied 1996: 27). In the European Union where the unanimous consent of member-state governments is required to change the Treaty for a strong sanction of the Commission, this gives the Commission considerable room for manoeuvre with now 15 member states. However, in view of the long-term nature of their cooperation it could be that the member-state governments realize this danger and block this possibility to gain, knowing that in the long run only the Commission is to profit if the governments do not solve their cooperation problem (Pollack 1995a). The case is very different if the governments change the default condition for the Commission (Ostrom 1986), as they have done with the 1988 structural fonds. Instead of having the programme as the status quo, its renewal in 1993required unanimous approval of the governments. This put the Commission at a serious disadvantage as it had to conform to the interests of all its principals in order to get their consent (Pollack 1995b). ₁

For the ability of the Commission to act its agency relationship to the principals is only one - highly important - consideration. Operating in a multi-level system of governance (Scharpf 1994), also the Commission's relationship to sub-national and non-state actors are relevant. This category includes diverse actors such as operators and manufacturers, sub-national bodies as well as European associations. Despite this heterogeneity the central characteristic of these actors is that they stand in a relationship of dependence with the member-state governments, for which most of them are, in turn, relevant principals. Several studies have shown that support of these actors can be crucial for the Commission to overcome the resistance of member-state governments to its proposals (Marks 1993; Sandholtz/ Zysman 1989). Thus, in these cases the national governments as agents are put under pressure by some of their principals, who have recognized parallel interests with the Commission. In sum, then, for its ability to act it is useful to see the Commission as embedded in a vertical relationship with the national governments as well as with private actors, which implies constraints as well as opportunities for its action.

This vertical relationship is only one side of the framework under which the Commission operates. Similarly important is its embeddedness in a horizontal dimension. Firstly, this concerns its direct relationship to the other European institutions. In part, the European Parliament and the European Court of Justice share the interest with the Commission in the strengthening of European integration. In part, they have also conflicting interests which is particularly apparent in the case of the European Parliament whose interests, to give an example, in a strengthened agenda-setting role implies direct losses for the Commission. The interests of the ECJ and the Commission are much more compatible. Nevertheless, the Commission cannot take it for granted to receive the support from the ECJ on which it crucially relies in most conflicts with its principals as well as with third parties. The ECJ enjoys greater independence and therefore leeway from the member-state governments than the Commission because of the tradition of an independent judiciary (Burley/ Mattli 1993). Though it should therefore in principle feel free to support the Commission's actions, it may at times come to a different assessment of its long-term interests in the development of the European legal order and its own legitimacy. But the ECJ is not only a constraint placed on the Commission. Its judgements often serve as "windows of opportunity" (Kingdon 1984) for the Commission. The new competences awarded to the Commission in the merger regulation in 1989, for instance, can be seen to root in the ECJ's ruling in the Philip Morris case, which opened such a window for the Commission, upon which it successfully acted (Bulmer 1994). Importantly, through such sequential action the Commission may benefit from the greater independence enjoyed by the ECJ.

The relationship of the Commission to the European Council and the Council of Ministers is characterized by the fact that for the Commission these bodies assemble representatives of its principals. The responses of these bodies therefore give indications on the Commission's appropriate behaviour and on possible sanctions. Most of all this is true of the European Council as the heads of governments are responsible for Treaty amendments and thus ultimately for the constitution of the Commission. For the Council of Ministers this only holds partly. Here in many cases, sectoral interests of ministers combine and at times overtake the aggregated national interests, depending on the degree of ministerial discretion. Several studies have pointed out that decisions in the Council of Ministers can serve to overcome existing national constraints, for instance by bypassing vetoes of other departments. This implies for the Commission (and, partly for the ECJ) - as well as for the analyst - that deliberations and decisions in the Council are not always representative of the principals. At times support that is promised on this level may be withdrawn subsequently by the government. However, such open conflicts may be expected to be rare as an interest can be assumed on the part of the national governments to absorb conflicts within the principal internally, keeping the external position consistent and accountable. Existing sectoral self-interests in the Council therefore can be generally seen as an opportunity for the Commission rather than as a constraint.

In a multi-level system of governance also the horizontal levels below the Commission can be expected to be relevant for its actions. The relations among the member-state governments but also among private actors may open opportunities for the Commission, or imply significant constraints. For instance, whenever cooperation problems (prisoners' dilemmas) arise among the national governments the Commission can strengthen its role by presenting solutions, offering the monitoring of agreements or the enactment of regulatory decisions; a scope for action which was already referred to when summarizing the rationale for delegation. Given that solutions to coordination problems are self-enforcing and require no monitoring, these offer less scope for supranational action, but here the Commission may point out possible solutions (cf Martin 1992). Its possibilities are quite parallel with view to coordination and cooperation problems of private actors instead of governments. However, next to presenting opportunities the interdependences and relations among private actors as well as governments may also pose constraints on the Commission for which cartel-like cooperations or veto coalitions are examples.

Figure I summarizes the framework of analysis

European telecommunications and electricity policy

The first major step to a European telecommunications policy was taken when the European Commission published its "Green Paper on the development of the common market for telecommunications services and equipment" in 1987, in which it called for a common reform of the regulatory framework. An important background to this document was that the ECJ had established in 1985 in the British Telecom case that the provision of telecommunications services was subject to the competition rules of the EC Treaty and had to be regarded as a normal economic sector with no exceptional status (Ellger 1992). The document was followed up by a broad consultation process of relevant organisations in the field and in February 1988 the Commission published ambitious proposals for the implementation of its measure. Even before the governments had endorsed these plans through a Council resolution in June 1988, the Commission issued a directive for the liberalization of terminal equipment based on Art. 90.3 of the Treaty of Rome. Thereby it employed a legal provision which has been used only once before.

