52001DC0745

GREEN PAPER on the Review of Council Regulation (EEC) No 4064/89

(presented by the Commission)

TABLE OF CONTENTS

GREEN PAPER on the Review of Council Regulation (EEC) No 4064/89

GREEN PAPER on the Review of Council Regulation (EEC) No 4064/89

EXECUTIVE SUMMARY

I. INTRODUCTION

II. JURISDICTIONAL ISSUES

A. Community dimension

1. Background

2. Summary of Issues to be Addressed

3. Assessment of the conceivable options for amending Article 1(3)

a) Alterations to the five constituting criteria of Article 1(3) are unlikely to provide any significant improvement

b) No modification of Article 22 likely to provide any significant improvement (and would require harmonisation of national laws)

c) Automatic Community Competence over Cases Subject to Multiple Filing Requirements

d) Conclusion on the conceivable options for amending Article 1(3)

4. Case Allocation in an Enlarged Community - a Long-term Perspective

B. Referrals to Member States, Article 9

1. Background

2. Summary of Issues to be Addressed

3. Possible amendments

a) Simpler Criteria for Referring a Case:

b) Referral at the Commission's initiative :

c) Timing

C. Joint Referrals to the Commission, Article 22(3)

1. Procedural Weaknesses of Article 22(3)

2. Operational Weaknesses of Article 22(3)

3. Conclusion on Article 22(3)

D. The concept of "concentration": relevant issues for review

1. Minority Shareholdings

2. Strategic Alliances

3. Article 2(4)

4. Partial-function Production Joint Ventures

5. Multiple Transactions

6. Venture Capital Investments, Article 3(5)

7. Convergence - "Control" vs. "Group"

8. Summary of Conclusions

III. SUBSTANTIVE ISSUES

A. The substantive test

B. Merger-specific efficiencies

C. Simplified Procedure

IV. PROCEDURAL ISSUES

A. Notification - Triggering event

B. Suspension of Concentrations

C. Calculation of time limits

D. Administrative efficiency

E. Completeness of Notification

F. Commitments Procedure

1. Amending the Applicable Procedure

G. Article 8(4)

H. Enforcement Provisions

I. Filing Fees

J. Due Process and "Checks and Balances"

V. INVITATION TO COMMENT

ANNEXES

GREEN PAPER

on the Review of Council Regulation (EEC) No 4064/89

EXECUTIVE SUMMARY

The proposed revision of the Merger Regulation is undertaken with a view to ensuring that the Merger Regulation will continue to be an effective instrument of merger control in a changing economic and political environment, both in Europe and worldwide. While building upon the successful application of the Regulation over the past 10 years, the proposals also seek to meet the challenges posed by global mergers, monetary union, market integration, enlargement and the need to cooperate with other jurisdictions. The Commission believes that the reform should be built on the principles underlying the Merger Regulation, i.e. the need to ensure effective, efficient, fair and transparent control of concentrations at the most appropriate level, in accordance with the principle of subsidiarity.

Jurisdictional issues

Community dimension

The Commission's review of the functioning of the Merger Regulation has revealed that the threshold levels in Article 1 (2) as well as the 2/3 rule function satisfactorily. However, as regards Article 1(3), while its main underlying principle (namely, that the Commission should deal with cases that affect three or more Member States) remains valid, the jurisdictional criteria contained in the provision have not solved the multiple filing problem which they were designed to tackle. As a result, cases of Community interest still remain subject to notification and scrutiny in a number of EU jurisdictions. Of particular concern is the trend indicating an increase in multiple filings to three or more Member States, and the fact that enlargement of the Community must be expected to exacerbate this trend. The Commission's proposal therefore focuses on the amendment of Article 1 (3), which should be operational before the enlargement of the Community.

Having analysed various numerical modifications to the combination of thresholds and other requirements of Article 1 (3), the Commission invites comments on the possibility of introducing an automatic Community competence over cases subject to multiple filing requirements to three or more Member States.

Articles 9 and 22 on referrals maintain their importance as guarantors of a degree of flexibility in the allocation of competence in EU merger control. The Commission investigation revealed a generally positive attitude towards a simplification of the test used in order to formulate a referral request under Article 9, for instance by disjoining the request from evidence of a threat of creation or strengthening of dominant position and/or by eliminating the need to establish that the requested market is not a "substantial" part of the common market. Such an amendment would allow cases whose effects do not extend beyond national borders to be more readily referred, thereby expediting procedures. For consistency of argument, the paper proposes the possibility for the Commission to refer cases, on its own initiative, to Member States when the established criteria are met. The Commission also proposes that the provisions in Article 22 on referral of cases to it by Member States should be modified along the same lines.

The concept of concentration

In recognition of experience gained as well as evolutions in business practices, the paper explores a number of potential adjustments to the concept of concentration set out in Article 3 of the Merger Regulation. With regard to minority shareholdings and strategic alliances, while acknowledging the potential structural effects of such transactions, the paper describes the difficulties in drawing borderlines with sufficient legal certainty. On Article 2 (4), concerning full-function cooperative JVs, it is concluded at this stage that more experience should be gained before considering an amendment, while it appears that, for the time being, there is no convincing argument for extending the Merger Regulation to partial function production JVs. The paper proposes to amend current provisions on multiple transactions in order to ensure more consistent and effective application of the merger control rules to three specific types of such transactions and also raises questions as to the applicability of the Merger Regulation to certain types of venture capital transactions. Finally, this section considers whether the group concept of Article 5 (4) should be harmonised with the concept of control in Article 3 (3).

Substantive issues

The substantive test

The Green Paper launches a debate on the respective merits of the "dominance test" as laid down in the Merger Regulation and of the "significant lessening of competition test" used in certain other jurisdictions. The value and effectiveness of the dominance test is recognised and it is pointed out that international convergence is already occurring to a considerable extent, independent of any actual harmonisation of laws. It could be argued, however, that a review of the Regulation text in this respect may be desirable, mainly to ensure that major jurisdictions will be working on the basis of the same test or to create a more explicit basis in the law for assessing efficiency considerations.

Simplified procedure

The Green Paper advances proposals which would take stock of the successful and smooth application of the Commission's 2000 Notice on simplified procedure. It is suggested that this practice might be consolidated into a "block exemption" regulation, with the advantage of further removing regulatory burden from transactions that are harmless to competition and of focusing available regulatory resources on those cases that require more thorough attention.

Procedural issues

Commitments

In the field of procedure, the most important proposal concerns a reorganisation of the time schedule for the submission and discussion of commitments in the first and second phases of investigation, with a view to allowing more time for all involved to make considered contributions. Specifically, the paper proposes a stop-the-clock provision. This would operate at the parties' request, thereby avoiding any ex officio prolongation of the procedure.

Due process

The paper moreover sets out the various mechanisms of relevance to due process in merger proceedings, to which the new rules on treatment of commitments would be added. In this context, it also addresses the question of the effectiveness of judicial review and invites comments on all these aspects.

Other procedural issues

Other issues with a bearing upon procedure are further considered. In specific, the Green Paper discusses current practices concerning the notification triggering event and the timing thereof as well as the "standstill obligation" (Article 7). Still on the subject of notification, the paper invites comments on electronic filing, submission of notification copies directly to Member States by the parties themselves, filing fees and the modalities of the declaration of incompleteness. The debate also embraces certain aspects of the application of Article 8 (4), and a reconsideration of the enforcement procedures in the light of the "modernisation" proposal concerning the implementation of Regulation 17. Finally, the paper considers the introduction of working days for the calculation of deadlines.

Comments are invited

Interested parties are invited to submit their views on any of the issues raised in the Green Paper, as well as on other matters of relevance to the implementation of merger control in Europe. Such comments should reach the Commission no later than 31 March 2002.

I. INTRODUCTION

1. Council Regulation 4064/89 on the control of concentrations between undertakings ("the Merger Regulation") was adopted on 21 December 1989 and entered into force on 21 September 1990. The Merger Regulation sets out rules applicable to large concentrations, the market impact of which is presumed to go beyond any single Member State.

2. Over its first 10 years of application, the Merger Regulation has received much acclaim for its ability to provide effective merger control at the European level. The introduction of short, strict legal deadlines was and remains one of the key aspects of the Regulation. The Commission believes that the Regulation continues to form an appropriate regulatory approach towards the continuing integration of markets across Europe, and in particular that it is capable of constant adaptation to new economic, legal and political developments - both within Europe and more widely.

3. The Merger Regulation was based on an understanding that the establishment of the internal market would lead to major cross-border corporate reorganisation, and that a level playing field was necessary to ensure that such transactions would not result in lasting damage to competition. The level playing field should ensure that the same notification requirements, procedure and legal standards apply to all concentrations with significant cross-border effects.

4. For this purpose the Merger Regulation conferred exclusive jurisdiction on the Commission over "concentrations with a Community dimension". This "one-stop shop" principle serves a dual purpose. First, in the spirit of subsidiarity, it builds on the realisation that merger control at Community level is justified in view of the inability of any single Member State to deal comprehensively with the cross-border scale and effects of such transactions. In addition, the single "stop" of the Regulation simplifies administrative procedures, thereby enabling both competition authorities and companies to minimise the costs of merger control.

5. Since the adoption of the Merger Regulation, the European Union has expanded from 12 to 15 Member States, whose markets are becoming increasingly integrated. Moreover, following the entry into force of the 1994 Agreement on the European Economic Area (EEA), the European Commission's exclusive competence for concentrations meeting the thresholds has been extended to cover the whole EEA territory. European cross-border corporate reorganisation has received further impetus from the introduction of Monetary Union in 1999.

6. We are now facing the prospect of a European Union with significantly more Member States as of 2004, and the completion of the monetary union with the final phase of the introduction of the Euro due to take place in 2002. At the same time the trend towards internationalisation, or even globalisation, of companies and markets is continuing at an increasing pace.

7. In a parallel development, an ever-greater number of pre-merger control regimes are being introduced across the globe (currently, about 60), with a consequent increase in the costs associated with multiple filing requirements.

8. Since 1990, some 1800 concentrations have been notified to the Commission (by July 2001). Still, that figure represents only a fraction of the total number of transactions that were notified to the EU's national competition authorities during the same period [1]. Since 1990, many Member States as well as Candidate countries have introduced merger control rules (or amended previous existing ones). As a consequence many national merger control systems largely follow the principles of the Merger Regulation. Despite this "soft harmonisation", various degrees of discrepancy persist, notably in terms of procedure.

[1] In 2000, a total of 3021 notifications were made to the EU's national competition authorities, compared to 345 cases notified to the Commission in that year.

9. All of these factors point to the desirability of revising the overall system of European merger control, so that the Commission and each national authority, individually and together, can utilise their resources in the optimal way for protecting competition in the Community, while at the same time reducing any unnecessary burden on industry, in terms of compliance costs and increasing legal certainty. This is particularly true for medium-sized companies which, owing to their limited size, fail to meet the current thresholds of the Merger Regulation, but who still remain subject to the burden of multiple national filings.

10. The experience gained over the first ten years of applying the Merger Regulation also indicates that certain improvements may be possible in both the scope and the functioning of the Regulation. As regards its scope, one of the basic concepts, namely that of a "concentration" has been re-assessed from a number of different perspectives. Similarly, the procedural setting of the Regulation has been re-examined and a number of options appear to exist for further improving both the effectiveness of the rules and legal certainty for those concerned by them.

11. Since its adoption in 1989, the Merger Regulation has been based on a principle by which concentrations are appraised in accordance with a criterion of market dominance: concentrations that create or strengthen such dominance should be declared incompatible with the common market (and therefore unlawful). Overall, this appraisal criterion has proved effective in averting potential competition problems. Indeed, many other jurisdictions have followed suit, introducing the same appraisal criterion into their rules. The main alternative merger control test, used in a number of major jurisdictions and currently contemplated in some Member States, is that of substantial lessening of competition. According to certain commentators, the lack of global harmonisation of the appraisal criteria for mergers is problematic. Therefore, and as significant experience has now been acquired in applying the dominance criterion, the Commission considers this an appropriate opportunity to open up a wider debate on the respective merits of the two tests, particularly in view of the acknowledged desirability of striving towards a greater degree of global convergence in merger control standards.

12. In this context it is also appropriate to re-examine provisions in the Regulation that have a bearing on due process. This involves a wider-ranging number of issues, relevant to the investigation, the adoption of decisions, and the right to judicial review. It is moreover recognised that ensuring transparency at each step of the procedure is of paramount importance.

13. This review of the Merger Regulation complements the previously launched review of Regulation 17 (on the implementation of the anti-trust rules in Articles 81 and 82 of the Treaty) [2]. There is nonetheless a fundamental difference between the two systems, namely the fact that the Commission has exclusive competence over concentrations with a Community dimension, while competence is shared between the Commission and the Member States in the field of anti-trust; indeed, the Commission is proposing to strengthen this parallel competence. Notwithstanding this difference, due consideration is given to the review of Regulation 17, allowing the two review exercises to form part of a comprehensive modernisation of the European legislative framework in the field of competition.

