h w< Edited by Professor of Law, University of Liege and College of Europe, Belgium Research Fellow, University of Liege, Belgium Nicolas Petit Research Fellow, University of Liege, Belgium Cheltenham, UK • Northampton, MA, USA 8. Coordination of European and Member State Regulatory Policy: Horizontal, Vertical and Transversal Aspects Pierre JLaroitche INTRODUCTION Economic regulation has experienced spectacular growth in the EC since the 1990s. Riding on the wave of liberalization efforts, whole new regulatory frameworks were put in place in areas such as telecommunications (now extended to electronic communications), broadcasting, posts, air transport, rail transport and energy. Each of these frameworks entailed - more or less explicitly - the creation of a national regulatory authority (hereafter 'NRA') to administer that specific framework. On the competition law side, in addition to the growth of merger control, recent reforms have opened the door to a significant increase in activity in. the enforcement of Articles 81 and 82 EC as well, through the empowerment of national competition authorities (hereafter NC As) and national courts as fully-fledged enforcers. Indeed the current institutional picture is as follows. At the European level, the Commission is active both as an enforcer of EC competition law and as a major player in the development of the various sector-specific regulatory frameworks. The Court of Justice plays its usual roles. In each Member State: - An NCA is in charge of applying EC and national competition law; - For each sector-specific regulatory framework (usually), an NRA is competent to apply national law (a large part of which often simply implements EC legislation); - Finally, national courts handle numerous cases either under competition law (as a first instance alternative to the NCA or the Commission) or under Coordination of European and Member State Regulatory Policy 165 sector-specific regulation (usually by way of judicial review). For the sake of keeping this chapter within bounds, national courts will not be dealt with in much detail. Bringing them into the analysis adds a further layer of complexity. Taking the EU enlargement into account, this means that hundreds of authorities are involved in one way or another in the operation of economic regulation in Europe. This chapter deals with the most obvious issue that can be expected to arise in view of such a substantive and institutional inflation, namely coordination. For the sake of limiting complexity, it focuses on only three frameworks, namely general competition law and sector-specific regulation in the telecommunications and broadcasting sectors. Furthermore, it does not go into a discussion of which of the abovementioned authorities qualify as agencies or any other category. It simply assumes that all of the above are relevant players in the operation of economic regulation. Finally, it is concerned with EC law. In order to organize the discussion, a distinction will be made - to the extent necessary - between three dimensions: - Horizontal coordination refers to the coordination between authorities from the various Member States active under the same framework (e.g. amongst all NCAs or amongst all NRAs for electronic communications, etc.); - Vertical coordination refers to the coordination between the Member State and European levels, here as well as in a given framework; - Transversal coordination refers to the coordination between authorities active under different frameworks (e.g. between the NCA and the NRA for electronic communications). After a brief recall of why we came to this state of affairs (I), this contribution looks at whether coordination is needed (II) and if so, how it is talcing place (III). FACTORS LEADING TO THE PROLIFERATION OF RELEVANT AUTHORITIES It is interesting to note that, while following different paths altogether, competition law and sector-specific regulation came to the same result, namely that national authorities should play the leading role in the enforcement of the law. For competition law, the path was somewhat longer. The original enforcement system was fairly centralized, with the Commission holding a monopoly on 166 Regulation through Agencies in the EU exemptions pursuant to Article 81(3) EC and the NCAs not being empoweied by EC law itself to apply Articles 81 and 82 EC. The new enforcement system resulting from Regulation 1/2003, coming into force on 1 May 2004, matks a sea change in that Article 81(3) EC is made directly applicable (thereby terminating the whole notification and exemption system), and the NCAs aie empowered by Regulation 1/2003 itself to apply Articles 81 (in its entirely) and 82 EC. The main reasons put forward for this change were that: - After 40 years, EC competition law had been fairly well mapped out, so that the monitoring and control function of the notification system were no longer necessary to ensure a consistent development and application of the law; - The Commission services were overburdened with notifications which were for the most part relatively uncontroversial and of limited interest. Given that the resources made available at EC level for the enforcement of EC competition law were not going to be increased, it was necessary to reduce that burden and free up resources for truly significant cases. As a consequence, the enforcement burden was shifted in great part onto NCAs and national courts, and beyond that (since notifications cease to be available) onto the shoulders of the private parties themselves. The latter indeed now have to figure out and decide for themselves whether their course of conduct breaches competition law and risks incurring the wrath of the competent authorities. For