To sanction what they saw as an illegitimate expansion of the Commission's powers, France supported by Italy, Belgium, Germany, and Greece called upon the ECJ. Though the liberalization of terminal equipment was not contentious among the member-state governments, they argued that only the Council could enact such a measure under Art. 100A of the Treaty. Notwithstanding this legal conflict, the Commission soon started to define another directive on the basis of Art. 90.3 to liberalize telecoms services. This time the measure was also contentious in a material sense, since the Commission wanted to restrict the monopoly to the telephony service while several southern European governments, including France, wished to maintain restrictions on data services. All during 1989 the governments tried to negotiate a more favourable solution with the Commission, who took a harsh stand. Finally, a compromise was struck between the Commission and the Council of Ministers in December 1989 to allow licensing conditions for data services in the liberalization directive, which was to be issued together with the regulatory framework (called the Open Network Provision [ONP]) of the Council. Both directives were published in June 1990. But again the directive was challenged before the Court, this time by Spain - with the support of France -, Belgium, and Italy. In early 1991, the Court backed the Commission's choice of Art. 90.3 in the terminal case. In 1992, the services case was also upheld by the Court. ₂

Based on this legal support of the Court the Commission has repeatedly taken recourse to Art. 90 to liberalize aspects of telecommunications sytems. In fact, all liberalization decisions have been based on this provision, and the strategy of the Commission has been to incrementally change the existing two directives to include further areas. All these changes, which were often prepared by a Green Paper, have been accompanied by broad consultations with the national governments, the EP, and private actors, which has strengthened the legitimacy of the Commission to act unilaterally. Thus, the liberalization of satellite services and equipment was included in October 1994; the liberalization of cable TV networks in October 1995; and in early 1996 the liberalization of restrictions on mobile telecommunications operators followed as well as the full liberalization of services and networks from 1998 onwards. While this last change was fully backed by resolutions of the Council, the directive also included the liberalization of all existing (so-called alternative) networks for liberalized services starting in mid-1996. This, as well as the liberalization of cable TV networks, and for mobile telecommunications operators was quite contentious among the member states but the Commission acted nevertheless. ₃

In telecommunications policy, in sum, the Commission has used its competition law rights quite forcefully to enact measures by itself. Thereby member-state governments have been set under pressure very effectively: Especially those governments reluctant to liberalize are forced to cooperate at least in the harmonization measures agreed on at the European level. To respond to the Commission by vetoing its Council proposals is not a feasible counter measure for them as this would mean their worst possible option - a pure market regime for telecommunications services. Given the universal trend towards liberalization started by the American, British, and Japenese reforms, however, telecommunications do not form a very difficult case to demonstrate the amount of agency loss and ability to act of the Commission. Changes in telecommunications regimes are highly overdetermined. Therefore, it is useful to turn to a sector where the Commission has met with less favourable conditions. For this, European electricity policy is a useful example.

The 1988 Commission report on "The internal market for energy" can be seen as an equivalent to the first telecommunications Green Paper. It took stock of the situation in the different segments of the energy sector and of the major barriers to the internal market, emphasizing the need to apply existing Community law - in particular the rules for competition and state aid - to the energy sector, and above all to network-based energy (electricity and gas). In contrast to telecommunications, however, the Commission could not refer to a central judgement of the Court to support these obligations (Slot 1994). The document was not superseded by a broad consultation of relevant actors as the Commission expected little support.

The Commission foresaw a series of liberalization proposals to gradually open electricity and gas networks to third-party use, to help increase trade, and to establish the internal market (Dohms 1994). In particular, the document's reference to an obligation of network owners to give third parties the opportunity to use their networks for the transport of electricity and gas, called "common carriage," ₄ aroused European-wide controversy. Most national governments were profoundly against the plan, and so was the electricity supply industry. Because price differentials were in many ways caused by differences in politicla regulations and levies rather than by divergent levels of efficiency, it was argued that the harmonization (instead of liberalization) of national electricity policies would get much closer to the root of the problem. Moreover, more trade could be easily realized by strengthening the existing collaboration of suppliers in the UCPTE. ₅ However, the Commission enjoyed the strong support of France, which demanded access to foreign networks as a means of exporting its surplus electricity. In fact, France helped to introduce the Commission's proposals by lodging a formal complaint with the Commission about German coal subsidies. It claimed they were a distortion of the EC electricity market that harmed French export interests. Portugal, which wanted to import French electricity through Spain, and the already liberalized UK also supported the Commission's call for liberalization.

First, modest measures were agreed: A directive to increase price transparency for large users of gas and electricity was adopted by the Council in June 1990, and two directives for electricity and gas to increase the transit possibilities of suppliers through each other's networks followed in October 1990 and May 1991, respectively. ₆ Another proposal, to improve the Commission's information on investment plans in the oil, gas, and electricity sectors, did not find support.

The overall plan of the Commission was to gradually abolish or adapt the existing monopolies for the import and export of electricity and gas, followed by those for the production, transmission, and - ultimately - for distribution, while establishing a European-wide regulatory framework on rights for production and access to networks. In view of its significant success before the Court in the telecommunications terminal case, the Commission decided in the summer of 1991 to draft Art. 90 directives as a way to sidestep the resistance of the member states to the liberalization of electricity and gas networks. In addition, it initiated infringement proceedings based on Art. 169 against existing monopolies for the export and import of electricity and gas in France, Denmark, Spain, Italy, Ireland, the Netherlands, Belgium, Greece, Portugal, and the UK (Slot 1994: 525). By this time France had already ceased its support of European policies, correctly perceiving that it would not only benefit but also be forced to enact considerable domestic changes. In view of the strong opposition of member-state governments, the European Parliament, and electricity suppliers, the Commission dropped its plan to use Art. 90 to enforce liberalization and presented proposals for the second stage of the programme as Council directives in February 1992. These included the opening of the networks for large consumers of electricity and gas, now called third-party access (TPA), the right of which was to be extended in the future third phase of liberalization. Further measures of this second phase included the permission to build networks and the organizational unbunding of production, transport, and distribution (Argyris 1993: 34ff; Hancher / Trepte 1992: 155-159).