[2] See Commission proposal of 27.9.2000 for a Council Regulation on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty and amending Regulations (EEC) No 1017/68, (EEC) No 2988/74, (EEC) No 4056/86 and (EEC) No 3975/87, COM(2000) 582 final

14. In conclusion, it is for the above reasons necessary, in order both to take stock of the accumulated experience and to prepare European merger control for the challenges of the coming years, to initiate a reform process. The Commission believes that the reform should be built on the principles underlying the Merger Regulation, i.e. the need to ensure effective, efficient, fair and transparent control of concentrations at the most appropriate level.

II. JURISDICTIONAL ISSUES

CONCENTRATION WITH COMMUNITY DIMENSION

15. The Merger Regulation gives the Commission sole competence to deal with concentrations with a Community dimension. The Regulation presumes that such concentrations have a market impact going beyond the national borders of any one Member State (Recital 9).

16. The allocation of cases between the Community and the Member States in the area of merger control was thus inspired by the same principles that underpin the notion of subsidiarity. According to this notion, action should be taken at the most appropriate level of jurisdiction, in view of the objectives to be attained and the means available to the Community and the Member States [3].

[3] Article 5 EC Treaty requires that action be carried out at the level at which objectives can be most sufficiently achieved. As specified in Commission Report to the European Council on the Adaptation of Community Legislation to the Subsidiarity Principle (COM(93) 545 final, 24.11.93), action should be taken at the mostly appropriate level of jurisdiction, in view of the objectives to be attained and the means available to the Community and Member States.

17. Experience has confirmed that the Commission generally is better placed to deal with concentrations where the merging parties are active on markets that are wider than national. For cases involving global or European wide markets, the Commission's powers of investigation and remedial and enforcement action are clearly more appropriate than the more limited means available to the Member States. The same often also holds true for cases which generate competition concern spreading across various Member States. Such cases may have significant competitive repercussions in other parts of the Community, for example by impeding entry by competitors from other Member States into the affected countries.

18. A particular feature of the creation of the internal market within the EU is that a number of markets may be seen as being in a transitional state. Pirelli-BICC [4], illustrates the result of a process where European deregulation and harmonisation effectively led to a widening of the relevant anti-trust markets. In this case the market was indeed confirmed to be EEA-wide on the basis that customers increasingly source power cables at a European level on the basis, inter alia, of the procedures provided for in the Community public procurement directives. The Pirelli case contrasted with earlier cases involving the same industry: in 1992, the same markets were still considered to be national in scope. The Commission is normally particularly well placed to handle cases where such a transition is occurring, as it is able to fully consider all effects at both national and European levels.

[4] Case n° COMP/M.1882 -Pirelli-BICC. Commision decision of 19.07.2000 under Article 8(2) of Regulation (EEC) n° 4064/89.

19. Although each Member State may be able to address issues such as these in so far as they relate to its own territory, the Commission is clearly better placed to address the totality of such concerns throughout the Community. Community competence in such cases would therefore be in line with the subsidiarity principle as it represents the most effective means of avoiding the creation of obstacles to further European integration. In addition, reasons of efficiency militate against parallel treatment of the same issues in a number of Member States.

20. Conversely, Member States are generally better placed to deal with cases which, owing to the activities of the companies involved, produce their effects within a Member State, that is where for instance no significant foreclosure or spillover effects occur. The effective treatment of such cases does not require access to the broader powers of the Commission, and the national authorities are normally better placed when it comes to gathering the relevant market information.

A. Community dimension

1. Background

21. In its Report of 28 June 2000 to the Council on the application of the Merger Regulation thresholds ("the 2000 Report") [5], the Commission outlined its preliminary findings concerning the functioning of the thresholds in Article 1 of the Merger Regulation, including "multiple filings", i.e. concentrations notified in at least two Member States. Following the adoption of that Report, the Commission has undertaken additional studies. The results of these further investigations are set out and discussed in detail in Annex I, which provides a comprehensive background to the conclusions set out in this section.

[5] COM(2000) 399 final - 28.06.2000

2. Summary of Issues to be Addressed

22. In the course of its review, the Commission has re-assessed the continued appropriateness of the turnover thresholds in Article 1 in the light, not only of their effectiveness, but also of their predictability and transparency. It should be pointed out that failure to attain the jurisdictional criteria in Article 1 normally means that the transaction will be subject to mandatory notification requirements in one or more Member States.

23. As to the functioning of the current thresholds, the review has not, as set out in further detail in Annex 1, indicated any urgent need to amend Article 1(2). This Article continues to provide results that are generally in line with subsidiarity, and a revision of the threshold levels set out therein would not be well suited to dealing with the multiple filing problem. Nor has the review brought to light any urgent need to modify the level of the 2/3-rule. The rule, which adopts a centre of gravity approach to the division of competence, appears to provide results that are generally in line with subsidiarity. Furthermore, to the limited extent that the 2/3-rule leads to multiple filings, it is very rare for this to go beyond notifications to two national authorities.

24. On the other hand, it must be concluded that Article 1(3) has fallen short of achieving its underlying objective. At the time of its adoption in 1997, this provision was intended to confer Commission competence over cases that affect three or more Member States. While the review broadly supports the validity of this objective on the grounds that these are generally cases with a Community interest, the provision has not lived up to expectations. Only a small proportion of such cases have fallen within the Merger Regulation. In 2000, only 20 cases were notified under Article 1(3), compared to 75 multiple notifications to three or more Member States. The failure of Article 1(3) is also shown by the fact that in 2000 only some 5% of all notifications were made under this Article, representing a significant decrease from the already low level in 1999 [6].

[6] In 2000, only 20 out of 345 notifications were made under Article 1(3). For 1999, the corresponding figure was 12% or 34 out of 292 notifications.

25. The review has provided indications that many of those multiple filing cases involved cross-border interests. One such indication in itself is the fact that three or more national notifications were required. Furthermore, in a majority of these cases, some or all of the Member States involved assessed and/or defined the relevant geographic market as wider than national. It is also notable that the majority of these cases satisfied the jurisdictional test in so far as it relates to the Community level, while remaining far from having a national centre of gravity (i.e. the 2/3-rule). The fact that many of these concentrations involved cross-border interests beyond the countries where notification took place is also supported by the fact that many of these cases involved business activities in several Member States where the concentration was not notified.

26. From the viewpoint of the companies involved, the review has also provided indications that multiple filings in three or more Member States give rise to a number of additional difficulties, in terms of length of process, costs and legal certainty (see Annex 1 for details of the results of the studies).

27. This conclusion is further strengthened by two additional factors. First, a trend may be observed according to which multiple notifications involving three or more Member States are steadily increasing, both in absolute terms and in relation to the number of cases notified under the Merger Regulation. This shows that this problem is growing, even within the current EU15. The approaching enlargement of the Community as from 2004 is likely to further accentuate the negative effects of multiple filings involving a significant number of Member States.

28. The Commission is aware of, and fully supports, the recent initiative by the competition authorities of the Member States for closer cooperation in dealing, inter alia, with merger cases that are notified in more than one country. Such cooperation should certainly have a beneficial impact in terms of strengthening the level of protection of competition, while at the same time increasing procedural efficiency. It nevertheless appears doubtful whether such cooperation, even if significantly developed, could be seen as an equivalent substitute for one-stop shop control of mergers with cross-border effects.

3. Assessment of the conceivable options for amending Article 1(3)

29. The solution to the above identified weaknesses of Article 1(3) should aim at ensuring effective application of Community competition rules to cases with a cross-border interest, while, in a balanced way, reducing the administrative burden for the involved companies. This solution, which should be in place prior to the accession of new Member States, should moreover be designed in such a way as to strengthen the underlying principles of the Merger Regulation, namely furthering effective protection of competition in line with the subsidiarity principle.

30. In addressing the identified weaknesses of Article 1(3), a reasonable approach would appear to be to evaluate whether its effectiveness could be improved by altering one or more of its five constitutive criteria. A careful analysis of the available information indicates that modifications to the current criteria in Article 1(3) are unlikely to provide the envisaged results, principally because the complexity of this provision allows for a significant diversity of reasons why companies fail to meet the current criteria of Article 1(3). This is explained in more detail below.

31. Further, it has been suggested to the Commission that it may not be necessary to adapt Article 1(3) at all, but that the desired effect may be attainable through modifications to Article 22. This suggestion has also been carefully analysed, and, again, the results of the review, which are further elaborated below, show that this suggestion would be unlikely to provide a workable solution.

32. At this stage, the Commission is therefore of the opinion that the desired results could best be achieved through the introduction of amendments that might be described as effects-oriented. They would seek directly to ensure a result where cases that typically involve a potentially significant cross-border effect would fall under the Commission's jurisdiction. The Commission seeks comments on the possibility of Article 1(3) being replaced by a provision whereby notification to three Member States can be seen as providing a reliable indication of the manifestly cross-border nature of a concentration [7]. This possibility is built upon the ideas underlying the Commission's proposal in its 1996 Green Paper.

[7] It may be noted that the available information does not support a proposal for a "straightforward" reduction of the worldwide thresholds or of that for Community turnover in Article 1(2). While such measures, depending on the values chosen, obviously would have the effect of extending the Commission's jurisdiction, it appears unlikely that a way could be found for such measures to specifically target those concentrations with a cross-border impact.

33. A more long-term discussion on work sharing in an enlarged Community has also been included at the end of the section below. As the measures outlined therein would require at least some minimum level of harmonisation of national merger control rules, this discussion remains outside of the current exercise.

a) Alterations to the five constituting criteria of Article 1(3) are unlikely to provide any significant improvement

i) Modification of Article 1(3) by Adjusting the Various Threshold Levels

34. It is important to note for this discussion that the threshold levels in Article 1(3) were set without having conducted a full investigation of the likely effectiveness of the changes in catching cases with Community interest. Nor does it seem likely that any useful data of this kind could be established. Turnover is and will remain a proxy for the likely impact of concentrations, but will continue to produce effects that differ according, inter alia, to the economic sector involved. Nevertheless, there is wide agreement that turnover is the best proxy for establishing which concentrations have a Community interest.

35. Against this background, it should come as no surprise that the conclusion from the Commission's studies is that there is no clear-cut answer to the question of the level at which the Article 1(3) thresholds would catch all Community interest cases. Moreover, it should be stressed that Article 1(3) presents what could be described as a matrix of provisions, which could be modified in an endless number of combinations. Thus, it is relatively straightforward to indicate the effect of, for example, an isolated lowering of the EUR 2,5 billion criterion in Article 1(3)(a) to EUR 2 billion. In theory, such an amendment could of course be coupled with any number of amendments to subparagraph (b), (c) and/or (d) of Article 1(3). It is clearly impossible (and probably not very useful), however, to attempt any overly sophisticated study of what might be the ideal level and combination of these criteria.

36. The presentation below will therefore focus on explaining the exclusionary effect of each of the criteria of Article 1(3), as concerns multiple filings which are made in three or more Member States [8].

[8] For a more detailed presentation, see Annex 1, Section B, on the results of the investigations undertaken by the Commission.

37. The data available indicate that 71% of all cases notified in three or more Member States failed to meet the EUR 2,5 billion criterion in Article 1(3)(a). The exclusionary effect would be reduced to 58%, if this criterion were to be replaced by an EUR 2 billion requirement. To illustrate further the difficulty of finding an ideal level, the following example can be provided. The Article 1(3)(a) threshold would have to be lowered to EUR 500 million or even lower in order to catch all cases that were notified in three or more Member States. Even then, however, a number of those cases would fail to satisfy sub-paragraph (b), (c) and/or (d).

38. About 40% of the cases notified in three or more Member States failed to meet the aggregate EUR 100 million criterion in at least three Member States (Article 1(3)(b)). The exclusionary effect would be reduced to 31% if this criterion were replaced by an EUR 50 million requirement.

39. For cases notified in three or more Member States, 68% failed to meet the EUR 25 million criterion in Article 1(3)(c). The exclusionary effect would be reduced to 38% if this criterion were replaced by an EUR 15 million requirement.

40. About 35% of all cases that were notified in three or more Member States failed to meet the EUR 100 million criterion in Article 1(3)(d). The exclusionary effect would be reduced to 25% if this criterion were replaced by an EUR 50 million requirement.

41. It should be stressed that the above figures must not be read cumulatively. This means that if one estimation shows that a reduction in, for example, the Article 1(3)(a) threshold were to make, say 15% of all multiple filings eligible for the Merger Regulation, it cannot be added to a modification of, say, Article 1(3)(b), which in itself would be estimated to make 10% of multiple filings eligible. In other words, both measures may be alternative ways of making some or even all cases identified by the first measure eligible.