sector-specific regulation, the situation is simpler. In line with a basic tenet of EC law, the enforcement of the new regulatory framework was left to the Member States. At least as regards electronic communications, the possibility of EC-level enforcement, through a regulatory agency (e.g. a European communications authority), was discussed a number of times, but there was never any political will on the part of Member States to depart from that basic tenet and move part of the enforcement at the EC level. Accordingly, the NRAs ended up being the first-line enforcers of EC sector-specific regulation. In both areas, a common picture emerges, namely that of national authorities (be they NCAs or NRAs) as the centre of gravity for the 'day-to-day' enforcement of EC law, together with national courts. The Commission remains somewhat removed, dealing with major policy issues as well as (in competition law) the most significant cases. THE NEED FOR COORDINATION At first sight, one might be tempted to conclude immediately that coordination between the various competent authorities is direly needed. Yet it is worth Coordination of European and Member State Regulatory Policy 167 dedicating some attention, as a preliminary matter, to whether coordination is needed at all.. Horizontal and Vertical Coordination As a starting point, jí is trite to say that the most sectors of the economy now have a global dimension, and that firms, as soon as they reach a certain size, plan and operate at a regional if not global level. The logical consequence is that regulation should correspondingly be coordinated - at the very least - at the level at which firms operate. For a pan-European or global firm, contradictory or even incompatible decisions arising from the respective authorities of various Member States engender significant costs, since they complicate the entire range of activities, all the way from business planning down to operations. They also give nse to distortions of competition, when competing firms conduct a large part of their respective operations under differing national legal and regulatory frameworks, some of which are more favourable than others. The situation is typical of failures in the internal market, and the solution involves greater coordination at EC level and beyond. At the same time, an examination of the literature on regulatory competition illustrates that there are some reasons for refraining from coordinating the actions of national authorities more than necessary: - Learning-by-doing: There is widespread agreement on the general competition law and regulatory principles (e.g. access on a nondiscriminatory, transparent, cost-oriented basis). However, more specific decisions are notoriously difficult to take, given policy, technical and economic considerations (e.g. What type of access must be given? Which conditions are not discriminatory? What price level qualifies as cost-oriented?). A number of options are open, none of which can be dismissed out of hand or picked as the obvious winner. In such cases, authorities which have to take decisions on the same or similar issues can benefit from the experience of those which already had to commit themselves to a decision. With time, one or more best practice(s) should emerge. - Closeness to the playing field: National authorities are usually in a better position to know the situation in their respective jurisdictions, if only because their resources are devoted solely to that jurisdiction. - Responsiveness and flexibility: Furthermore, the various decision-making procedures available under EC law are known to take some time: coordination at the EC level is thus likely to lead to longer decision-making processes. As a result, regulatory intervention will then be less responsive and flexible, since both its onset and its termination will not follow as closely the development of the market. Even under EC competition law, where the 168 Regulation through Agencies in the ED Commission holds decision-making powers in individual cases, it is hkelv that national authorities will be in a better position to respond more quickly to problems arising within their jurisdiction. This literature has begun to find its way into the case law of the ECJ, as the Court has started to emphasize that the mere fact that national laws differ is not as such sufficient to justify harmonizing them, in the absence of evidence that the difference leads to significant barriers to intra-Community trade or appreciable distortions of competition.' in other words, liiere are some benefits linked with leaving matters to be dealt with at national level, and a clear case must be made that the costs of doing so offset those benefits before harmonization at EC level can be envisaged. In the end, no simple solution can be put forward to ensure the appropriate level of coordination. Nonetheless, some guiding principles can be put forward: - The benefits of a more decentralized approach, as described above, will not arise automatically from the mere fact that decisions are taken by national authorities. For instance, there is little to be gained if a national, authority takes a decision in complete isolation, without examining what other authorities have done or are contemplating doing. Similarly, taking a different decision for political reasons (power struggle, etc.) unrelated to the situation of the market is counterproductive. It would follow that, at the veiy least, national authorities should be obliged to take into account the work of other authorities, and that they should give reasons, especially if they choose a different route than their counterparts. - The benefits of coordination can equally be lost through overly long or complicated procedures. The