In the Council, the TPA proposal remained very controversial, only having the supprt of the UK and Portugal. After the European Parliament had delivered its highly critical opinion in October 1993, the Commission presented an amended proposal in December that weakened the regulated TPA to a negotiated TPA. The laggard progress of legislation was paralleled by the infringement procedures, now only against import and export monopolies in France, Spain, Italy, Ireland, and the Netherlands, which were not brought before the Court until 1994. Yet it was reiterated that the monopolies for production, transmission, and finally for distribution would eventually be tackled in the same way. Before, the Competition Commissioner had tried repeatedly to use the legal proceedings as a lever to move the governments to make advances in the Council. Under pressure, France, the major opponent of the liberalization policies, presented an alternative proposal to the TPA system in the Council discussions during 1994, called the Single Buyer concept. This proposal left the monopoly of Electricité de France (EdF) basically intact, but gave third parties the right to buy cheaper electricity from alternative sources and sell it to the network operator (for instance EdF) as the "Single Buyer". Users would profit form these transactions instead of from liberalized electricity. Simultaneously, especially French actors have made the determined effort to use the 1996 Treaty revision as a way to delimit the future impact of competition law in general, and the Commission's competences under Art. 90.3 more particularly. By including the concept of public services into the Treaty the pressure on member-state governments could be reduced considerably, however such an inclusion as such seems most unlikely.

It took the Council two more years and the determined attempts of four presidencies (German, French, Spanish, and Italian) to finally reach a compromise in June 1996. Until the very end the agreement was uncertain as several countries backing the modified TPA proposal (UK, Germany, Portugal, Sweden, Finland, and the Netherlands) wanted further liberalization while France was reluctant to concede more alterations to the single buyer proposal, which was also backed by Belgium, Spain, Italy, Greece, and Ireland. Questions of reciprocal market opening were paramount in the deliberations, backed by a study of the Commission that explicated the shortcomings of the single buyer. To put further pressure on the Council the Commission warned to issue an Art. 90 directive should the Council drag its feet, but it missed its self-set deadline of December 1995. That a compromise could be found was in many ways due to Franco-German endeavours to overcome their contrary positions. France had been very eager to find an agreement before the Court would submit its ruling. Similarly, the European Parliament had tried to put pressure by criticizing that a court ruling was in danger to become decisive in such a politically contentious area.

In the end an icnremental opening of the market was agreed based on the two models of negotiated TPA and the single buyer. Starting in 1999 the market share equivalent of users with more than 40 GWh/ year consumption will be opened, a level which is lowered after 3 years to 20 GWh/ year, and after a further 3 years to 9 GWh/ year. While users aboved 100 GWh are automatically eligible, national governments will designate eligible users to achieve the appropriate market share. ₇ Greece, Belgium, and Ireland gained exemptions for one to two years. In 2006 the Council will negotiate further steps. Following pressure of the German electricity suppliers a reciprocity clause was included into the compromise to avoid too many asymmetries in the opening of markets caused by the two different models. With this compromise - which received the backing of all the governments in the Council and the Commission, and to which no major objections of the EP are expected - the very long discussions on the internal market for electricity could be brought to an end, to which all parties contributed concessions. ₈ Thus the Council managed in the end to find a political compromise before the Court of Justice - whose judgements on the Commission's infringement procedures against import and export monopolies was expected in August 1996 - gave its interpretation of the Treaty obligations.

In sum, electricity policy paints a very different picture of the Commission's ability to act than the telecommunications case. Despite several threats, the Commission did not dare to overcome the clear lack of support in the Council with unilateral action, although legally it stood good chances on the basis of the telecommunications precedent. Instead it only mustered to reluctantly pursue a limited aspect of the existing monopolies through single cases. Thus, this case points to much stronger control of the agent Commission through its principals, so that the comparison of the Commission's action in the two cases promises to allow to identify some conditions for the Commission's ability to act under competition law.

Figure II summarizes the developments for both sectoral policies

The Commission's Ability to Act - Evidence from the Cases

On the basis of this framework for analysis important conditions for the Commission's ability to act can now be singled out. Starting with the telecommunications case, the Commission's action was much helped by the ECJ's ruling in the Case British Telecom in 1985 that telecommunications was a regular economic activity falling under the provision of the Treaty. Because of the superiority and the direct effect of European law this implied that private actors could call upon the Commission or upon national courts to complain about existing restrictions imposed by the telecommunications monopolies. Using this window of opportunity the Commission soon afterwards followed up private complaints on the existing equipment monopolies in several member states (Germany, Netherlands, Belgium, Italy, and Denmark). Already before the publication of the first Green Paper several member-state governments and national monopolists had therefore met with uncertainty as to the longer term impact of European law.

Simultaneously the Commission used its possibilities at informal agenda-setting and took up the new instrument of publishing Green Papers. This allowed it to circulate its analysis of ongoing changes, detailing existing Treaty obligations. At the same time a comprehensive policy programme was put forward. Thereby, the Commission achieved, on the one hand, that the existing uncertainty on the future of telecommunications systems was heightened for the national governments and monopolists. Rapid technical change, international pressure for reform, and newly enforced Treaty obligations implied reform needs. On the other hand the awareness of private actors was raised as to their existing legal rights for expanding activities in telecommunications.