42. Within the confines of what can be deduced from the available information, it seems unlikely that any modifications to the five criteria in Article 1(3) could be targeted to allowing multiple notifications to three or more Member States to come under the Commission's jurisdiction. If anything, the available information would indicate that a reduction in the level of turnover needed to satisfy Article 1(3)(a) and (c) would hold the greatest scope for improvement. This results from the indication that specific reductions to these criteria appear to have a higher impact on cases notified to three or more Member States. It appears that modifications of the current threshold values will nonetheless leave an inestimable number of concentrations with particularly strong Community interest (in terms of the number of affected Member States) still failing to meet the thresholds [9].

[9] At the same time, any such modifications would also most likely allow, inadvertently, a number of cases that are not subject to multiple filings in three or more Member States to come under the Commission's jurisdiction.

ii) Modification of Article 1(3) from a Three-country to a Two-country Requirement

43. One conceivable alternative to an outright lowering of the various thresholds in Article 1(3) that has been suggested would be to modify the current three-country requirement to a two-country requirement. The current requirement means that the undertakings concerned must have a combined (aggregate) turnover of at least EUR 100 million in three or more Member States, and that at least two undertakings concerned must reach an individual turnover of EUR 25 million in three of those Member States. The latter requirement, in particular, appears to be one of the most common reasons why multiple filing cases fail to meet the criteria in Article 1(3), often with the result that notifications in three or more Member States are required.

44. It is nevertheless recalled that Article 1(3) was intended not to be applicable to concentrations which would only have to be notified in two Member States. Although it is clear that the actual effect of the three-country requirement has gone beyond its original purpose, it is, however, at the same time not likely that the conversion of Article 1(3) into a "two-country requirement" would have more than a relatively limited effect in terms of reducing the number of cases subject to filing requirements in three or more Member States [10].

[10] The available information does not allow for any estimation about the likelihood that the shift to a two-country requirement under Article 1(3) (b) and (c) would bring cases that currently are subject to a single national notification under the Commission's jurisdiction.

iii) De-linking of Criteria in sub-paragraphs (b) and (c) of Article 1(3)

45. Another conceivable alternative to an outright lowering of the various thresholds in Article 1(3) would be to de-link the criteria in Article 1(3)(b), which aims at ensuring that a significant volume of aggregate business activities is present in at least three Member States, and that of Article 1(3)(c), which aims at ensuring that at least two of the undertakings concerned have a significant presence in those Member States.

46. In theory, it could be argued that the EUR 25 million requirement in Article 1(3)(c) represents a relatively modest level of turnover, and that it would be surprising if concentrations that raise a Community interest would not satisfy this requirement. In that respect, it is notable that the available information suggests that the proportion of cases meeting this requirement would increase to some 54% if it were to be de-linked from the criterion in Article 1(3)(b). Thus, while the current link between the two criteria is aimed at excluding that the entire aggregate turnover taken into account for the purposes of Article 1(3)(b) is achieved by one of the parties, this link also appears to have the effect of being overly exclusionary in certain situations.

47. It is difficult to accept that the above-mentioned 54% of excluded cases, which fulfil the three times EUR 25 million presumption, would - as a rule - nevertheless only have a de minimis cross-border effect. The fact that cases in this category frequently are subject to notification requirements in three or more Member States strengthens this impression.

48. A further sign pointing in the same direction is that the currently applicable Member State rules often require notification of cases involving less than the EUR 25 million required in Article 1(3), thereby introducing a presumption that significant effects can occur also below this level.

49. A modification that would allow a more significant proportion of the cases which can be taken to have a significant effect in at least three Member States would be to remove the Article 1(3)(b) criterion. This would allow for possible Community jurisdiction in all cases where one of the undertakings concerned, typically the acquirer, meet all criteria in Article 1(3), while the other undertaking concerned, typically the target, meets the criteria in Article 1(3)(c) and (d). The available data, however, do not reveal any conclusion as to the degree to which such a modification would have repercussions in the form of catching cases that are currently not subject to multiple filing requirements.

50. Alternatively, it would be possible to remove the current cumulative character of Article 1(3)(b) and (c), making these two criteria alternative ones instead. However, in order to avoid that all turnover required under Article 1(3)(b) is achieved by one of the undertakings concerned, additional criteria would be needed. It is unclear whether this could be achieved within the logic of Article 1 without referring to a threshold value. It would, for example, seem doubtful from a legal certainty viewpoint to simply require that at least two of the undertakings concerned are active in the three Member States where the EUR 100 million is achieved.

iv) Conclusion on the potential for alterations to the five constituting criteria of Article 1(3)

51. It results from the above that modifications to the current criteria in Article 1(3) are unlikely to provide the envisaged results. Any modifications would most likely, on the one hand, not provide a guarantee against continuing multiple filings in three or more Member States, and, on the other, most likely confer Commission jurisdiction over a number of cases that today are not subject to such multiple filings.

b) No modification of Article 22 likely to provide any significant improvement (and would require harmonisation of national laws)

52. It has been suggested that the problem of multiple filings could be addressed by amendments making joint referrals under Article 22 of the Merger Regulation a more readily available tool. As the difficulties in applying Article 22 are extensively discussed in the below section on referrals, all the arguments will not be repeated here. For the discussion on amending Article 1(3) it is sufficient to note that the possibilities appear limited for making making Article 22 a useful measure of general application to multiple notification cases.

53. Among the main reasons why Article 22 has not led to any joint referrals appear to be the remaining technical differences in national merger control procedures, notably concerning the event triggering the notification and the rules concerning timing of notifications. In order to address these issues, it is likely that some degree of harmonisation of national laws would have to be made. This clearly falls outside the scope of the current review exercise, and would in any case necessarily involve a relatively time consuming procedure.

c) Automatic Community Competence over Cases Subject to Multiple Filing Requirements

54. In order to address the problem of multiple national filings, the Commission proposed in its Green Paper of January 1996 to extend the Commission's competence to those concentrations below the thresholds that come within the jurisdiction of more than one national system.

55. The proposal was for the Commission's competence to be extended to cover multiple notification concentrations falling between the Article 1(2) thresholds and some lower thresholds, which were indicatively set at EUR 2 billion (world-wide) and 100 million (Community-wide). In short, the proposal implied that Article 1 of the Merger Regulation would be amended to provide that concentrations which do not meet the thresholds laid down in Article 1(2) have, nevertheless, a Community interest when it comes within the jurisdiction of three Member States.

56. In subsequent discussions, the Commission's proposal, however, did not achieve the required support. The main criticisms related to issues such as how to deal with the existing disparity between Member State laws in areas such as the definition of a concentration and in obligatory/voluntary notification. This disparity was found not only to raise issues of legal certainty, but also to make the Commission's assessment of whether national rules applied to a certain case overly complex.

57. Since the time of the discussions concerning the Commission's 1996 proposal, a significant harmonisation of national merger control rules has taken place on a de facto basis. This should have reduced the difficulties applying the proposed system that were identified at that stage. It therefore appears appropriate to re-open the discussion about the merits of this proposal. For that purpose the Commission would retain the basic orientation of the 1996 Green Paper, i.e. that the cases targeted by such an amendment would be those involving three or more national notifications. However, on the basis of the more detailed knowledge accumulated on the operation of Article 1(3), there appears to be no empirical basis for limiting the applicability of such a system to cases satisfying certain turnover thresholds.

i) Procedure

58. Once the Commission's competence on the basis of the multiple notification requirements had been established, the Commission's powers and the procedure to be followed would be the same as for cases falling under Article 1(2). However, in addition to the normal requirements establishing the Commission's competence, it would be necessary to determine whether the concentration met the notification thresholds of at least three Member States.

59. Based on the discussions following the 1996 Green Paper, it can be assumed that such a system would have to be based on the principle that the relevant Member States should confirm to the Commission the parties' interpretation of the national thresholds [11].

[11] Any alternative approach would imply that the Commission would have to interpret the national merger laws. To the extent that national laws are not yet fully harmonised even when it comes to technical concepts such as calculation of turnover, this would raise issues about potentially divergent interpretations of national laws by the Commission and the Member States.

60. In a system where notification to the Commission would be made obligatory for cases meeting the notification requirements in three or more Member States, the parties would certainly need to demonstrate, no later than in the formal notification, that the operation meets the notification thresholds in three Member States. There is however no reason why the parties could not provide the Commission and the relevant Member States with this information at an even earlier stage. In either case, a system would need to be established by which the Member States concerned would confirm the parties' analysis to the Commission within a set period of time. In order not to create legal uncertainty, this would necessarily have to be done within a relatively short period of time (e.g. within one or two weeks). The system could, for example, take the form of a non-opposition procedure, meaning that if the Member States concerned did not oppose the analysis of the parties within the set period, the competence of the Commission would be established.

61. In the consultation that followed the 1996 Green Paper it was argued that it may be difficult to determine quickly whether a concentration meets the national notification thresholds, especially in those Member States where other criteria than turnover are used (e.g. market-share thresholds). However, to the extent that the application of some national thresholds may be difficult, this would not seem to be aggravated by the proposal, as such an assessment would in any case have had to be done if the Merger Regulation did not apply. Moreover, as the number of Member States that apply non-turnover based criteria has been reduced since 1996, this objection may no longer be perceived to pose a similar problem. Finally, to the extent that a serious legal certainty issue were to arise in this regard, it might be addressed, for example, by providing that the one-week requirement in Article 4(1) would not apply to this category of cases.

d) Conclusion on the conceivable options for amending Article 1(3)

62. The above proposal is an effect-based model, under which the decisive criterion under Article 1(3) would no longer be directly based on reaching certain specified levels of turnover. Instead, the fact that a merger gives rise to three or more national notification requirements would be taken as a sufficient presumption of a Community interest. This model would require the establishment of a transparent and rapid method for Member States to confirm the applicability of their national rules to a given case.

63. Interested parties are encouraged to indicate their views on the outlined proposal and the way it could most usefully be pursued, as well as to comment on other possibilities that may exist. Comments are also encouraged on the most suitable procedure for establishing the applicability of the merger control rules in three or more Member States as well as on other technical or procedural matters. Finally any comments as to any perceived inconvenience in amending Article 1(3), or generally as to its desirability are welcomed.

4. Case Allocation in an Enlarged Community - a Long-term Perspective

64. The data collected for this review also suggest that a more overarching solution could be considered in the longer term. A system such as the one outlined below would clearly require amendments to Community and national rules that would go beyond the scope of the current review. This section is therefore introduced in order to commence a larger debate on the appropriate shape and form of merger control in Europe in the years following enlargement of the Community.

65. The focus of such a debate should be on how to design a system that would best serve the principle of subsidiarity, whilst allowing each competition authority to make best use of its limited resources, responding at the same time to legitimate industry requests for greater predictability and reduction of unnecessary costs.

66. Despite having provided results generally in line with subsidiarity, the jurisdictional criteria of the Merger Regulation could, in principle, be amended so as to provide a more direct and case-related test for establishing the cross-border nature of any concentration.

67. One possibility would be to create a system where the turnover thresholds are removed from the ECMR, and where the triggering event for a Commission notification would be the fact that the case is subject to multiple filing requirements in the EU. In this way, the determination of which cases can be assumed to have a Community interest would be directly connected to the fact that the case will have an impact in several Member States. One of the key conditions necessary for such a system to function would, however, be that further harmonisation of national merger control thresholds could be achieved. The Commission's surveys indicate that the currently non-harmonised notification requirements in Member States do not permit reliable identification of the Member States in which a particular concentration will have a significant impact. A more systematic approach to the setting of notification thresholds in national merger laws could, however, allow these thresholds to serve as direct measurements of the cross-border impact of individual concentrations.

68. In order to develop a network approach to merger control in Europe it would also be useful to strengthen the existing degree of de facto alignment of merger control rules within the Community. This applies in particular to certain key-issues such as the concept of a concentration and important parts of the procedural framework. Another topic for consideration would be the extent to which further harmonisation of the applicable competition test would be beneficial to effective and transparent protection of competition and the maintenance of a level playing field. Such a harmonised environment would also be beneficial to the creation of a more seamless network of competition authorities, where ultimately the Commission and one or more national authority could share the task of assessing the totality of impacts from a concentration on anything from global to local markets.

B. Referrals to Member States, Article 9

1. Background

69. The referral mechanism set out in Article 9 of the Merger Regulation is intended to fine-tune the effects of the turnover-based system of thresholds for establishing jurisdiction, allowing the best placed authority to deal with the case. This is done by means of a test oriented towards the concentration's effects upon competition in a market within a Member State. If the test is satisfied, concentrations may, even if the thresholds in Article 1 are fulfilled, be dealt with at national level. This arrangement is in the spirit of the subsidiarity principle.

70. Article 9(10) states that referral provisions may be re-examined at the same time as the thresholds referred to in Article 1. Following the Commission's report in June 2000, Member States supported such a re-examination and also suggested that the Commission further investigate potential additional costs entailed by referrals.