coordination procedure must therefore aim to keep costs and delays to a. minimum. The amount of coordination can vary depending on the level of the decision, which ranges from general policy decisions to individual cases. At the level of general policy decisions, i.e. legislation, a strong case can be made for coordination. At the opposite end of the range, i.e. decisions in individual cases, the benefits of a decentralized approach are overwhelming. Yet one of the main characteristics of the newest regulatory framework (electronic communications2) is that a number of significant decisions are taken at an intermediate level, somewhere between legislation and individual decisions. For instance, decisions concerning the main aspects of interconnection with the incumbent (tariffs, conditions of colocation, etc.) are derived from higher legislative principles (cost-orientation, non-discrimination); at the same time, they go beyond the individual case, since they influence the operations of the whole industry. Similarly, under competition law, a number of significant decisions will be made Coordination of European and Member State Regulatory Policy 169 in leading cases (often processed by the Commission), whose impact is felt much beyond the parties to the case. That intermediate level definitely .benefits from coordination, yet at that level coordination must be more carefully balanced against the benefits of a more decentralized approach. Transversal Coordination While a measure of 'regulatory competition' within a single framework might be healthy and beneficial, the situation is different when it comes to transversal coordination. First of all, transversal coordination issues only arise to the extent that the various authorities can render decisions which diverge and cannot both be complied with. For instance, it is unlikely that transversal coordination problems would arise between the NRAs in charge of post and energy (which does not prevent their decisions from contradicting each other at a more theoretical level, e.g. in the choice of accounting standards). Transversal coordination is most likely to be an issue between the authorities in charge of applying competition law (Commission, NCAs, national courts), on the one hand, and each of the sector-specific authorities (NRAs), on the other hand. Indeed, in the current state of EC law, competition law has taken an expansive course whereby a large number of regulatory problems (access, discrimination, pricing, etc.) can also be dealt with under competition law, thereby creating a substantive overlap between competition law and each of the main areas of SSR under study here. Moreover, transversal coordination can also be an issue between NRAs dealing with neighbouring sectors, first and foremost electronic communications and broadcasting. Secondly, the various authorities have different remits, however much their respective regulatory frameworks might converge. Accordingly, they do not truly 'compete' with each other. It is conceivable that their respective decisions might diverge because each of them sees itself as having a specific mandate, with no added value coming out of this divergence in die form of learning-by-doing, closeness, responsibility or flexibility. Thirdly, unlike horizontal coordination where the national authorities exert their jurisdiction over different geographical areas, transversal coordination can involve authorities whose competence extends to the same territory. In this case, the consequences of a lack of coordination are far more severe than simply having to incur extra costs to comply with contrasting decisions in different territories. From a practical perspective, the only redeeming feature of multiple authorities exercising overlapping jurisdiction over the same territory is that claimants enjoy a choice of fora in which to pursue their grievances; in each case, they will presumably pick the one which they perceive as the most favourable to their 170 Regulation through Agencies in the EU claim. Obviously, this may contribute to keeping the pressure on prospective defendants to comply with applicable law and regulation. For prospective defendants, however, it can become difficult to manage legal and regulatory matters when claims can arise in a number of fora. Most significant firms are likely to fall on the claimant side in some Member States and on the defendant side in others, so that they experience both advantages and disadvantages. The risk of contradictory decisions from these various instances affects claimants and defendants alike, however. Claimants who would have erred in their choice of forum might see their competitors obtain better results in another forum, More significantly, defendants might be put in a position where they have to comply with contradictory or perhaps even incompatible rulings. In fact, uncoordinated action by the various authorities undermines the very basis for regulatory intervention: because of excessive costs arising from contradictions and incompatibilities, regulatory intervention could very well fail to deliver any overall benefit. Moreover, even in the absence of contradictions and incompatibilities, a lack of transversal coordination can seriously undermine the efficiency of enforcement. When two or more authorities working under different regulatory frameworks are involved in settling a given case, procedures can drag on for years. While it is important to safeguard the rights of defending parties, this should not come at the expense of effective enforcement of regulatory policy and of the rights of claimants. In particular, under EC law, it must not be forgotten that, as a matter of principle, EC law must be given its full effectiveness (effet