Both parallel actions complemented and strengthened each other. Private actors became aware through the Green Paper discussions and signalled open support to the Commission's programme. Thus the Commission effectively mobilized the national governments' principals. Simultaneously, the Commission recruited potential complainants about existing legal restrictions, which could either lead to separate legal proceedings culminating in a judgement of the ECJ, or on which infringement procedures or single decisions of the Commission against isolated member-state governments could be based. This possibility of case-specific enforcement of European law played a crucial role. In procedures initiated by the Commission national governments are isolated in confronting excessive demands by their agent, and therefore cannot block these as effectively as if all principals were confronted at once. Even if several governments are concerned - as was the case for terminal equipment monopolies - there does not seem to be the attempt of the affected governments to strengthen their position against the Commission by seeking contact and cooperation. Therefore the Commission is in a relatively good position to pursue single cases. Importantly, this helps it in the subsequent attempt to broaden the policy to all member states. Having been forced to alter their domestic legal frameworks despite the fact that other governments continue with comparable though not quite identical restrictions, the concerned governments generally have the incentive to support the Commission in order to see reciprocal conditions realized in the other member states.

Throughout its telecommunications policy the Commission has successfully followed this two-tiered approach of publishing Green Papers and policy documents with detailed policy proposals (altogether five) while starting to pursue some single cases. Crucially, as is very apparent, the Commission thereby relied on the support of non-state actors, interested in the new possibilities opened by technical change. But the skillful use of its rights is only part of an explanation of the successful European telecommunications policy. European policy change based on case-specific measures is very cumbersome and lengthy, resulting in a very fragmented legal order characterized by uncertainty. Thus the possibility of the Commission to impose a unitary policy on the member states through the use of Art. 90 directives was at least as important. The Council of Ministers would not have agreed to realize a similar extent of liberalization through different directives within such a short perioid of time, however many single cases the Commission would have followed up in different countries. Telecommunications liberalization raises significant concerns on employment losses and a geographically even development of infrastructure. The resulting domestic opposition to reform, which has been significant already in highly developed member states such as Germany and France, would have made it difficult to reach a qualified majority in the Council.

Also the use of Art. 90.3 for Commission directives placed high demands on the Commission's skills. It had been used just once before for a directive which only demanded certain information requirements of the national governments. In telecommunications, in contrast, national monopolies were abolished and organizational requirements (the separation of regulatory and operational functions) were specified. For the Commission, and more specifically for the relevant units in DG IV and DG XIII, this meant that broad support for such a bold measure was necessary both within and outside of the Commission. In this respect it was highly important that the Commission received strong support for its overall policy programme in general, and for the first measure of terminal equipment liberalization frodm the national governments more specifically. Notably many high-level officials welcomed the pressure imposed by the Commission on the member states, knowing the constraints placed on telecommunications reform by domestic opposition.

It may be said that the strong material interest of the national governments in a European telecommunications policy provided the opportunity for the Commission to take the procedural risk with the first Art. 90 directive - for the governments did not want to jeopardize the policy programme for general institutional concerns. At the meeting of the telecommunications Council just after the directive, it seemed that no government wanted to initiate legal proceedings. The fact that soon afterwards five countries (France supported by Italy, Belgium, Denmark, and Greece) did call upon the ECJ shows that in this case the telecommunications Council did not represent the interests of the Commission's principals.

The case was slightly different with the Commission directive for services liberalization that was issued in mid-1990 before the ECJ's ruling in March 1991. This measure was materially contentious amidst the governments as the mediterrean countries opposed the liberalization of data services. In this case one finds an even stronger reliance of the Commission on the fact that divergent interests among its principals would prevent it from strong sanctions. Already in the previous terminal equipment case it is quite apparent that a unanimous appeal of all governments to the ECJ out of constitutional interests would have had a much stronger impact - both on the Commission's future actions and on the Court. But several governments were too interested in seeing the material policy succeed to intervene - with the exception of the Netherlands who openly favoured delegating the independent rights of Art. 90 to the Commission. For the liberlization of services some governments much more clearly supported the Commission knowing that this was the only way to realize their interest in European-wide liberalization. The following note by the Financial Times elucidates this for the case of the UK and Germany:

,,The West Germans refuse to be pinned down in public on whether the Commission should abandon its controversial approach of forcing through its programme and put it to a vote. Mr Schwarz-Schilling would not give a clear answer on this in spite of repeated questioning. In private, however, officials admit that they want the Commission to reserve the right to force the reform through in the event of a compromise not being reached. They are not willing to say so in public because of fears that a similar technique will be used in other areas where they disagree. The British government is taking the same approach of supporting the Commission in private put not committing itself in publi." ₉

The conflict on the services directive led to a compromise at the December 1989 meeting of the telecommunications Council where the Commission agreed to allow for licensing requirements for data services. In addition, an institutional compromise was achieved that in future the Commission would consult with the member-state governments when taking recourse to Art. 90 directives. The governments took this compromise to imply that the Commission would henceforth only use Art. 90 under their assent. With such an accord the use of Art. 90 became acceptable to the governments. Nevertheless, Spain supported by France, as well as Belgium, and Italy appealed to the ECJ; but this seems to have been motivated strongly by interests of consistency with the still ongoing terminal equipment case. At least France, the main initiator of the earlier case, quite apparently felt it to be necessary to appeal again so that it would not weaken its chances in the ongoing terminal equipment case. In addition to this indirect support of national governments it was crucial for the Commission that the ECJ backed the use of Art. 90.3 in its judgements in 1991 and 1992, otherwise the Commission could not have pursued liberalization on this basis.

For its following Commission directives - for the liberalization of satellite communications in 1994, and the three directives to liberalize different aspects of networks in 1995 and 1996 - the Commission has aimed to strengthen its position with this instrument. On the one hand, it has been very careful, issueing each directive as an amendment to the existing two directives, with which it could avoid renewed discussion on the procedural basis of Art. 90.3 versus Art. 100A. On the other hand, it has started to institutionalize procedural rules for the adoption of Art. 90 directives as a way to strengthen their legitimacy and possibly their future use also in other areas. Starting with the services directive the Commission has always attempted to first gain the Council's approval through the adoption of a resolution on an overall policy programme. Often prepared by a Green Paper Council resolutions endorsed broad packages of measures, of which the Commission could subsequently present its directive as a part. In a way, then, the broad Council resolutions exhibited a lack of policy clearness the Commission could exploit. After the services directive, moreover, the directives have increasingly been published before their final acceptance by way of separating a first from a second reading of the Commission. The last proposals have been published in the Official Journal with an invitation for comments from all concerned actors. Also the Parliament as well as the Council have been consulted.