71. The review has looked at cases in which merger notifications were the subject of referral requests over the last five years (November 1996 to January 2001). The results are set out in Annex II.

2. Summary of Issues to be Addressed

72. The results of this most recent survey, although providing little evidence that referrals entail additional costs, nevertheless point to certain aspects of the current referral mechanism that could benefit from amendment. These are, notably, the time schedule, the criteria for referring a case, and the problems caused by partial referrals. Additionally, there is some concern about the treatment of referred cases by National Competition Authorities ("NCAs"), mainly with regard to aspects of procedure. What clearly emerges from the Commission's survey is a desire for the establishment of an efficient procedure which ensures sufficient transparency between the companies involved, the Commission and Member States, which limits duplication of efforts and waste of time, and which guarantees timely and consistent decisions.

73. A recurrent remark on the time schedule concerns the three-week deadline available to NCAs for submitting a referral request. Many respondents considered the length of this period to be longer than necessary for the relevant issues to be addressed. This applies in particular where a request concerns markets that a national authority is undoubtedly better placed to deal with. Some indicated that this additional period may be harmful, as it can upset the timing of national bidding procedures. Problems could also be created by a request arriving close to the expiry of the deadline while parties may be focusing on preparing remedy proposals for discussion with the Commission. Finally, timing considerations become important in relation to the varying length of deadlines for a final decision by NCAs, in particular, in partial referrals.

74. As regards the test for the definition of concentrations to be referred (Article 9(2) of the Merger Regulation), most respondents would favour the establishment of "predictable", "simple" or even "automatic" criteria. Proposed examples include "clearly distinguishable product and geographic markets", "absence, or very low flow, of intra-Community trade", one clearly effect-oriented test such as creation or strengthening of a dominant position, or any combination of the above. Article 9(2)(b), which was introduced to facilitate referral requests by establishing a less demanding test, has only once been successfully used as a basis for a request. [12]

[12] Case Govia/Conex (2001) referred to the authorities of the UK.

75. Recurrent views among companies which have experienced partial referrals include that references diminish the benefits of the one-stop-shop principle, and that they may also have a negative impact upon timing, costs and legal certainty. This is all the more so if the assessment of various parts of an operation is referred to more than one Member State. Most companies were of the view that referrals generate additional financial and human costs resources especially after a request has been accepted, though these costs were not quantified. It should however be borne in mind, as confirmed by the survey, that companies are generally able to forecast when a case is likely to be the subject of a referral, and can factor this into their cost calculations in advance.

76. There is widespread concern about national rules, especially procedural ones, applying to referred cases. Though referred cases all appear to have been assessed by national authorities on competition grounds, uncertainty persists regarding the length of national procedures and the protection of business secrets in the treatment of referred cases. Many would like to see more alignment of national procedural rules, at least to a minimum standard.

77. A considerable number of respondents felt that the assessment of concentrations at national level could be susceptible to political influence depending on the degree of maturity of national competition law, the ownership regime of the undertakings concerned and, finally, the political or social importance of the relevant sector for the Member State in question.

78. Once the case has been referred, the Merger Regulation imposes two obligations on Member States. First, Article 9(6) obliges the NCA to report its findings within a maximum period of four months. It has been suggested that the current wording of this rule leaves scope for NCA's to use more time for the national review than would have been the case had the case remained with the Commission, and that this has occurred. Second, Article 9(8) requires Member States to take only the measures strictly necessary to safeguard or restore effective competition in the market concerned. While it is left to NCAs' discretion to decide how to comply with the latter Article, it does introduce an obligation to adhere to the principle of proportionality. As the Member States' use of this discretion is subject to the control of national courts, there would not, as has been suggested by some industry representatives, appear to be any need to amend this provision by introducing some form of supervision by the Commission. Two recent cases provide examples of the role of national courts in referral cases [13].

[13] Case COMP/M.2216 ENEL/FT/Wind/Infostrada, referred to Italy on 19 January 2001, decision appealed before the competent national court; Case COMP/M.2044, Interbrew/Bass referred to the UK on 22 October 2000, decision on remedies quashed by competent national court on procedural grounds only.

3. Possible amendments

79. A simplification of the requirements for the submission of a referral request appears to be in the interests of procedural efficiency, and may allow the current timeframe for the submission and grant of a request to be shortened. The latter objective is largely dependent on the feasibility of shortening the applicable procedural requirements at the national levels. In this respect, there appears to be little or no scope for reduction of the three-week period for so long as the national authority needs to demonstrate the threat of creation or strengthening of dominance (Article 9(2)a). However, a simplification of the requirement for the submission of a referral request should reduce the need for specific investigation by the national authorities into the likely effects of a concentration before making such request. Furthermore, it would appear possible to limit the time needed for internal consultation within the administrative structure of the Member State, given the fact that a request would no longer represent a preliminary conclusion that the transaction is likely to give rise to anti-competitive effects.

80. In line with the objective of facilitating the referral of cases which, due to a lack of significant cross-border effects, would be most appropriately assessed at national level, it would be reasonable to provide the Commission with the possibility to refer such cases on its own initiative. Such a procedure would naturally involve prior consultation between the Commission and the relevant Member State authority and could spare the Member State the procedural steps needed in preparing a request. In addition to simplifying the referral procedure, such a possibility for the Commission would, in cases with sub-national effects, allow for a more level playing field across the Community. The Commission's option to transfer such cases to Member States would basically mirror the current option for Member States to refer appropriate cases to the Commission under Article 22 of the Merger Regulation.

81. In order to further increase the transparency it may be appropriate for the Commission to adopt special guidelines on the treatment of referrals under an amended system. Such guidelines could, in a form similar to the recently adopted Notice on remedies, provide explanations on the principles underlying referrals, and on how to define cases of local or regional interest, coupled with practical examples.

a) Simpler Criteria for Referring a Case:

It is proposed to maintain only Article 9(2)(b) but facilitate its use. This means to disjoin the referral request from evidence of a threat of creation or strengthening of dominance. A substantiated claim of effect on competition in a distinct market within the Member State should suffice. As regards the size and relevance of the market in question the criterion should not be the extent to which this market is a substantial part of the common market. Instead, a simpler geographic criterion should be established. For the purposes of a referral, there would not seem to be any need for even defining the geographic scope of the relevant market/s, provided that the effects do not extend beyond the Member State's borders. It should therefore be sufficient that a request establishes that the alleged effect on competition, which would justify the referral, does not extend to significant effects in terms of foreclosure, spillover on related markets of greater geographic scope or similar cross-border effects.

b) Referral at the Commission's initiative :

In order also to fully protect competition on markets within a Member State, the Commission should, in line with the principle of subsidiarity, be able to refer a case or part of a case to national authorities where the same criteria as described above are fulfilled.

c) Timing

As the need to demonstrate the creation or strengthening of dominance would be removed, the three-week period for a referral request, for the Commission's exercise of an initiative to refer, or both could be shortened to two weeks. Notifying parties wishing to enhance legal certainty could ensure that more time is available for discussions with the NCAs concerned and the Commission by engaging in early pre-notification contacts. In order to maintain the added value of this more flexible and confidential stage of the procedure, it should however not be formalised in the Regulation.

82. On a general level, it appears justifiable that notifying parties should not be put in a worse position by the fact that a case is subject to a referral. This applies to substantive issues in particular, but also to procedural issues that may have an impact on the notifying parties. On the other hand, referred cases should arguably be treated on an equal footing to all cases subject to a merger control procedure within the specific reviewing authority. Nevertheless, there may be some merit in seeking to harmonise the timeframe in which the final decision is taken. A possibility would be to clarify the current rule in Article 9(6) so that a decision of a definitive nature comparable to an Article 8 decision under the Merger Regulation would have to be adopted within the same timeframe as would have applied for the Commission. Another, more far-reaching possibility would be to provide, in the Merger Regulation, that any national authority dealing with a case that has been referred to it should do so under the procedure indicated in the Merger Regulation. Either of these amendments would be likely to require some subsequent amendments of national merger control procedures.

83. With the purpose of initiating a debate on these issues, the Commission invites interested parties to comment on how the current system for work sharing between the Commission and national competition authorities could be improved. Commentators are invited to consider the twofold aim of better safeguarding competition, even where possible negative effects are not of a cross-border nature, and of improving the efficiency of the current system. Observations as to any perceived inconvenience in such an amendment, or generally as to its desirability, are also welcome.

C. Joint Referrals to the Commission, Article 22(3)

84. Prior to its amendment through Regulation 1310/97, Article 22(3) served the purpose of allowing Member States to ask the Commission to "step into its shoes" and assess the effects of concentrations below the Article 1 thresholds. As such, Article 22(3) was originally conceived for the benefit of Member States that did not yet dispose of the appropriate tools to investigate mergers at the national level. It has been questioned whether this original purpose of Article 22(3) was compatible with the principle of subsidiarity, as - in the absence of this Article - such concentrations would not have been subject to any control mechanism at the national level.

85. However, as has been mentioned above, all Member States except Luxembourg have now introduced merger control rules. In practice, this means that the potential scope for use of Article 22(3) in its original form is very limited [14].

[14] In fact, the latest Article 22(3) referral was made in 1997 by the Dutch Authorities in Case IV/M.890 - Blokker/Toys "R" Us. In all only four cases have been referred to the Commission under this procedure, two from the Netherlands and one each from Finland and Belgium-

86. The amendments that took effect in 1998 were intended to give Article 22(3) a new function as a means of allowing two or more Member States to make joint referrals to the Commission where they felt that the Commission was better placed to act. Thus, the intention was to strengthen the application of Community competition law in cases with cross-border effects, to strengthen the one-stop shop principle and to alleviate the problem of multiple filings. In a sense this amendment can be seen as complementary to the introduction, at the same time, of the Article 1(3) thresholds, which was likewise intended to address the same issues. Unfortunately, however, the amendments in Article 22(3) have proved even less effective than has Article 1(3). In fact, the Commission has until now not received any joint request to deal with a case.

87. The Commission has asked companies that have been involved in multiple filings whether, to their knowledge, the relevant NCAs considered the possibility of a joint request for the Commission to deal with the case. Only in 6% of cases notified to three or more Member States were notifying parties made aware of the fact that the relevant NCAs considered a joint request. None of the companies notifying in two Member States were aware of any such consideration.

88. In addition, the same companies were asked whether they were aware, in the course of the multiple filing proceedings, of any contacts between the involved authorities (e.g., by requests to agree to an exchange of confidential information between the authorities). In 8% of the cases notified to three or more Member States the notifying parties were aware of some co-ordination among NCAs. The equivalent figure for companies notifying in two Member States was 2%.

1. Procedural Weaknesses of Article 22(3)

89. While it should be noted that the source of the above information is limited to the subjective impression of notifying parties responding to the questionnaire addressed to companies involved in multiple filings, it nevertheless appears quite clear that, in most multiple notification cases, the first pre-condition for a joint referral, namely timely contacts between those concerned, is not being met. For this reason and in order to strengthen the function of Article 22(3) as a remedy to the multiple filing problem, it would probably be necessary to devise a mechanism of information exchange between all concerned. Such a mechanism would necessarily have to involve the notifying parties, all Member States concerned and the Commission.

90. In a limited number of cases, the Commission has had informal contacts with NCAs, where the possibility of a joint referral has been discussed. These contacts have highlighted three additional elements, relating principally to the procedural rules contained in the second sub-paragraph of Article 22(4), that can frustrate the attempts of Member States to make a joint referral.

91. The first problem relates to the provision that a request must be made within one month of the date when the concentration was either made known to the Member State or to all Member States making the joint request, or effected. The period begins on the date of the first of these events. There is no definition of what is meant by "making a concentration known to a Member State". Although the provision would probably benefit from clarification in that respect, it appears natural to use the date of a national notification as the triggering date in Member States where a notification requirement exists. The reason for this interpretation is that it appears unreasonable to use any other, less formal announcement as a basis, where the national law contains specific requirements for how a concentration should be brought to the Member State's attention [15]. An issue more open to discussion is how "making a concentration known to a Member State" should be interpreted when the national law contains no requirements for a notification [16]. In addition, it is conceivable that a concentration could be effected on different dates in different Member States.

[15] This interpretation, on the other hand, has been criticised on efficiency grounds, as it arguably might not be necessary to request the parties to undertake the often detailed work involved in preparing several national notifications, if the case is ultimately likely to be referred to the Commission.

[16] Following the entry into force of the new French merger control system, this difficulty will only be generally applicable in the UK and in Luxembourg (but will, potentially, apply to any merger below the national merger control thresholds).