utile) by Member States, which would imply a certain level of transversal coordination. Accordingly, in contrast with the more nuanced situation as regards horizontal or vertical coordination, it would be tempting to issue a strong plea for transversal coordination. HOW COORDINATION IS ENSURED On the assumption that coordination (horizontal, vertical and transversal) would be required, the next issue is then how it is taking place. In order to fully understand how EC law seeks to ensure coordination between regulatory authorities, it is necessary to examine both substantive and procedural avenues. Substantive Means to Ensure Coordination Usually, one would think that the coordination of regulatory policy as between the various authorities would be conducted through institutional or procedural measures. However, both EC competition law and electronic communications Coordination of European and Member State Regulatory Policy 171 law show that it is possible to achieve some coordination via substantive law. These measures typically 'harmonize' the law to be applied by the various aulhorities or the substantive tasks of these authorities. Under EC competition law, it is interesting to look at Article 3 (1) of Regulation 1/2003, which provides that national authorities (NCAs and courts) must apply EC competition law as soon as it is applicable in a given case (i.e. if there is an effect on trade between Member States), irrespective of whether they also apply tlieir own national competition law as well. Even though the final version of that provision differs from the original Commission proposal,3 in practice the effect is not much different: given the primacy of EC law in case of conflict, there is a fair chance that in the long run national authorities will focus most of their attention on the application of EC law, at the expense of national law. The Commission retains a significant role in the enforcement of EC competition law. It intends to continue dealing with major cases and issue communications and notices, so that it will still be the leading authority on EC competition law. Accordingly, the Commission will be able to influence the work of national authorities by shaping the evolution of the EC competition law which all authorities are bound to apply. This can already be seen in the Commission's Guidelines on the application of Article 81(3) EC,4 where it is attempting to reshape the case law on that point in such a way as to orient the work of the NCAs and national courts. As regards sector-specific regulation, the trend is to align it in substance with competition law. Indeed, Commissioner Monti has been advocating such an alignment in recent pronouncements. This trend is most visible in the regulation of electronic communications; other regulatory frameworks find themselves at a prior stage, where the regulation is infused with economic analysis but without explicitly using competition law concepts. The mainstay of electronic communications regulation, namely the heavier regulatory framework applicable to operators with Significant Market Power, has now been aligned with competition law. It runs along the three stages of market definition, market analysis and remedies, and at each stage reference is made to the corresponding competition law concepts (relevant market, dominance, etc.). While it is not empowered to take any leading decisions in individual cases, here as well the Commission is actively shaping the work of NRAs through its Guidelines on Market Definition and Analysis5 and its Recommendation on Relevant Markets.6 A common point between the two areas surveyed above is therefore the use of EC competition law to minimize the influence of national competition law or to streamline the substance of sector-specific regulation. Given that the Commission remains the key player in the development of EC competition law,7 it is then in a position to effect a certain coordination of regulatory policy (horizontal, vertical and transversal) without using any institutional or procedural means. 172 Regulation through Agencies in the EU Some reservations should nevertheless be made. First of all, the influence of the Commission as the coordinating instance is felt mostly through soft-Jaw instruments, in EC competition law, soft-law instruments (notices, guidelines) have traditionally been issued on the basis of decision practice in individual cases, where the Commission views were confronted, with those of interested parties and ultimately subject to the control of the ECJ. in more recent soft-law instruments, the Commission tends to set out its views as regards newer issues which have not yet been dealt with in individual cases. These views being persuasive if not outright authoritative, they are often not challenged subsequently in individual cases.8 In the absence of 'battle-testing', there is a risk that the quality and accuracy of soft-law instruments would decrease. Secondly, the coherency of competition law is likely to suffer if it is used as an instrument to coordinate regulatory policy across different frameworks. In the longer term, it could be that either competition law concepts become formless or that 'sector-specific competition law' islands begin to emerge. Recommendation 2003/311 on relevant product and service markets for the purposes of electronic communications regulation,0 for instance, pays lip service to the concept of relevant market under competition law. In fact, however, it is an exercise in identifying areas of the telecommunications sector where regulation is likely to be needed because of high and persistent barriers to entry, behind which effective competition is unlikely to arise and where the application of competition law does not