With the last three directives to liberalize alternative networks already from 1996 onwards the Commission has once again tried to broaden its independence by proceeding despite the lack of a previous Council resolution. As a result, criticism of the national governments arose, especially from Greece, Portugal, Spain, Italy, Luxembourg, and Belgium, which found one expression in the call upon the ECJ by Spain and Portugal to rule against the cable TV directive. This emphasized how much the Commission is seen to be dependent on their prior agreement for a directive to be issued. But the Commission has not backed down. Since the less developed member states and Luxembourg (due to its size) can get exemptions for several years, the Commission has, however, effectively disturbed very few national interests. In addition, the little time left until full liberalization means that non-implementation by national governments cannot be sanctioned. Fundamentally, all during the last Commission proposals for the liberalization of telephony and networks, it has become increasingly apparent that in view of the current pace of technical change national governments can only trade in the coordinated, European liberalization under the purview of the Commission with the de facto circumvention of restrictions which cannot be enforced any more.

In sum, in European telecommunications policy the Commission has shown a significant ability to act, furthered by adroit strategic action, which combined the rallying for internal Commission support with the one of external actors through the publication of Green Papers, as well as the pursuit of case-specific applications of competition law. The Commission has, however, met exceptionally favourable circumstances for its actions given the significant pressure for reform of telecommunications and the rapid technical change that effectively breaks existing regulatory restrictions. To a remarkable extent, thus, telecommunications liberalization is overdetermined. Because of the significant interests in liberalization, the Commission found strong support by private actors that issued complaints or lobbied governments for acceptance. Neither were the governments bypassed. Rather, the analysis shows considerable support of national governments, despite the fact that their constitutional interests are violated. Partly, the Commission relies on the defection of single governments who refuse to sanction the Commission since they can realize immediate self-interests. This is most clearly the case with the UK, and with Germany but these are only the most apparent countries. Simultaneously, less developed countries are being compensated by the packaging of liberalization with re-regulatory and developmental measures as well as the granted suspensions. National officials and ministers with sectoral interests as well as interested private actors are keen to evade Council decisions on liberalization knowing that the existing political costs would prevent capitalizing on all the potential benefits of liberalization. By having the Commission decide blame can be avoided, although this comes at the possible expense of future costs imposed by a strengthened Commission.

The electricity case immediately shows that these costs are largely under the control of the national governments. The mere existence of the precedent of telecommunications and the confirmation of its rights through the ECJ has not released the Commission from the need of following closely the interests of national governments and other relevant actors. Already its initial act - the publication of a policy document - met with little support. Only large industrial users, the UK, and Portugal signalled support, besides France who soon withdrew. The lack of support for the Commission in this sector corresponds to a less immediate need to enact reforms compared to the telecommunications case.

For electricity, the significant price differentials between the member states affect long-term investment decisions and therefore have put governments in high-price countries such as Germany under increasing pressure to act, which is, however, much less imminent than in telecommunications. Because of its earlier unilateral liberalization decision also the UK has been keen to open the continental market for its electricity supply industry (ESI). The UK example thereby illustrates well the constraints of unilateral reforms which exist in the Community. Because of an intervention of the Commission the UK government was required to subsidize imported French nuclear power on a par with British nuclear power. ₁₀ Otherwise, the Commission would have prohibited the measure as state aid under European competition law. The impossibility to violate against the internal market principles is a much more significant constraint on the continental member states which - in contrast to the UK - have interconnected high power grids allowing for much more trade. Given that France has an oversupply of cheap nuclear power, bordering countries face the risk of significant imports if they do not come to a multilateral reform agreement that, at least, establishes reciprocal conditions.

But not all governments have a slowly evolving longer-term interest in a European policy framework. Especially France has a successful, domestic system that relies, however, on monopolized planning. With the support of some other countries such as Italy, Ireland and Greece, France proposed a very different European policy model, the single buyer - if a change was to be enacted at all. While a two-option approach can be adapted much better to heterogeneous national conditions, it raises difficult questions of reciprocal market access between the countries. Despite formal qualified majority voting it was however hardly an option to attempt to isolate France, not only because of its political importance but also because several member states are dependent on its electricity exports. The long discussions on the electricity directive have thus emphasized the difficulties of European integration, if conditions in the member states are very heterogeneous and strongly entrenched. The eventual solution that was found by including a reciprocity clause in the directive has the drawback to partly violate the logic of the internal market so that it may not withstand legal action.

Because of the national heterogeneity and the only long-term interest in reform it is not surprising that the Commission's attempt in 1991 to realize the internal market on the basis of an Art. 90 directive was unsuccessful. The strong opposition of the national governments against the Commission's attempt to enact far-reaching changes that had not their prior consent implied that the draft already failed to attract enough support inside the Commission for the idea to be pursued any further. Consequently, the legislative proposals were handed over to the Council.

In telecommunications, the adoption of Art. 90 directives was only one part of the Commission's strategy which also relied on case-specific measures to increase pressure and support for the internal market policy. Also in the electricity case the Commission started infringement procedures against existing import and export monopolies for gas and electricity against several national governments. These cases exerted considerable pressure because of the drawbacks of a case-based policy change to all parties concerned in terms of legal certainty and coherence throughout the Community. But in view of opposition from member states the Commission repeatedly postponed its steps, so that there was uncertainty in how far its case-specific measures could be controlled by the principals.