92. A second problem, partly related to the one described above, is that Member States' merger control systems differ significantly in the requirements as to when a notification should or may be made. National systems show considerable variation in this respect. In practice, this means that it is difficult to foresee when the one-month deadline in Article 22(4) will expire as concerns any of the Member States. This difficulty is then obviously multiplied by the number of Member States that may participate in the joint referral. Another difficulty in itself is that there is no objective way to predict how many Member States could participate in a joint referral. The end-result is that depending on which Member States participate in a joint referral, the deadline in Article 22(4) may differ significantly, as this in turn will depend on when the concentration was made known to each of them.

93. The third and possibly most problematic issue, which again is connected to the lack of procedural harmonisation among Member States, is that there is no provision for suspending the applicable national deadlines pending a joint referral. This means, for example, that country X may be legally bound under its national law to adopt a decision on the notified transaction before the expiry of the deadline for country Z to submit its request to the Commission. This difficulty can obviously be aggravated if there would be more than two countries considering a joint referral, as well as by the fact that the Commission, in accordance with Article 22(3), has discretion whether or not to accept a referral request.

94. In conclusion, in order to make Article 22(3) operational as a generally applicable corrective mechanism to the multiple filing problem, the system would most likely need amendment of more than just the Merger Regulation. Iit is doubtful whether the Merger Regulation could be amended so as to introduce mandatory consultation between Member States about notifications of cases with no Community interest (as defined in Article 1). The same applies to the above-described problems stemming from Member States' varying provisions concerning the triggering event for notification.

95. For these reasons it is likely that Article 22(3) could only be made operational following a sufficient degree of harmonisation of national laws. However, even if this were to be achieved, the system would still suffer from certain operational inconveniences that are set out below.

2. Operational Weaknesses of Article 22(3)

96. From an efficiency viewpoint, the Article 22 referral provision inherently implies longer overall procedures and less legal certainty for all involved. As any referral provision will, by its very nature, always involve an assessment by two or more competition authorities, it is unlikely that a system can be devised which would not involve longer waiting-periods than the immediate one-stop shop system. Moreover, as, at least in the current system, each involved competition authority has a discretion whether or not to request/accept a referral, the procedure is bound to be surrounded by yet more uncertainty. The merging parties, who may be able to use the existing differences between national laws in order to promote a specific outcome, may contribute additional uncertainty. Whereas these weaknesses may have to be accepted in a system that is designed to function as a corrective mechanism in exceptional cases, they go beyond what can be defended in a system aiming to more generally alleviate the multiple filing problem.

97. A second operational weakness of Article 22(3) is that a joint referral only gives the Commission competence to take the measures necessary to maintain or restore effective competition within the requesting Member State(s). This may be logical for a system based on the idea that the Commission in such cases will replace the requesting Member State(s). Nevertheless, it is liable to create results that are incompatible with the one-stop shop principle (as non-requesting Member States may pursue parallel procedures), or, worse, may lead to the result that the Commission is unable to address an identified competition concern (if non-requesting Member States do not investigate).

3. Conclusion on Article 22(3)

98. As has been indicated above, some of the current weaknesses of Article 22(3) could be overcome by clarification (e.g. of the expression "making the concentration known to a Member State"). In addition, as Article 22 has always had the role of a mirror image to Article 9, the same type of modification as has been outlined in respect of the latter could also be envisaged (e.g. the facilitation of the applicable test).

99. Still, in view of the above described weaknesses, it is doubtful whether this provision could be modified in such a way as to provide an efficient and general method for alleviating the multiple filing problem. Interested parties are nevertheless invited to set out their views on this, including any suggestion for making this provision part of a coherent system for the allocation of competence.

D. The concept of "concentration": relevant issues for review

100. In addition to mergers, the concept of concentration covers the acquisition by one or more companies of legal or de facto control over one or more companies, including the creation of full-function joint ventures.

101. It follows that a transaction, which does not entail an acquisition of such control falls outside the scope of the Merger Regulation. This means, for example, that transactions involving the acquisition of non-controlling joint or sole minority shareholdings are not covered by the Regulation, and this is regardless of whether the minority stake is acquired by a competitor. As long as the control criterion is not satisfied, the same principle also applies to strategic alliances, a business concept that has become more frequent since the introduction of the Merger Regulation. As a rule, the Commission currently examines such transactions under Articles 81-82 EC. The Commission is conscious of the fact that such transactions may have a "structural" impact on the markets involved. It is also conscious, however, of the difficulty in sufficiently defining the transactions in question for the purposes of mandatory ex-ante notification.

102. As noted above, the Merger Regulation applies to full-function joint ventures. In 1998 the scope of application was extended, through amendments to Article 3(2) and the introduction of Article 2(4), which enabled the assessment to include co-operative aspects of a joint venture. Experience gained during the first years of applying this new provision will be assessed below. Moreover, the 1999 White Paper on modernisation of the rules implementing Articles 81 and 82 raised the question of whether partial-function production joint ventures should also be included under the Merger Regulation. However, in its proposal of 27.9.2000, the Commission deferred further examination of this issue to the review of the Merger Regulation.

103. An issue that has on occasion given rise to difficult questions of interpretation relates to concentrations that consist of multiple transactions. Put simply, the question is: under which conditions should two or more legally separate transactions be considered, for the purposes of the Merger Regulation, as constituting a single notifiable transaction - The answer to this question can have repercussions on the allocation of competence between the Commission and national authorities. In order to guarantee effective and efficient control of concentrations at the most appropriate level, the Commission believes that it should be ensured that economically inter-dependent transactions are treated comprehensively by one and the same authority. At the same time, the system should exclude artificial combination of distinct transactions.

104. Another issue which is linked to the concept of a concentration is whether the Merger Regulation can be said to cover certain transactions, in particular in the financial sector (venture capital, etc), which in reality do not give rise to a concentration of supply or demand in any market. Article 3(5) of the Merger Regulation provides an exemption for certain types of such transactions in. However, in view of developments that have occurred since 1990 in the financial markets and in corporate financing, it is appropriate to reassess whether this Article continues to fulfill its purpose, or whether it requires revision.

105. Finally, an issue with relevance to the concept of a concentration is whether there could be more convergence between this concept and the group concept in Article 5 of the Merger Regulation. The current distinction between these two concepts has, on occasion, given rise to controversy and it is therefore desirable to assess whether utilising a common concept under both these Articles might facilitate the application of the Regulation.

1. Minority Shareholdings

106. In a limited number of cases the Commission has taken account of the divestment of minority shareholdings or the abandoning of interlocking directorships in first phase decisions, rendering unnecessary an analysis of the potentially complex issues arising from the assessment [17].

[17] See for examples IV/M.1080 - Thyssen/Krupp, Decision of 2 June 1998, COMP/M.1712 - Generali/INA, Decision of 12 January 2000 and COMP/M.1980 - Volvo/Renault, Decision of 1 September 2000.

107. The Merger Regulation, however, does not apply to acquisitions of minority shareholdings, unless, due to other factors, a situation of legal or de facto control is established. Still, it may be the case that a minority shareholding (potentially coupled with interlocking directorships) may alter the linked companies' incentives to compete and thus have an impact upon market conditions. Although the Merger Regulation is not applicable in such cases, Articles 81 and 82 EC can, according to established case law, be used to assess minority shareholdings and the competition problems they may entail.

108. The Commission's attention has been drawn to minority shareholdings and interlocking directorships as possible factors facilitating joint dominance, and the fact that a number of other jurisdictions apply merger rules to the acquisition of minority shareholdings independently of the acquisition of control [18].Some have suggested that the Community rules may benefit from a similar extension.

[18] German, Austrian and Irish merger assessment rules, for example, apply to an acquisition of a 25% shareholding, regardless of changes in control in the meaning of the Merger Regulation. In these systems, a second notification may be required if, subsequently, sole control is acquired. Similarly, the UK merger rules apply, in addition to acquisitions of control, to transactions whereby one party acquires the ability to "materially influence" another party. The US merger rules (Section 7 of the Clayton Act) also apply to situations that would not meet the definition of a concentration under the Merger Regulation. Issues relating to interlocking directorships are covered by Section 8 of the Clayton Act. The US rules employ, inter alia, a "size-of-the-transaction test", which can apply to transactions that do not involve a transfer of control (in the Merger Regulation meaning), provided that the acquired securities exceed the $50 million threshold.

109. At this stage the Commission is not in the possession of comprehensive data as to the prevalence of minority shareholdings and interlocking directorships. However, based on current experience, it appears that only a limited number of such transactions would be liable to raise competition concerns that could not be satisfactorily addressed under Articles 81 and 82 EC. Under this assumption it would appear disproportionate to subject all acquisitions of minority shareholdings to the ex ante control of the Merger Regulation. At the same time it appears doubtful whether an appropriate definition could be established capable of identifying those instances where minority shareholdings and interlocking directorships would warrant such treatment.

110. The Commission nevertheless invites interested parties to describe their experience with the treatment of minority shareholdings and interlocking directorships under Community competition rules, as well as their views on the most appropriate future treatment of such arrangements.

2. Strategic Alliances

111. Strategic alliances are co-operative arrangements of varying scope involving the creation of several links which are usually contractual but may also have structural aspects. They may range from the creation of a joint venture, to specialisation in certain markets, joint R&D, technology transfer, cross-supply arrangements, commitments to co-operate in other fields in the future, or even the acquisition of shareholdings. [19] Strategic alliances are often concluded among actual competitors who are seeking to compete with other similar alliances. They have tended to be most common in recently liberalised markets where they provide a means to compete in markets integrating on a global scale. Competition law mainly seeks to ensure independence and effective competition between such alliances.

[19] See 24th Report on Competition Policy (point 156).

112. The agreements involved in a strategic alliance (notably, exclusive licences, purchase, distribution, non-compete clauses) may be restrictive of competition and consequently may be caught by Article 81 of the Treaty. So far, only one, namely the strategic alliance between Alitalia and KLM, [20] has been assessed on the basis of the Merger Regulation. In that case the Commission considered that, for a number of reasons, [21] the alliance fulfilled the criteria of a full-function joint venture, as defined in the Commission Notice.

[20] JV.19 - KLM/Alitalia

[21] Notably the facts: that the parents would cease to operate in the markets pooled in the alliance; that they would run jointly the day to day business and would adopt jointly the main strategic and commercial decisions; that tangible assets of each of the parties would be exclusively dedicated to the Alliance; that the duration of the agreement would be long.

113. According to recital 23 of the Merger Regulation it is appropriate to define the concept of concentration in such a manner as to cover only operations bringing about a lasting change in the structure of the undertakings concerned. Strategic alliances are normally not designed to bring about the structural change envisaged in Article 3(1) of the Merger Regulation and may not necessarily result in the creation of an autonomous economic entity as this is defined in the Notice on the concept of full-function joint ventures. Article 81, therefore, still appears to be the most appropriate legal instrument for assessing such transactions.

3. Article 2(4)

114. One of the key amendments made to the Merger Regulation in 1998 consisted in the inclusion of certain joint ventures, which previously had been considered as falling outside the scope of the Regulation (so called full-function co-operative joint ventures "ffCJV's").

115. Between March 1998 and December 2000, a total of 47 ffCJV's were notified. Out of these, only one notification was based on the turnover criteria set out in Article 1(3), while the remaining 46 cases fulfilled the criteria in Article 1(2).

116. The extension of the Merger Regulation to include all ffCJV's is generally considered as having been appropriate. Experience has shown that these cases are well suited to the more structural type of assessment of the Merger Regulation. Their inclusion under the Merger Regulation has reduced the cost and delays involved in achieving regulatory clearance for the companies involved in such transactions, and Article 2(4) allows all the competition aspects of such operations to be evaluated in a single administrative procedure.

117. From a substantive viewpoint, the likelihood that ffCJV cases produce competition concerns can be illustrated as follows. Seven of the above-mentioned ffCJV cases were found to raise dominance concerns in either the first or second phase. The equivalent figure for all other cases that were notified under the Merger Regulation in the same period was 75 (out of 798 notifications). Putting these figures in comparison, dominance concerns arose in some 15% of the ffCJV cases, and in 9% of the other cases.

118. Over the same period, two ffCJV cases [22] raised issues related to the co-ordination of competitive behaviour of undertakings that remain independent (i.e. Article 81 concerns). In these cases, these issues were not further investigated because modifications brought by the parties to their initial project addressed them, rendering a full investigation unnecessary. To date, no ffCJV case has been found incompatible with the common market.

[22] Case M.1327 - NC/Canal +/CDPQ/Bank America and Case JV.15 - BT/AT&T.

119. While it appears that the type of concerns which Article 2(4) was introduced to control occur relatively infrequently, it must nevertheless be acknowledged that the period within which this provision has been in force is relatively short. More practical experience should therefore be gathered before making any more detailed evaluation.