appear to provide a solution. This is a solid economic approach, but it does not entirely square in with competition law. Institutional and Procedural Means to Ensure Coordination Beyond substantive measures, coordination of regulatory policy is mostly conducted through institutional and procedural measures. Here it is convenient for the purposes of analysis to revert to the distinction between horizontal and vertical coordination, on the one hand, and transversal coordination, on the other hand. Horizontal and vertical coordination Coordination amongst national authorities and between the authorities and the Commission was a. key point in the discussions leading up to Regulation 1/ 2003 and to the new regulatory framework for electronic communications. As concerns Regulation 1/2003, the whole decentralization endeavour received frequent and vocal criticism for creating a risk that the uniform application of EC competition law would be unwound through the independent actions of the various authori ties. A cacophony of diverging or even conflicting interpretations - all of them valid on their face - would arise out of the work of the NCAs and national courts. In order to counter that risk, Regulation 1/2003 Coordination of European and Member State Regulatory P.olicy 173 requires close cooperation between all authorities involved.'0 As a last resort, should it strongly disagree with the course of action envisaged by an NCA, the Commission holds the express power to relieve an NCA from a case and take it over." Leaving aside these extreme cases, Regulation 1/2003 rather rests on a premise of constant and institutionalized cooperation. In particular, it provides for the creation of a network regrouping the Commission and the NCAs, the European Competition Network,'2 whose task is to ensure consistency in the work of the various authorities.13 As regards sector-specific regulation, the new EC regulatory framework for electronic communications strengthens the position of the NRA as the main authority hi charge of the application and enforcement of the law. At the same time, in order to alleviate concerns not unlike those expressed as regards the decentralization process under competition law, the work of the NRAs is subject to a number of control mechanisms in order to ensure consistency across the EU and prevent over- or underenforcement. In particular, NRAs are bound to consult the other NRAs and the Commission before taking decisions in application under the new SMP procedure (whether concerning market definition, market analysis or remedies).I4 In some cases, the Commission holds something of a veto right over draft measures from NRAs.'5 Outside of these exceptional cases, the new electronic communications framework also assumes constant and institutionalized cooperation between the authorities. It makes room for the creation of a network of regulators comprising the Commission and the NRAs, called the European Regulators Group (ERG).16 Despite differences in institutional direction17 and substance,18 the reform of EC competition law and the new electronic communications framework evidence a number of similarities when, it comes to the institutional structure: - National authorities are at the forefront, and EC law devotes considerable attention to their setup and powers; - The Commission stays in the background and deals with the most important cases (competition law) or with the broad lines (electronic communications); - The Commission uses soft-law instruments to set out its position; - The Commission monitors the work of national authorities and can if necessary impose its will against a national authority if it believes that the proposed course of action of that authority would impede the consistency of Community law; - The Commission and the national authories cooperate within a network that is meant to ensure consistency without the Commission having to use the hard powers mentioned just above.19 Under both competition law and electronic communications regulation, there is some tension between the horizontal and vertical dimensinn« nf onwHinQfi^ 174 Regulation through Agencies in the EU Vertical coordination is ensured with relatively classical means - rules of conflict in favour of the Community level20 and ultimately a power to prevent national authorities from taking decisions that would endanger consistency in the eyes of the Commission. The Commission is thus clearly placed in a superior position with respect to the NCAs and NRAs.2! Horizontal coordination, in contrast, is entrusted to a fairly innovative institution, namely the network of national authorities. It is in the essence of such a setting that the authorities cooperate on an equal footing, as peers. Yet the Commission is also part of the network. In view of this tension, it appears too early to predict how the system of coordination will work. On the one hand, it could be that the horizontal cooperation mechanism - the network - would become prevalent (leaving the harder vertical instruments unused), so that the leadership for the evolution of the law would truly pass to the authorities acting collectively. This setup would, also allow some measure of regulatory competition (which can also be construed as a flaw in coordination), since network members are not obliged to agree. On the other hand, it could be that the vertical coordination mechanisms would turn out to be used more often than originally envisaged, ensuring perhaps greater consistency but turning the national authorities into some kind of branch office. Transversal coordination It was seen above that the case for transversal coordination might be stronger than for horizontal or vertical coordination. Yet here EC law is weaker. Of course, EC