To explain the Commission's relative inability to act it is necessary to take into account the parallel discussions in the Council, which were initiated with its proposal. The ongoing deliberations by the national governments meant that the problems resulting from available Commission measures were always transparent, implying that any action of the Commission would have had to knowingly disturb the difficult process of searching for a compromise and to act openly against its principals' interests. Given its responsibility as initiator, this would be inappropriate behaviour for a considerable time, at least as long as national governments were seen to search for a compromise in earnest rather than to simply obstruct the Commission's proposals. In addition, already the cases against the import and export monopolies with their relatively mild repercussions meant that the Commission was forced to take sides. The heterogeneous situation in the member states implied that only those five countries were affected which had codified monopolies, notwithstanding the fact that such monopolies de facto existed as well in the other member states. The Commission therefore had to balance its disturbance of the Council discussions and the implications of the existing opposition with the need to keep up some pressure and to safeguard its own credibility. Its double role as initiator and enforcement agent was partly in conflict, and the Council discussions allowed governments a check on the case-specific enforcement they normally do not enjoy.

Nevertheless, the Commission did not fail in its attempt. This became apparent with the French proposal for an alternative single buyer system. Because of the parallel infringement procedures France could no longer stick to its first strategy of simple opposition in the Council. The Commission made it clear that without any reaction it could not drop the cases, so that some changes were likely under the ECJ's ruling. With an increasing willingness of other governments to reform, France could also not trust on not being isolated eventually in the Council under qualified majority voting. Had an acceptable compromise in the Council been reached in 1994, it is realistic to assume that the Commission would have left the cases. It is hardly conceivable that the Commission would pursue a more far-reaching interpretation of Treaty obligations for national electricity systems shortly after a political compromise in the Council on the same matter. Rather it would have to withdraw its proposal to the Council if it saw it diverting from the Treaty's foundation.

In the deliberations on reform also the expected position of the ECJ has been a relevant variable. On the one hand, the Commission relies on its support. With no backing at least for the abolition of the import and export monopolies it would have no scope to apply pressure on the member states. Governments could still agree on a multilateral reform programme, but this would not be required by any Treaty obligations, and would therefore be much more unlikely. Depending on the assessment of the Court's position, therefore, the Commission or the national governments had reason to be more conciliatory. For example, for some French actors interested in the status quo, the single buyer proposal appeared as a mistake after the Court had issued two judgements in 1994 with relevance for electricity policy. The Corbeau, and the Almelo cases seemed to indicate that the Court now accorded much more scope to public service obligations, leaving its previous very liberal position (Hancher 1994; 1995). But the likely rulings of the ECJ are very difficult to assess for the governments. Thus, the outstanding judgement of the ECJ was a major incentive for France's then successful efforts to reach a political compromise.

Lastly the Commission's relative difficulties with case-specific measures have to be explained with view to the actions of private actors, mainly the electricity suppliers and the large-volume users as the benefitors from reform. The national electricity suppliers have long established links of cooperation in the UCPTE, which manages the trade on the high voltage grids. Their dominant interest in the maintenance of their monopolies and their well-established links have allowed them to block quite effectively most case-specific measures under European law. In most cases when the Commission has pursued complaints from third parties eager to profit from electricity liberalization and trade, the responsible supplier has made a matching special offer to the user. In addition, external suppliers have generally declined contracts with users located in another supply area. Given their significant exchanges it is not rational for instance for EdF to supply a customer of RWE which could harm its established links to that utility.

High-volume users, against this, would in general profit from case law losening the existing monopolies. However, they have faced a collective action problem in carrying through cases since it is individually rational for each affected user to agree to "sweetheart deals". Thus, the Commission has suffered from a lack of support from third-party complaints that would have strengthened its position against national governments, applying also independent pressure on these. In this respect, electricity policy is significantly different from telecommunications where complaints are generally based on the interest to complete with the dominant suppliers, giving the latter no scope to buy off the complaints.

In sum, the Commission has had to confront the difficult situation in electricity policy that neither governments nor third parties gave her much support. However slowly, this has changed and finally allowed for a political compromise on liberalization nine years after the start of the discussions. For this change several developments have been important. High price countries have been under increasing pressure to improve their locational conditions due to the internal market and the perceived threat of global competition. The difficulties of a unilateral reform within the European legal framework have heightened the attractions of a multilateral reform. In addition British actors have been interested to export their domestic liberalization, as have the new swedish and finnish members. However, given the central position of France and the French interests in the maintenance of the domestic system this support could not be sufficient. The initiation of infringement procedures against the import and export monopolies by the Commission was therefore crucial for the eventual political compromise. The national governments faced difficulties to prohibit these case-specific measures of the Commission, which induced a strong interest to pre-empt the impending court ruling with a political compromise. For the agreement to materialize it has also been important that the positions of the electricity suppliers have changed over the years. Despite their strong coalition and cartel-like behaviour especially some large suppliers have increasingly seen that change is both inevitable and desirable, as the larger utilities stand to gain from competition. ₁₁

Lastly, in this sector the Commission also profits from the diverging interests among its principals. Otherwise the national governments could have agreed to exempt the sector from the Treaty - as was quite clearly the intention when the exception of Art. 90.2 was drafted originally - in view of the difficulties of finding a common policy. But interests of the UK to have its own liberalization spread via the Community, and of governments under reform pressure that can profit from Community requirements when enacting domestic changes, have prohibited this. These interests will also prove an obstacle to the ongoing effort to include a public service concept in the Treaty.

Conclusion

Based on the ruling of the ECJ in the case British Telecom the Commission has used the very favourable circumstances in the telecommunications sector to acquire impressive new competences which were again backed by the Court: the unilateral adoption of directives under Art. 90 with no formally required previous consultation of other European bodies. With its innovative interpretation of a long-existing Treaty article the Commission has gone well beyond what the founding members envisaged, and what many contemporary observers believed to be possible. For its successful application of competition law competences the Commission has benefited from its agenda-setting rights which helped it to initiate learning processes as to the necessary reforms and to rally the support of private actors.