4. Partial-function Production Joint Ventures

120. As mentioned above, the Commission's modernisation proposal for Articles 81 and 82 left the question of the possibility of a further extension of the concept of a concentration to this review. The background was that the Commission's white paper had indicated that it appeared desirable to maintain a prior authorisation system for partial-function production joint ventures. The basis for this assumption was the substantial investment and far-reaching integration of operations involved in such operations. It was envisaged that partial-function joint production ventures, as full-function joint ventures, would be subjected both to the dominance test and to Article 2(4) of the Merger Regulation.

121. The subsequent consultation procedure, however, revealed several important criticisms of the proposal. One problem is that it would be very difficult to find an unambiguous legal definition of the concept of a partial-function production joint venture, particularly in to the context of service markets. This indicates that such operations may not be suited to control under a system of mandatory ex ante notifications.

122. Even more importantly, the consultation did not produce indications that such production joint ventures are inherently more suited to ex ante control than other operations that may involve large-scale investments (e.g. R&D joint ventures or distribution systems). For example, as partial-function production joint ventures would by definition not be active on any market, the only real assessment of such cases, should they be brought under the Merger Regulation, would relate to co-ordination between the parent companies. As this means that the test to be applied would remain the same (i.e. Article 81), it is important to note that there are no indications that parties concerned have a strong preference for ex ante control under the current system for applying Article 81: it does not appear that parties to partial-function joint production ventures frequently notify under Article 81 and await the Commission's decision before implementing such transactions.

123. In addition, partial-function joint production ventures would continue to be covered by block exemption regulations under Article 81. Of particular relevance in this context are the block exemptions on research and development agreements and specialisation agreements. Regulation 2658/2000 on the application of Article 81(3) to categories of specialisation agreements [23] block exempts, inter alia, joint production agreements, whereby two or more undertakings agree to produce certain products jointly. The exemption is subject to a market share threshold of 20%. Regulation 2659/2000 on the application of Article 81(3) to categories of research and development agreements [24] block exempts joint research and development of products or processes as well as joint exploitation of the results thereof. If the parties to the agreement are competitors, the block exemption is subject to a market share threshold of 25%.

[23] OJ 2000 L 304, page 3.

[24] OJ 2000 L 304, page 7.

124. In conclusion, based on currently available information, there does not seem to be any compelling reason to extend the scope of the Merger Regulation to partial-function production joint ventures. The Commission nevertheless invites any interested party who feels that such an extension would be appropriate to describe how, in their view, such a system, incorporating sufficient standards of legal certainty, might be envisioned.

5. Multiple Transactions

125. In the past, the Commission has had to decide on a number of occasions whether a number of transactions, which are linked as to a variety of aspects, should be considered as constituting a single concentration.

126. The question is relevant in determining the jurisdiction of the Commission. A transaction that by itself might not have Community dimension because it does not meet the thresholds of the Merger Regulation, might nevertheless fall under the jurisdiction of the Merger Regulation as part of a concentration that is made up of more than one transaction.

127. Article 3 applies to the acquisition of "direct or indirect control of the whole or parts of one or more other undertakings". Apart from this general and wide definition, the only other provision directly concerned with the situation of two or more transactions constituting a single concentration is Article 5(2)(2). This provision was designed primarily to prevent the circumvention of the Merger Regulation by separating one concentration into numerous transactions. The Merger Regulation has established a legal presumption that all transactions meeting the requirements of Article 5(2)(2) are to be considered as one concentration.

128. In conformity with the one stop shop principle, there is no reason in principle why other multiple transactions, which do not appear to constitute a single concentration under the present Merger Regulation, should not be assessed as a whole if they involve an economic link revealing economic unity among the transactions in a way that is equivalent to a single concentration.

129. The concept of economic unity should be assessed against the goals pursued by the parties. The relevant factors to establish that the situation is equivalent to a single concentration necessarily include a time connection as well as an identity link, both in terms of the participants in the transactions and in the scope of the transactions.

130. The application of the one stop shop principle to the types of multiple transactions described below would better serve the overall aim of maintaining effective competition, as it would ensure that the totality of effects of such concentrations are subject to one coherent assessment. For the same reasons as apply to any other concentration, a more stringent application of the one-stop shop principle should also enhance legal certainty for those involved or affected by the concentration. In order to avoid artificial separation in the treatment of multiple transactions, which from the perspective of the parties and/or of the market are characterised by economic unity, it appears appropriate to further specify the circumstances in which multiple transactions should be taken as a whole for the purpose of the Merger Regulation. This seems to apply to the following three scenarios [25]:

[25] A possible fourth scenario would involve joint bids. These cases are already treated as a single concentration for the purposes of the Merger Regulation, unless the joint bidders have agreed an immediate break-up of all the acquired assets according to a pre-defined plan. In the latter scenario, point 24 of the Notice on undertakings concerned specifies that each such acquisition of assets will be considered as a separate concentration. As this main principle is already established, there would be no need to amend the existing legal provisions. Nevertheless, it may be appropriate to further clarify the limitations of the exemption in point 24 of the Notice on undertakings concerned in the next revision of this Notice. Such clarification would explain that, to the extent that there is any uncertainty concerning the future disposition of the jointly acquired assets, the joint bid will be considered, in its totality, as a single concentration, the assessment of which will include any known dispositions to break up the acquired assets.

131. Operations involving the acquisition of joint control of one part of an undertaking and sole control of another part. This is, for instance, normally the case when the direct acquisition concerns the parent company of a group. In such a scenario, the target company may often have one or several jointly controlled subsidiaries which, as a consequence of the direct acquisition, will become (jointly) controlled by the new owner of the parent company. In such a scenario it would clearly be artificial not to regard the acquired group as one economic unit, and as long as the situation at hand, from an economic viewpoint, is the same, there would be no reason not to apply the same general principle. This can be demonstrated by way of a simple example. Consider a situation where Company A intends to sell the following assets (which, for the sake of the argument, may be active in the same relevant market and have been run as a single entity by A) to Company B:

* 100% of its subsidiary A1,

* its 50% stake in A2 (a jointly controlled company), and

* its 25% stake in A3 (not a controlling stake).

132. Under the current rules the acquisitions of A1 and A2 would be considered as separate concentrations, whereas the acquisition of the stake in A3 would not be characterised as a concentration at all. The two first-mentioned would, depending on the turnovers involved in each of those two transactions, be subject to assessment under the Merger Regulation or by (multiple) Member State(s). The result would be completely different if A, prior to the sale, would put A1-A3 in a holding company (for example, for tax reasons) that subsequently was transferred to B. In this case there would only be one concentration under which B would acquire sole control over the holding company. Moreover, all turnover related to the assets held by that company would be taken into account for the decision on jurisdiction, which as a result would mean that the entire transfer of assets would be subject to an assessment at either Community or Member State level. In order to capture fully all the economic activities being combined through the concentration it therefore appears appropriate to amend Article 5(2)(2) of the Merger Regulation to fully extend its coverage to this type of scenario [26].

[26] Consequently changes would also be required to the Commission's interpretative Notices (see Notice on the concept of a concentration, point 16).

133. The exchange of assets (swaps) between two companies will normally have as an end result that two separate entities remain on the market (albeit with modified asset constellations). A swap will often be determined by a single contract and conclusion of each transaction will, almost by definition, be conditional upon conclusion of the other. Until now, the Merger Regulation has not recognised that such links are sufficient to treat swaps as a single concentration. However, in order to provide for a coherent assessment of the entire transaction, a change of policy appears justified. In addition, one or both swaps can be partial, in the sense that the previous owner may retain an economic interest in the swapped business. As regards the assessment, each side of a swap may certainly have an impact on the assessment of the other one. These reasons, and the fact that swaps, from the parties' viewpoint, normally represent an economic unity, would advocate in favour of treating such transactions as equivalent to single transaction concentrations. It therefore appears appropriate to amend Article 5(2)(2) of the Merger Regulation to fully extend its coverage to this type of scenario [27].

[27] Again, subsequent changes would be required to the Commission's interpretative Notices (see Notice on the concept of undertakings concerned, point 49-50)

134. "Creeping" take-overs via the stock exchange is another example of multiple transaction concentrations. Such transactions can be implemented in a number of more or less sophisticated ways, ranging from relatively straightforward direct share purchases from a number of previous shareholders to transaction structures that involve any number of financial intermediaries using a variety of financial instruments. This type of transaction will often occur in "hostile" situations, i.e. where the target company and/or some of its previous shareholders are less than fully supportive of the take-over. In such scenarios, it will normally be both impractical and artificial to consider the concentration as occuring via the acquisition of the particular share or block of shares that will put the acquirer in a situation of (de facto) control over the target company. Instead, it will normally be clear from the viewpoint of all parties involved that a number of legally separate acquisitions of rights, from an economic viewpoint, form a unity, and that the intention is to acquire control over the target company. It therefore appears appropriate to amend Article 5(2) (2) of the Merger Regulation to fully extend its coverage to this type of scenario [28].

[28] See also paragraph 0 (and following) on the relationship in these situations between Articles 4 and 7 of the Merger Regulation.

135. The Commission is of the view that the above-mentioned types of transactions could be brought within the scope of the Merger Regulation through adjustments to Article 5(2) (2). In order not to over-extend the application of the principle of Article 5(2) (2), it however appears appropriate to limit the scope of this provision to transactions relating to the same economic sector. Following such an amendment Article 5(2) would read as follows (amendments in bold):

2. By way of derogation from paragraph 1, where the concentration consists in the acquisition of parts, whether or not constituted as legal entities, of one or more undertakings, only the turnover relating to the parts which are the subject of the transaction shall be taken into account with regard to the seller or sellers.

However, two or more transactions between the same persons or undertakings which take place within a two-year period shall, unless they concern unrelated industrial sectors, and regardless of the type of control involved, be treated as one and the same concentration arising on the date of the last transaction.

The principle set out in the second subparagraph shall apply mutatis mutandis to exchanges of assets (swaps) and to multiple acquisitions of the securities, as defined in Article 7(5), second subparagraph, of an undertaking.

136. Comments are invited on this proposed amendment to Article 5(2) of the Merger Regulation. Interested parties are also invited to comment on whether, in their view, there are other types of multiple transaction concentrations that would benefit from further clarification as to their treatment under the Merger Regulation.

6. Venture Capital Investments, Article 3(5)

137. Article 3(5) describes certain narrowly defined situations when a concentration shall not be deemed to exist. One such exception relates to the normal trading activity of financial institutions. However, experience shows that the limitations of this provision (in terms of non-voting rights and as regards the maximum period for holding on to the securities) mean that it is rarely applicable. At the same time, restrictions of this type are intended to discourage circumvention of the Merger Regulation, and thus to safeguard its effectiveness as well as the level playing field.

138. The merger review has drawn the Commission's attention to the existence of new forms of financing in the equity markets, and these should be considered when reviewing the scope of the Regulation. Venture Capital ("VC") as a phenomenon has increased significantly in Europe since the adoption of the Merger Regulation. According to one distinction, VC can be divided into two main categories; transaction investments and growth capital/technology investments.

139. Transaction investments normally relate to takeovers of established, quoted or private, companies and are in many respects similar to traditional industrial transactions. They do not appear to raise any particular issues of relevance for this review.

140. Growth capital/technology investments, however, do present some characteristics that may deserve closer examination. Such investments are normally made in order to provide capital for a new business during its early development phase, with the ultimate aim of floating the company on the stock exchange in the medium term. Typically, the entrepreneur will retain full operational control over the business and the company may not even have a Board. It has been argued that this type of venture capital should be seen as equivalent to traditional loan facilities provided by banks. The main feature of relevance under the Merger Regulation is that VC investors will hold shares in the new business and typically have at least veto rights over its budget and business plan. Under traditional financing methods banks would normally not hold shares, but would instead retain a possibility to refuse renewal of loan facilities.

141. The reason why these growth capital/technology investments may fall under the Merger Regulation is that they are normally syndicated, which means that two or more VC investors will share the investment. Owing to this construction, such transactions may, even in the case of start-up businesses with no sales, be notifiable under the rules applicable to full-function joint ventures. There are several reasons why both the VC investors and the entrepreneur may prefer syndication, notably the wish of both sides to lower their exposure to risk and to combine particular experiences.

142. This type of VC investments will often relate to small and medium size enterprises and, moreover, will encourage entrepreneurial activity that in the medium to long term is likely to benefit competition, as well as the economy in general. Under the current rules, such VC investments are likely to meet the criteria for simplified treatment under the Merger Regulation (see below). However, even if it could be accepted that competition concerns are unlikely to occur in this type of VC investments, as they provide new businesses with capital during an initial stage, there are several difficulties in defining the scope of a specific exemption. One such difficulty relates to the fact that there is no common definition of a VC investor. In this respect, it is certainly clear that there is less scope for competition concerns differ if the VC investor is a purely financial company (such as a bank), than if it is a subsidiary of an industrial company [29]. Furthermore, it is clearly of relevance for the assessment of whether competition concerns may arise whether the syndicate of VC investor takes a more or less active role in the business. Competition concerns are unlikely to arise where the VC investors take a passive role and do little more than provide funding in a way that is comparable to traditional bank facilities. The same may not necessarily be the case where the VC investors, who are often sector specialists and invest in a large number of businesses in a particular sector, take a more proactive role, possibly including the transfer of assets and/or information between various investments.