law seeks to avoid conflicts between the NRA and NCA. The Framework Directive on electronic communications, for instance, provides that Member States must ensure that the NRA consults the NCA and cooperates with it.22 Moreover, the NRA and NCA must exchange the information required for the application of the regulatory framework.23 In any event, even without specific provisions in secondary law, primary EC law would oblige the NRA to work together with the NCA, since the NRA must avoid contradicting or undermining EC competition law. Most Member States have implemented the above through a cooperation agreement between the NRA and NCA, which typically assigns common cases to one of the authorities - usually the NRA, with rules to refer cases from one authority to the other accordingly. When dealing with such a case, the leading authority must consult with the other. It is interesting to look closer into such a rule to try to assess whether it is a mere priority rule or a jurisdictional rule of more fundamental significance. A few cases can help in this assessment. Firstly, the NRA can simply fail in its mission by not launching proceedings or dragging its feet. In such cases, the NCA or the Commission should be able to intervene.24 Coordination of European and Member Stale Regulatory Policy 175 Secondly, the NRA can discharge its task but reach a result with which the NCA cannot agree (or even reach an agreeable result for reasons which the NCA cannot subscribe to). From a very formalistic perspective, it could be held that the NCA should be able to intervene to restore the primacy of competition law. However, competition law can often receive many interpretations, especially on difficult issues such as arise before NRAs. It could be that the NRA's decision is consistent with another interpretation of competition law than the NCA's, and that both interpretations are equally reasonable. In such a case, the two authorities should settle their divergences before the court on review from the NRA decision (where the NCA could intervene in the proceedings), instead of the NCA opening its own proceedings to stake its claim to a different interpretation of competition law. Thirdly, the NRA could discharge its tasks in a manner agreeable to the NCA, but see its decision subsequently quashed by the courts. If the NRA decision is invalidated on the merits, it is difficult to see how the NCA could carry on in the same case and reinstate the NRA decision in practice, on the basis of competition law.25 if, on the other hand, the NRA decision is invalidated on formal grounds (including lack of jurisdiction on the part of the NRA), the NCA should be able to take up the case, even if the parties are then being dragged through yet another proceeding.26 In the light of the foregoing, it would seem that cooperation agreements between NCA s and NRAs cannot do more than establish a priority rule, leaving the NCA with many possibilities to intervene after, or in the place of, the NRA. From the perspective of judicial and administrative efficiency, this is disappointing. While it might be important to offer adequate avenues for the enforcement of economic regulation, sequential interventions on the part of the NRA and the NCA in the same case, each with the possibility of judicial review, brings with it considerable delays, costs and inefficiencies. Compliance with EC law could actually require more than just a cooperation agreement between the NRA and the NCA. Indeed, it is not just a matter of implementing secondary legislation (the Framework Directive), but also to ensure the effectiveness (effel utile) of EC law, as discussed earlier.27 Lengthy and complex procedures do not meet that last standard. Among alternative options, it could be possible to integrate the NRA and NCA. For instance, the UK has given its NRA the power to apply competition law in addition to regulation. This solution does not eliminate the transversal coordination problem unless die NCA sees its jurisdiction curtailed accordingly. It would be counterproductive to do so (given that the NCA is otherwise competent across the whole economy), and further it would almost certainly lead to litigation on jurisdictional grounds. The Netherlands considered a more radical option, namely to bring the NRA within the NCA (as a special chamber within the NCA). This Oütionhas the great advantao-^ nf ;*h™-««i™«~ *i------'-■ 176 Regulation through Agencies in the EU of transversal coordination. At the same time, the NRA is also entrusted with tasks that are less immediately related to competition law.2S A unified authority would conceivably neglect these tasks; perhaps they could be left with a 'rump NRA, but then the coherency of regulatory policy would also suffer. Another option could be to improve the division of labour between NRAs and NCAs on the substance. In essence, the NCA would have the last say over market definition and analysis, while the NRA would focus on remedies. Such a division of labour can be observed in certain instances. Under the electronic communications framework, the influence of competition law is greatest at the level of market definition and analysis. These two areas are mapped out for the NRAs through the Guidelines on Market Definition and SMP assessment29 as well as the Recommendation on Relevant Markets30 issued by the Commission.31 In these two areas as well, the NRAs are subject to greater control by the Commission.32 Presumably, the interplay with the NCAs will also be most intense for market definition and analysis. Jn contrast, the NRAs enjoy more leeway as regards the imposition of remedies. The regulatory framework does not provide for the issuance of any notices on that point,33 and the