But this precedent set in the telecommunications case could not be used by the Commission subsequently in other contexts, incrementally expanding its mandate. The electricity case shows that national governments enjoy considerable control, notwithstanding the legal backing of the ECJ, and allows to draw some conclusions as to the circumstances under which the Commission can use its competition law powers.

In its telecommunications policy, the Commission managed to combine its different rights very successfully, so that its different actions complemented and strengthened each other. Support of private actors could be garnered as well as the willingness of influential principals to cooperate with the Commission against the opposition of other principals. Because of the strong rationale for liberalization the plans of the Commission met the interests of private actors and governments. The latter's domestic ability to act could be strengthened with the help of the Commission, allowing stronger commitment to reform and the avoidance of significant political costs. This finding reveals an interesting addition to the argument made by Scharpf (1996) on the loss of the member-state governments ability to act because of European legal constraints. Thus, with regard to Art. 90 directives the Commission's action can be better understood as controlled delegation of rights than as a significant example of agency loss.

That the Commission has largely acted as an agent in telecommunications policy despite of the defection of some principals, the exploitation of information asymmetries, and the Commission's structural advantage in long-term planning becomes most apparent when comparing the example of electricity policy. Under less favourable conditions and support for its plans, the Commission had to give up on a commission directive. Also with its other instruments it met with difficulties. Having initiated Council discussions it could not simultaneously enforce the competition law in single cases, as the fabrication on legal facts would have pre-empted the search for a political compromise. Next to this role conflict the cartel-like cooperation of electricity suppliers could undermine the support of large users for such enforcement. But eventually the horizontal interdependence with the other European organs helped the Commission's policy. Under the shadow of an impending judgement, agreement in the Council became finally possible. Because of the greater independence of the European Court of Justice governments cannot influence its rulings but in an indirect way. While it is highly unlikely that the Court's interpretation of the Treaty would openly contradict an agreement of its principals on secondary law, the unanimity requirement to change the Treaty implies very high costs on the losing party, once a judgement has been made.

Much more than its very far-reaching rights to issue directives under Art. 90.3 the Commission's possibility to examine isolated cases thus seems to give it scope to further its policy plans. Here the sources of agency loss come into full play as the principals are incompletely informed about ongoing examinations and have difficulties to assess their long-term consequences. Since - given the diversity of interest - principals are unlikely to unanimously agree on explicit limits to the Commission's rights, it is well-placed to incrementally enact changes.

Bibliography

Argyris, Nicholas, 1993: Regulatory Reform in the Electricity Sector: An Analysis of the Commission's Internal Market Proposals. In: Oxford Review of Economic Policy 9(1), 31-43.

Bayliss, B.T./Ali M. El-Agraa, 1990: Competition and Industrial Policies with Emphasis on Competition Policy. InAli M. El-Agraa (ed.), The Economics of the European Community. New York: Philip Allan, 137-155.

Bulmer, Simon, 1994: Institutions and Policy Change in the European Communities: The Case of Merger Control. In: Public Administration 72(3), 423-444.

Burley, Anne-Marie/Walter Mattli, 1993: Europe Before the Court: A Political Theory of Legal Integration. In: International Organization 47, 41-76.

Dohms, Rüdiger, 1994: The Development of a Competitive Internal Energy Market in the European Community. In: Conneticut Journal of International Law 9(3), 805-856.

Ellger, Reinhard, 1992: Telecommunications in Europe: Law and Policy of the European Community in a Key Industrial Sector. In:William James Adams (Hrsg.), Singular Europe. Economy and Policy of the European Community after 1992. Ann Arbor: The University of Michigan Press, 203-250.

Fiorina, Morris P., 1982: Legislative Choice of Regulatory Forms: Legal Process or Administrative Process? In: Public Choice 39, 33-66.

Garrett, Geoffrey/ George Tsebelis, 1996: An Institutional Critique of Intergovernmentalism. In: International Organization 50(2), 269-290.

Graupner, Frances, 1973: Commission Decision-Making on Competition Questions. In: Common Market Law Review 10, 291-305.

Hammond, Thomas H., 1996: Formal Theory and the Institutions of Governance. In: Governance 9(2), 107-185.

Hancher, Leigh, 1994: Case C-320/91 P, Procureur du Roi v. Paul Corbeau. In: Common Market Law Review 31, 105-122.

Hancher, Leigh, 1995: Case C-393/92, Gemeente Almelo and Others v. Energiebedrijf IJsselmij NV. In: Common Market Law Review 32(1), 305-325.

Hancher, Leigh/Peter-Armin Trepte, 1992: Competition and the Internal Energy Market. In: European Competition Law Review 13(4), 149-160.

Héritier, Adrienne, 1993: Policy-Netzwerkanalyse als Untersuchungsinstrument im europäischen Kontext: Folgerungen aus einer empirischen Studie regulativer Politik. In: Politische Vierteljahresschrift 34, Sonderheft 24/1993, 432-447.

Héritier, Adrienne, 1996: The accommodation of diversity ion European policy-making and its outcomes: regulatory policy as a patchwork. In: Journal of European Public Policy 3(2), 149-167.

Kingdon, John W., 1984: Agendas, Alternatives, and Public Choices. Boston: Little.

Kydland, Finn E./Edward C. Prescott, 1977: Rules Rather than Discretion: The Inconsistency of Optimal Plans. In: Journal of Political Economy 85(3), 473-491.

Majone, Giandomenico, 1992: Market Integration and Regulation: Europe after 1992. In: Metroeconomica 43, 131-156.

Majone, Giandomenico, 1995: Mutual Trust, Crecible Committments and the Evolution of Rules for a Single European Market. EUI Working Paper RSC No.95/1. Florenz: European University Institute.

Majone, Giandomenico, 1996: Independent Agencies and the Delegation Problem: Theoretical and Normative Dimensions. Manuscript 1996.