[29] The European trade association for VC investors (EVCA) includes, inter alia, bank subsidiaries and fund management companies, but also so called "corporate ventures" who are subsidiaries of industrial companies. The aim of the latter category may be to invest in new technologies that will support or develop existing industrial activities.

143. A different type of issue to be resolved for any type of exemption to be useful would appear to be the relationship between such an exemption and the merger control rules in the Member States. An exemption that would have as a consequence that syndicated Growth capital/Technology investments would be subject to multiple notification requirements at the national level would not only be incompatible with the one-stop shop principle, it would add to, rather than reduce, the regulatory burden involved.

144. Notwithstanding the above-described difficulties in defining a category of VC investments which would not give rise to competition problems, the Commission remains open to the possibility of enlarging the scope of Article 3(5). It therefore invites interested parties to submit their views, and in particular any proposals as to how this might be achieved without reducing the effectiveness of the Regulation. In providing such comments, interested parties are particularly invited to consider whether the proposals set out below concerning an extension of the scope of the simplified procedure might alleviate the need to address additional issues relating to VC investments.

7. Convergence - "Control" vs. "Group"

145. Article 3(3) defines the concept of "control" for the purposes of determining the circumstances in which a concentration is deemed to arise. The test applied is qualitative rather than quantitative, and can be established on the basis of both law and fact.

146. Article 5(4) defines the concept of the group of undertakings that, in addition to the undertaking concerned, will be included for the purposes of assessing whether the turnover thresholds in Article 1 are satisfied. The purpose of this provision is to capture the total economic resources involved in a concentration.

147. In a limited number of cases some controversy has arisen as to the compatibility of Article 3(3) with 5(4). This has been caused primarily by the following two facts:

- indent 1-3 of Article 5(4)(b) states that a company shall be considered as part of a group if more than half of its capital or assets are owned by the group, if more than half of its voting rights can be exercised by the group or if the group can appoint more than half of its Board;

- indent 4 of Article 5(4)(b) states that a company shall be included in the group if the group has the right to manage its affairs.

148. The first three indents of Article 5(4) provide what can be described as quantitative form-based criteria. The fourth indent is of a more effects-based nature and is therefore more similar to the qualitative test in Article 3(3). Therefore, even if the end-result of the two types of analysis may be the same in the vast majority of cases, it is clear that there are differences between the two provisions and that such differences, may sometimes create uncertainty.

149. It is, for example, established case-law that control in the sense of Article 3(3) can be acquired by a company holding significantly less than half of the voting rights in another company (assuming that it nevertheless is likely to hold a majority at the shareholders' meeting). On the other hand, it is not certain that control in the sense of Article 3(3) would be deemed to arise even after the acquisition of, say 51% of the share capital or assets (assuming, e.g., that another shareholder controlled more than half of the votes in the company).

150. Equally, Article 5(4), including indent 4, uses language that is more compatible with a situation of sole control than with joint control between several parent companies.

151. In view of the above, the Commission invites interested parties to submit comments as to whether, in practice, the current differences between Articles 3(3) and 5(4) is a cause of concern to them and, if so, whether, in their view, it would be appropriate to also base the group concept on the principles underlying Article 3(3). Equally, the Commission would be interested in receiving comments about any perceived inconveniences that might result from harmonising the two provisions.

152. A related issue concerns the Commission's practice, based on an analogy with Article 5(5) of including a proportion of the turnover of joint ventures for the purposes of calculating the total group turnover [30]. Although the analogous application of Article 5(5) in this way has not given rise to controversy, it nevertheless appears appropriate to clarify the approach in the provision itself.

[30] See point 40 of the Commission's Notice on the calculation of turnover.

8. Summary of Conclusions

153. This chapter has considered a number of possible amendments to the concept of concentration set out in Article 3 of the Merger Regulation. It first looked at the treatment of this concept in relation to minority shareholdings and strategic alliances. While certain of these transactions might have structural effects, this does not apply as a general rule and it does not appear possible to draw this distinction with sufficient legal certainty. Article 81 therefore appears to remain the most appropriate instrument for the assessment of such transactions.

154. This chapter also considered the issues of Article 2(4) and partial function joint ventures. Regarding the former, it was concluded that, although its introduction into the Merger Regulation in 1998 has not lead to a significant number of spill-over concerns being identified, it is nonetheless consistent with the one-stop shop principle and should therefore be retained. As regards partial function joint ventures, it was concluded that there appears to be no convincing argument for extending the Merger Regulation to such operations.

155. Tthe issue of so called multiple transactions was considered, and it was concluded that the current provisions should be amended to better ensure coherent and effective application of merger control rules to three specific types of such transaction.

156. The chapter also considered whether it would be appropriate to limit the applicability of the Merger Regulation to a particular type of venture capital transaction. It was concluded that the feasibility of such a change would depend on whether such a limitation could be defined in such a way as to exclude any negative impact on the effectiveness of the Merger Regulation.

157. Finally, the chapter considered the appropriateness of harmonising the group concept in Article 5(4) with that of control in Article 3(3).

158. In addition to being invited to comment on some the above-mentioned issues, interested parties are encouraged to comment on any other issues relating to the concept of concentration that, in their view, would benefit from clarification or review.

III. SUBSTANTIVE ISSUES

A. The substantive test

159. The substantive test according to which notified concentrations are appraised is set out in Article 2 of the Merger Regulation. In the course of this review it has been suggested that the Commission should use this opportunity to discuss the merits of the dominance test contained therein. Both procedural and substantive reasons have been advanced for a re-evaluation of the appropriateness of this test.

160. From a procedural viewpoint, the main reason proposed in favour of such a re-evaluation is that it could allow an alignment of the Merger Regulation's appraisal criteria with those applied in other major jurisdictions, such as the US, Canada and Australia, which rely on a concept of substantial lessening of competition ("the SLC-test"). Such an alignment towards a global standard for merger assessment holds certain attractions. It would, for example, facilitate merging parties' global assessment of possible competition issues arising from contemplated transactions, by obviating the current need to argue their case according to differently formulated tests. This would in turn provide competition agencies with a better basis on which to build effective cooperation in cases that are notified in several jurisdictions. Moreover, as a common test would tend to highlight the actual application of the test, rather than the test itself, it would provide for better bench-marking of the activities of competition authorities and courts, as well as facilitating the development of competition-oriented research and modelling.

161. It should nevertheless be pointed out that an amendment of the test in the Merger Regulation could also involve some drawbacks. Although this effect should not be exaggerated, interested parties may, at least for an initial period following such a reform, face greater difficulties in forecasting the likely outcome of merger control proceedings in Europe. The reason for this would be that the existing body of case-law (emanating from both the Commission and the courts) has been built up under the Regulation's dominance test. Another possible complication relates to the fact that most Member States (as well as the Candidate Countries) have aligned their merger control provisions to the current dominance test [31]. Thus, unless national rules were also amended, changing the Merger Regulation's competition test could have the awkward effect of creating greater alignment internationally, while leading to greater disparity within the Community.

[31] It is noted, however, that the United Kingdom and Ireland are contemplating reforms towards the introduction of an SLC test. Moreover, other Member States are interested in evaluating the possible advantages of moving to such a test.

162. From a substantive viewpoint it should be noted that there are many similarities between the dominance test and the SLC-test. Both types of test, will, for example, involve an investigation into the scope of the relevant market as well as an assessment of how the market(s) will be affected by the proposed concentration and which competitive constraints would be faced by the merged entity. It should also be noted that, despite the current difference in legal tests, the vast majority of cases dealt with by the Commission and other major jurisdictions using the SLC-test have revealed a significant degree of convergence in the approach to merger analysis.

163. Since the adoption of the Merger Regulation in 1989, the application of the notion of dominance has evolved, allowing it to be adapted both to developments in economic theory and to refinements of the now available econometric tools to measure market power. This implies that merger assessment today can be less reliant on the rather blunt and imprecise market share test than it was 10 years ago. The fact that the dominance test has undergone such an evolution is natural, and Article 2 has so far proved sufficiently flexible to accommodate an effects analysis made on the basis of more sophisticated micro-economic tools, instruments and models developed by econometric and industrial organisation research.

164. The perhaps most well-known example of this evolution is the European courts' interpretation of the Merger Regulation's competition test as applying to situations of collective dominance, in the judgments of the Court of Justice and the Court of First Instance in the Kali und Salz and Gencor cases.

165. It has nevertheless been suggested that the SLC-test might be closer to the spirit of the economically-based analysis undertaken in merger control and less (legally) rigid than the dominance test. As such, some consider it better adapted to an effective merger control, in particular in the context of growing industrial concentration. At the same time, it has also been suggested that adopting the more open-ended SLC-test would lead to a greater degree of legal uncertainty.

166. One of the more specific hypothetical questions that has occasionally been raised about the reach of the dominance test in the Merger Regulation is the extent to which it would allow for effective control in some specific situations where firms unilaterally may be able to raise price and thus exercise market power. The type of example that tends to be cited is of a merger between the second and third largest players in a market, where these firms are the closest substitutes. In such a scenario the merging firms may remain smaller than the existing market leader. The argument goes that the SLC-test would be better adapted to addressing such a situation, in particular if the market characteristics would not be conducive to a finding of collective dominance. While interesting as a hypothetical discussion, the Commission has so far not encountered a situation of this kind.

167. In conclusion, experience in applying the dominance test has not revealed major loopholes in the scope of the test. Nor has it frequently led to different results from SLC-test approaches in other jurisdictions. Still, in view in particular of the increasingly international scope of merger activity, the Commission believes that the time is right to initiate a thorough debate on the respective merits of the two tests for merger control.

168. In order to open up a full debate on these important issues, interested parties are invited to submit their substantiated views on any perceived advantages or disadvantages resulting from the current wording of Article 2 of the Merger Regulation, and to also assess the effectiveness of the test by contrast with the SLC-test.

169. As this discussion touches on principles of fundamental importance to competition law, not only at the Community level, but also at Member State level, it is recognised that it may not be possible to reach definitive conclusions within the timeframe available for this review exercise.

B. Merger-specific efficiencies

170. Some commentators moreover suggest that the dominance test does not allow for a proper consideration of efficiencies that may result from mergers. To date, however, the issue of efficiencies has only been raised in a limited number of decisions under the Merger Regulation, and the precise scope for taking such considerations into account may not have been fully developed. Some jurisdictions, by contrast, explicitly provide for merger-specific efficiencies to be taken into account in the context of merger control (see, for example, the US Horizontal Merger Guidelines). In that regard, an "efficiency defence" allows a merger to go ahead where the benefits to the economy resulting from the efficiencies are deemed to outweigh the harm to the economy resulting from reduced competition. These jurisdictions, regardless of the substantive test applied, tend to allow for a defence of this kind only in those rather exceptional circumstances where the efficiencies are likely to be passed on to consumers, and despite a situation of dominance or substantial lessening of competition.

171. In any event, the burden of demonstrating that a particular concentration is likely to lead to significant efficiency gains is naturally placed on the party forecasting those gains, i.e. it normally rests on the notifying parties, who are, in any event, best placed to explain the scale and scope of such efficiencies as well as why the proposed merger is necessary for their realisation.

172. The Commission is aware of and supports the ongoing debate on how, and the extent to which, efficiencies should be taken into account in competition analysis [32]. Accordingly, and independently of the discussion on the two substantive tests, views are invited as to the proper role and scope of efficiency considerations in the field of merger control.

[32] See, for example, OCDE/GD(96)65 "Efficiency claims in mergers and other horizontal agreements".

C. Simplified Procedure

173. The 2000 Report indicated that European industry, whilst supporting an amendment that would allow for Commission competence over all cases with cross-border effects, also suggested that a less restrictive approach should be adopted in relation to transactions that are unlikely to have negative effects on competition. For such cases it was proposed, in particular, that the level of information required in a notification should be limited and that the standstill provision in Article 7(1) should not apply to such transactions.

174. In September 2000, the Commission introduced a Notice on a simplified procedure for treatment of certain concentrations under the Merger Regulation [33]. First experience of applying these principles has been very positive and the Notice has significantly enhanced the efficiency of European merger control.

[33] Published in the Official Journal of the European Communities on 29 July 2000 (OJ C 217, 29.07.2000, p. 32).