Commission has fewer means of control.34 This state of affairs matches the evolution of competition law. It seems that the competition authorities, first and foremost the Commission, are content to use their superior investigative powers to gather market data and conduct the substantive assessment, leaving then the NRAs with the 'dirty work' of working out and implementing the remedies. This division of tasks was first evidenced in the inquiries opened in 1997, concerning international telephone prices and fixed and mobile telephony prices. In both cases, the Commission gathered substantial amounts of data on the strength of its investigative powers (essentially through questionnaires), analysed that data and came to the conclusion that there were serious competition law issues, it then proceeded to urge NRAs to act on these findings and adopt adequate remedies, which they did in most cases. The Commission was then able to close these inquiries without itself having to undertake enforcement action.35 More recent inquiries36 concerning leased lines or mobile roaming proceeded in the same fashion.37 Furthermore, mNewscorp/Telepiú, a merger decision at the beginning of 2003, the Commission called upon the Italian Communications Authority to supervise the proper implementation of the commitments entered into by the parties as a condition for approval of their merger.38 CONCLUSION The coordination of European and Member State regulatory activities is a comclex issue, which has only begun to be addressed. This contribution Coordination of European and Member State Regulatory Policy 177 attempted to structure the analysis by distinguishing between horizontal (e.g. between NCAs or between NRAs), vertical (e.g. between NCAs and the Commission) and transversal (e.g. between the NCAs and. NRAs) aspects of coordination. This distinction has proven useful, since on a number of points, the conclusion reached with respect to transversal aspects differed from that reached with respect to horizontal and vertical aspects. The latter two also tended to differ in some respects. In the academic and policy discussion so far, a number of issues are in risk of being forgotten. First of all, the need for coordination is often simply assumed, without any question being asked. However, it is not established as a general proposition that coordination between authorities is always desirable. In the horizontal and vertical dimension, in any event, the case for regulatory competition should be studied more closely. In the transversal, dimension, on the other hand, the case for coordination is much clearer. Secondly, primary EC law should not be discounted. Sometimes, the detailed provisions of secondary EC law hide die general principles of primary EC law, which always remain valid. For instance, even if secondary EC law instruments tend to contain provisions designed to ensure that the rights of parties (especially the defendants) are respected, primary EC law also required that the substantive rights given by EC law to complainants are effectively implemented (effet utile) in national law. While the requirements of secondary EC law might tend to produce lengthy or complex procedures, primary EC law puts a brake on this trend. Thirdly, using the substance of EC competition law as a tool for coordination, via soft-law instruments issued by the Commission, is a risky strategy. Such instruments will often not be 'battle-tested', and accordingly the legitimacy of competition law - as a relatively neutral policy framework evolving through major cases (as opposed to bureaucratic design) - could be endangered. Leaving these issues aside, the procedural and institutional aspects of the 'new' EC competition law and the recent regulatory framework for electronic communications evidence some progress in finding appropriate solutions to coordination issues. The introduction of networks, such as the European Competition Network or the ERG, could provide the institutional vehicle to reach the equilibrium between coordination and regulatory competition amongst national authorities. In both cases, however, the creation of the network (horizontal dimension) is offset by the 'hard' powers left with the Commission to enforce its position as against national authorities (vertical dimension). Accordingly, only after a few years of practical experience will one be in a position to reach a conclusion on this point. In comparison, the mechanisms put in place to conduct transversal coordination - which in theory is more desirable than horizontal or vertical coordination - still leave a lot to be desired. 178 Regulation through Agencies in the EU NOTES i See Germany v. Parliament (Tobacco Advertising) (Case C-376/98), EC.fi, 2000,1-8419 and subsequent case law. 2 The same can be said of oilier regulated sectors, such as utilities (energy, post), transport of financial services. 3 Which provided that once there was an effect on trade between Member States, 'Community competition law shall apply to the exclusion of national competition laws'. 4 Guidelines on the application of Art. SI (3) of the EC Treaty, O.J. C 101 of 27 April 2004, p. 97. 5 See Commission guidelines on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services, OJ. C 165 of U July 2002, p. 6. '' See Recommendation 2003/311 of 11 February 2003 on relevant product and service markets wilhin the electronic communications sector susceptible to ex ante regulation in accordance with Directive 2002/21, OX L 114 of 8 May 2003, p. 45. 7 It is too early to tell whether the new enforcement system of Regulation 1 /2003 will jeopardize this position, see infra. * Often, Commission notices or guidelines will precisely lead parties to avoid litigation and to adopt the solution set out therein instead. v See supra, note 7. 