Marks, Gary, 1993: Structural Policy and Multilevel Governance in the EC. In: Alan W. Cafruny/Glenda G. Rosenthal (eds.), State of the European Community: the Maastricht Debates and Beyond. Boulder, CO: Lynne Rienner, 391-411.

Martin, Lisa L., 1992: Interests, power, and multilateralism. In: International Organization 46(4), 765-792.

McCubbins, Mathew D./Roger G. Noll/Barry R. Weingast, 1987: Administrative Procedures as Instruments of Political Control. In: Journal of Law, Economics, and Organization 3(2), 243-277.

Moe, Terry M., 1984: The New Economics of Organization. In: American Journal of Political Science 28 (November), 739-777.

Moe, Terry M., 1990: The Politics of Structural Choice: Toward a Theory of Public Bureaucracy. In: Oliver E. Williamson (ed.), Organization Theory. From Chester Barnard to the Present and Beyond. Oxford: Oxford University Press, 116-153.

Moravcsik, Andrew, 1995: Liberal Intergovernmantalism and Integration: A Rejoinder. In: Journal of Common Market Studies 33(4), 611-628.

Ostrom, Elinor, 1986: An Agenda for the Study of Institutions. In: Public Choice 48, 3-35.

Pierson, Paul, 1996: The Path to European Integration: A Historical Institutionalist Perspective. In: Comparative Political Studies 29(2), 123-163.

Pierson, Paul/Stephan Leibfried, 1995: The Dynamics of Social Policy Integration. In: Stephan Leibfried/Paul Pierson (eds.), European Social Policy. Between Fragmentation and Integration. Washington D.C.: The Brookings Institution, 432-465.

Pollack, Mark, 1995a: Obedient Servant or Runaway Eurocracy? Delegation, Agency, and Agenda Setting in the European Community. Paper submitted for publication by the Center for International Affairs Working Paper Series, August 1995.

Pollack, Mark, 1995b: Regional Actors in an Intergovernmental Play: the Making and Implementation of EC Structural Policy. In: Carolyn Rhodes/ Sonia Mazey (eds.), Building a European Polity? Boulder, CO: Lynne Rienner, 361-390.

Ritter, Kurt Lennart, 1994: EC Antitrust Law and Energy. In: Barry E. Hawk (ed.), Antitrust in a Global Economy. Annual Proceedings of the Fordham Corporate Law Institute. Irving-on-Hudson, NY: Transnational Juris Publications, 127-157.

Sandholtz, Wayne/ John Zysman, 1989: 1992: Recasting the European Bargain. In: World Politics 42(1), 95-128.

Scharpf, Fritz W., 1994: Community and Autonomy: Multi-level Policy-making in the European Union. In: Journal of European Public Policy 1(2), 219-242.

Scharpf, Fritz W. (1996) Negative and Positive Integration in the Political Economy of European Welfare States. In: Gary Marks, Fritz W. Scharpf, Philippe C. Schmitter and Wolfgang Streeck, Governance in the European Union. London: Sage, 15-39.

Slot, Piet Jan, 1994: Energy and Competition. In: Common Market Law Review 31, 511-547.

Stigler, George, 1971: The Theory of Economic Regulation. In: Bell Journal of Economics and Management Science 2, 3-21.

Torenvlied, René, 1996: Political Control of Implementation Agencies. Effects of Political Consensus on Agenca Compliance. In: Rationality and Society 8(1), 25-56.

Williamson, Oliver E., 1985: The Economic Institutions of Capitalism. New York: The Free Press. 11202 words, 28.10.1996.

Figure I

Figure II

Note 1: In addition to these structural advantages agents can profit from unintended consequences of action arising from decision-making under uncertainty and, for instance, issue density that leads to repercussions of decisions on other sectors etc. (see Pierson/ Leibfried 1995 for such an argument). I do not consider these possibilities separately since there is no reason why unintended consequences - in principle - could not work both ways, requiring a symmetrical argument: issue density and uncertainty could further or hamper the Commission's cause. Only when combined with a structural advantage, such as information asymmetry, could the Commission be generally seen in an advantageous position. Back.

Note 2: Only a minor aspect of both directives - the definition of special rights - was declared void by the Court. Back.

Note 3: Because of it, Spain and Protugal sought again the assistance of the ECJ against the liberalization of cable TV networks. In view of the time lag and the previous judgements, this is however unlikely to alter much. Back.

Note 4: The Commission failed to specify the obligations implied by "common carriage." Later the term was substituted by third-party access (TPA). Back.

Note 5: The following countries belong to the UCPTE (Union pour la Coordination de la Production et du Transport d'Electricité): Albania, Austria, Belgium, France, Germany, Greece, Italy, Luxembourg, the Netherlands, Portugal, Spain, Switzerland, and the countries of the former federal state of Yugoslavia. Back.

Note 6: The transit directive was modelled on a disagreement between France and Spain on the terms of use of Spain's network for French exports to Portugal, which had been settled in the meantime (Ritter 1994: 139-140). Back.

Note 7: Different national consumption patterns stand behind this rule. The direct eligibility of customers with 40 GWh/ year consumption, for instance, would have opened 32.5% of the German but only 24.8% of the French market (FAZ 22.6.1996). Back.

Note 8: In fact the compromise seems to have been much due to the determination of Chancellor Kohl to get an agreement, since the CSU and the FDP had insisted on further French concessions. The utility RWE thus complained that a Court ruling would have been more consistent with German interests. But also EdF criticized that the French government had given up central positions. Back.

Note 9: FT 23.6.1989: Hugo Dixon: Paris and Bonn seek accord on telecom market. Back.

Note 10: The fossil fuel levy imposed on electricity sales for this subsidy amounted to about 10% of electricity bills. Back.

Note 11: The heavy involvement of electricity suppliers in telecommunications has made it at least more difficult for them to continue their opposition to liberalization. Back.