10 See Regulation 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, O.J. L 1 of 4 January 2003, p. 1, Art. 11,15. 11 Note that, for constitutional reasons, no similar power exists with respect to national courts. 12 Mentioned in Regulation 1/2003, supra note 10 and created in October 2002. 13 It can be noted that national courts are left out of this network, here as well for obvious constitutional reasons. 14 European Parliament and Council Directive 2002/21 of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive), O..I. L 108 of 24 April 2002, p. 33, Art. 7. 15 Directive 2002/21, ibid., Art. 7(4). This veto right concerns draft NRA measures that would stray from the Commission's recommendation concerning relevant markets or that would designate SMP operators (or abstain from doing so). See also Recommendation 2003/561 of 23 July 2003 on notifications, time limits and consultations provided for in Article 7 of Directive 2002/2 I/EC of the European Parliament and of the Council on a common regulatory framework for electronic communications networks and services, O.J. L 3 90 of 30 July 2003, p. 13. The Commission recently used this right for the first time to require the Finnish communications authority (F1CORA) to withdraw a draft measure. Ie Directive 2002/21, ibid., Rec. 36. The ERG was formally created through Decision 2002/627 of 29 July 2002 establishing the European Regulators Group for Electronic Communications Networks and Services, O.J. L 200, p. 38. 17 The reform of EC competition law is meant to decentralize enforcement, while the new electronic communications framework rather amounts to a re-balancing, with the Commission getting more power over the actions of NRAs. iK The core of EC competition law are directly effective EC Treaty provisions, while electronic communications regulation is secondary EC law, which must be implemented in national law. 19 This network can also have some noteworthy positive effects on the independence of its members as against both national administrations and market players (capture). By acting together, the network members see their positions on their respective national scenes strengthened. 20 Regulation 1/2003, see note 10, Art. 3 (2) for competition law, and for electronic communications regulation, the Obligation to heed EC competition law, which given the substantive alignment with competition law can also amount to a rule of precedence in favour of decisions taken at EC level. 21 The situation is different for national courts, as discussed above. 22 Directive 2002/21, see supra note 14, Art. 3(4). 23 Ibid., Art. 3 (5). This provision is bizarrely worded, given that the NCA is not entrusted with the app] ication of electronic communications, so that it is difficult to see why the NRA would d isclose Coordination of European and Member State Regulatory Policy 179 information to the NCA for that purpose. 24 Indeed this is the point of view of the Commission in the Notice of 22 August 1998 on the application of competition rules to access agreements in the telecommunications sector - framework, relevant markets and principles, O.J. C 265, p, 2, para. 11 and ff. 25 Unless sector-specific regulation so differs from competition law on the issue at stake, in that case that the NCA could base its decision on reasons that would not contradict the judgment that invalidated the NRA decision. In view of the trend towards the alignment of regulation and competition law, this appears unlikely. 2fi The case of mobile termination tariffs in the Netherlands offers a perfect example of this latter situation. 11 This principle is central to the ECJ case law, and it flows directly from primary EC law (amongst outers. Art. 10 of die EC Treaty). n Such as the administration of the universal service regime and the tasks relating to scarce resources. ?-'} Supra, note 6. 10 Supra, note 7. 31 Within the Commission, it is also significant to note that these two documents were elaborated by, or under the close supervision of, DG COMP. 32 See Directive 2002/21, supra note 14, Art. 7 (4), which gives the Commission a right to veto draft NRA measures which would stray from the Recommendation on relevant markets or which designate (or fail to designate) SMP operators. Here as well, it should be noted that the treatment of notifications concerning these two issues is primarily entrusted to DG COMP and not DGINFSO, within the joint task force set up by the two DGs. *3 See the ERG Common position on the approach to appropriate remedies in the new regulatory framework, available at http://erg.eu.int, which attempts to map out the exercise of that discretion. 34 Draft NRA decisions concerning remedies fall outside of Art. 7 (4) of Directive 2002/21, supra note 14 and are therefore subject to the consultative procedure of Arts. 7 (3) and (5) only. 35 See 'Commission sees substantial progress in its investigation into international telephone prices', Press Release IP/99/279 (29 April 1999) and 'Commission successfully closes investigation into mobile and fixed telephony prices following significant reductions throughout the EU' Press Release IP/99/298 (4 May 1999). 36 These were sector inquiries conducted pursuant to the Commission's competition law powers, formerly Art. 12 of Regulation 17, O.J. of 21 February 1962, Art. 17 of Regulation 1/20O3, supra note 10. 37 See the statements contained on the DG COMP's website at http://europa.eu.int/comm/ competition/antitnist/others/sector_inquiries/Iocal_loop/ and http://europa.eu.int/comm/competition/ antitrust/others/sector_inquiries/roaming/respectively. 38 Decision of 2 April 2003, Case COMP/M.2876, Newscorp/Telepiú, OJ. L 110 of 16 April 2004, pp. 73-125.