175. Between September 2000 and 30 April 2001, some 216 notifications were notified to the Commission under the Merger Regulation. About 39% of these were considered to fall under the provisions of the simplified treatment procedure. The average duration from notification till clearance was 25 calendar days. Thus, while in terms of procedural length, the new procedure has had little significant impact, it is likely that this situation will continue unless the current rule in Article 9(2) (which gives Member States three weeks from the date of receipt of a copy of the notification to request a referral) is amended. A possible amendment to Article 9(2) could be considered either in conjunction with a shortening of the current three-week deadline for a referral request under Article 9, as discussed above in the section on referrals. Alternatively, it could be provided that Article 9(2) would either not apply at all or apply, but with a shorter deadline, in cases where the parties make reference to the Notice on Simplified Procedure in their notification.

176. While the success of the new simplified procedure is borne out by experience to date, it is envisaged that measures could still be taken to further streamline procedures; it is, in particular, foreseen that a simpler Form-CO might be designed for such cases. This exercise will be undertaken in the context of a more general review of Commission Regulation 447/98.

177. It has been suggested that the practice on simplified procedure might be consolidated either into the Regulation itself or into a form of "block exemption" that could be built around the underlying principles of the Notice. The potential advantage would be to do away completely with the need to process harmless concentrations and with the need to adopt formal decisions in cases unlikely to have any significant value as precedents. Still, it might be prudent and in the interest of legal certainty to maintain some form of information requirement vis-à-vis the Commission and Member States.

178. From a substantive viewpoint it has been furthermore suggested by some that the Merger Regulation, as is now the case under some national rules, should allow for a de minimis threshold, with the result that the Commission would not examine dominance concerns that might arise in small markets.

179. Interested parties should submit their comments on how, in relation to the above-mentioned issues or any related ones, an amended Merger Regulation could best further the simplification of procedure. In particular, comments on aspects of transparency and legal certainty would be welcomed, but also on any of the other items outlined above.

IV. PROCEDURAL ISSUES

A. Notification - Triggering event

180. The 2000 Report indicated that industry representatives in certain circumstances felt a need for further clarification of the concepts in Article 4(1) that trigger the obligation to notify a concentration. These concepts have been further explained in the Commission's Notice from 1998 on the concept of a concentration. However, in view of the further experience that has been gained in applying the Merger Regulation, the need for further clarification will be discussed below.

181. This discussion will also cover the suggestion made by some industry respondents that notification should be possible even prior to the triggering points currently indicated in Article 4(1), i.e. before the conclusion of a legally binding agreement.

182. Unlike some other mandatory pre-notification merger control systems, Article 4(1) of the Merger Regulation explicitly specifies a point in time when the transaction should be notified. Failure to comply with this rule exposes the involved companies to the possibility of being fined under Article 14(1)(a). As it normally is in the parties' interest to notify as early as possible, with the aim of securing a decision as soon as possible, it has not in practice been necessary for the Commission to enforce the one-week requirement [34]. Moreover, it is established practice that the Commission will not enforce the one-week obligation, assuming that the parties do not take any steps towards implementation of the merger agreement, thereby allowing the parties the time necessary to prepare and file a complete notification. Various proposals have been made as to the latest point in time for a notification. These range from a complete removal of the deadline for notification to a codification of the current practice in this respect, or to an amendment whereby the one-week requirement would relate to providing informal information about the transaction while a longer period could be provided for the notification itself.

[34] Cases where fines have been issued for late notification also involved implementation of the concentration in contradiction to Article 14(2)(b) and Article 7(1), see Samsung and AP Möller.

183. Some industry representatives have proposed that Article 4(1) could be either repealed or amended as regards the earliest possible time for a notification. In practice, the Commission has regularly accepted notifications on the basis of an agreement between the Boards of companies, even where these may not be enforceable if certain conditions are not subsequently fulfilled, e.g. approval by a shareholders' meeting. One complication in interpreting Article 4(1) has been the existing national differences in the extent to which the management bodies of a company are able to bind the company. These differences mainly result from national provisions of company law, but may also be influenced by specific features in contractual traditions.

184. The main argument proposed in favour of relaxing the requirement for an agreement which is binding on the companies involved is that this would make it easier to co-ordinate notification to the Commission with notification to other jurisdictions, such as the US. It should nevertheless be pointed out that such co-ordination is possible also in the current system, i.e. by notifying to all authorities involved at the date when a binding agreement has been concluded. Still, it is to some extent understandable that business reasons may advocate commencing each regulatory procedure as early as possible.

185. The current policy of requesting that notifications are submitted after the conclusion of a sufficiently binding agreement (or public bid), is based on a number of assumption that, prima facie, remain valid. At the very least notifications must, as they trigger the start of the deadlines for the investigation, be made at a point in time when confidentiality reasons do not restrict the Commission's ability to fully investigate the transactions [35]. In addition, it might be inappropriate to launch investigations at a preliminary stage of the transaction if this could put any third party in a position of advantage or disadvantage compared to those making the notification. The possibility of earlier notifications may also raise an issue of efficient use of the Commission's resources (as well as the resources of others involved in such procedures). Finally, if notifications were to be allowed at an earlier stage of the transaction, parties may not be in a position to provide all information necessary for the Commission to complete the initial assessment within the one-month period set out in Article 10(1).

[35] The fact of the notification is, in accordance with Article 4(3) published in the Official Journal of the European Communities.

186. Notwithstanding these reservations, however, the Commission believes the possibility to introduce greater flexibility, which would allow better co-ordination of merger investigations in different jurisdictions, should be further examined. It invites interested parties to submit comments on any perceived advantages or disadvantages thereof, and in particular comments on how this could be achieved without resulting in negative consequences such as those outlined above.

B. Suspension of Concentrations

187. Article 7(1) provides that a concentration shall not be put into effect either before its notification or until it has been declared compatible with the common market (hereinafter "the stand-still obligation"). Under certain conditions, a derogation from the suspending rule applies by law to public bids, see Article 7(3). In other cases, the Commission can grant an individual exemption in accordance with Article 7(4). Breach of the standstill obligation is sanctioned by the penalty of substantial fines according to Article 14(2)(b). Moreover, under Article 7(5) transactions carried out in contravention of this obligation are void, unless the concentration is subsequently declared compatible with the common market.

188. It has been suggested that the relationship between, on the one hand, the legal exemption under Article 7(3) for public bids and, on the other, the situation for other acquisitions through the stock exchange would benefit from clarification. Acquisitions over the stock exchange are not, contrary to public bids, subject to mandatory rules that may oblige the bidder to implement the transaction at a certain date. Companies making such acquisitions have nevertheless argued that the provision in Article 7(1), for practical reasons, should not be allowed to pre-empt the completion of such transactions as long as the conditions set out in Article 7(3) are complied with. The opposite view, namely that the standstill provision in Article 7(1) should be fully respected, is usually adopted by the company which is the target for the acquisition.

189. As has been indicated above when discussing the concept of a concentration, it is considered appropriate to extend the scope of Article 5(2) to cover more clearly certain acquisitions through the stock exchange. Views are invited as to whether, in addition to this modification, it would be appropriate to clarify the scope of the current stand-still obligation with regard to such acquisitions or generally. In setting out their comments, respondents who are of the view that the one-week requirement in Article 4(1) could be abandoned are encouraged to consider the impact of such a modification and the appropriate scope of the standstill obligation.

C. Calculation of time limits

190. One of the most important aspects of the Merger Regulation is its creation of a system of short and definitive deadlines for the assessment of notified concentrations. To this end, the Merger Regulation contains numerous Articles specifying the period or point in time when certain events should occur. These deadlines are currently expressed in months, weeks and days. More detailed provisions for the calculation of these time limits are provided in the Commission's Implementing Regulation, which also sets out the principle of recovery of holidays.

191. Given the importance of compliance with all these deadlines for the merger control procedure and in order to increase transparency in their calculation, it seems appropriate to consider the introduction of a more simple method of calculation. Such a simplification could be achieved by consistently using a concept of working days in all relevant parts of the Regulation.

192. It is noted that a concept of working days is already used by some other competition authorities. In relation to the most important deadlines of the Merger Regulation, the concept of working days would indicatively translate into some 23-25 working days for a normal first phase assessment [36]. At the same time, the length of a normal second phase proceeding could be set at approximately 90 working days, with intervening deadlines, for example for the submission of proposed remedies.

[36] In view of the discussion on a possible extension of the scope of the simplified procedure (see above), it would be reasonable to use a 25 working day period for the reminder of the cases, as these, by definition, would involve a substantive analysis of competition effects. As a point of reference, it is noted that the existing proposal for revision of the UK merger control system would introduce a period of 30 working days.

193. Interested parties are invited to provide their views on the merits of a possible switch to the use of a working days concept in the Merger Regulation.

D. Administrative efficiency

194. Article 19 of the Merger Regulation imposes an obligation on the Commission to transmit copies of notification to Member States within three working days. The date of receipt by Member States of the notification is important, as it is decisive for the period according to which a referral request under Article 9 has to be made.

195. It is questionable whether these rules provide for the most efficient use of time and resources, not least as there have been significant improvements in the means available in transmitting information and documents since the rules were originally conceived.

196. It is therefore considered appropriate to assess whether these provisions could not be amended so as to provide that notifications are submitted directly by the notifying parties to the competent authorities of the Member States. At the same time, views are invited on what means might be available for providing a secure and reliable way to effect such transmissions, as well as on what means might be available for introducing electronic submission of merger notifications.

E. Completeness of Notification

197. The consultations for the 2000 Report also indicated that the legal certainty of notifying parties could be strengthened by clarification of certain aspects of the Merger Regulation procedure. One relatively widespread view was that there should be a legal deadline for the Commission's ability to declare a submitted notification to be incomplete (such a declaration re-starts the one-month period, see Article 4(2) of the Commission's Implementing Regulation, EC No 447/98). Various proposals were made as to the appropriate length of such a deadline (varying from 1-3 weeks after notification).

198. Following a constructive dialogue between the European legal community and Commission representatives, best practice guidelines were developed in 1999. These are available for consultation on the Commission's home page [37]. The discussions clearly revealed a common interest on the part of the Commission, as well as of the business and legal communities, in ensuring that declarations of incompleteness are kept to the minimum. Nevertheless it was recognised that it will not be possible for notifying parties to follow these guidelines in all circumstances, and that there may be situations where a declaration of incompleteness is unavoidable.

[37] See http://europa.eu.int/comm/competition/mergers/others/best_practice_gl.html

199. Declarations of incompleteness have occurred less frequently since publication of the best practice guidelines in 1999. In the years 1997-1999, the proportion of notifications that were declared incomplete was stable at around 10-11%. In the year 2000, it dropped to 6% and for 2001 it seems that a further reduction is likely.One of the main reasons for this positive development is that a further increase in the use by companies of pre-notification contacts with the Merger Task Force over the last years, advocated by the best practice guidelines.

200. The suggestion that the system be made more rigid requires careful consideration. It should, inter alia, be recognised that a more rigid system could lead to a less efficient use of the Commission's resources, and therefore ultimately to less protection of competition. In addition, the question must be raised whether a legal deadline for declaration of incompleteness would necessarily be to the advantage of notifying parties. If a new issue arises after the notification, the Commission will obviously do its utmost to resolve the issue by further investigation during the remaining time. It is probable, however, that there could be situations where the Commission, in the absence of a possibility to declare the notification incomplete, would be forced to open an in-depth investigation. This could happen regardless of whether the parties have made their best efforts to disclose fully and openly the relevant facts in the notification. As the opening of an in-depth investigation automatically triggers a number of procedural steps (Hearing, Advisory Committee, translations, adoption of decision by full Commission etc.), experience shows that it is very difficult to finalise the procedure within considerably less than the foreseen four months.

201. Thus, if the system were to be made more rigid through the introduction of a deadline for declarations of incompleteness, notifying parties would face an increased risk of prolonged procedures. In such a situation, their only option to avoid the prolongation would be to withdraw the notification and re-notify the transaction. This will normally not only involve additional costs and loss of time, but also contractual uncertainty, as a withdrawal of a notification without a simultaneous abandonment of the underlying transaction means that the notifying parties are in breach of Article 4(1), raising the question of a possible application of Article 14(1)(a).

202. In view of the above, the Commission is of the opinion that the current possibility to declare a notification incomplete fulfils a proportionate and appropriate objective in the rare cases where it is used. Comments on this subject are nonetheless invited.

F. Commitments Procedure

203. Comments indicating the need to review the procedural rules relating to the Merger Regulation's procedure for commitments have been submitted both by industry representatives and by Member States. For industry, the most relevant objectives in this respect can be summarised as increasing legal certainty and creating an atmosphere for meaningful discussions. Although, these concerns are more strongly felt in relation to the 2nd phase, owing to the finality of the decision taken at that stage, they apply also to commitment discussions at the end of the 1st phase. The interests of industry partly converge with those of Member States, who generally would like to see amendments that would ensure their full involvement in the procedure, as foreseen in Article 19 of the Merger Regulation [38].

[38] See also Commission Notice of 21.12.2000 on remedies acceptable under Council Regulation (EEC) No 4064/89 and under Commission Regulation (EC) No 447/98 ("the Notice").

204. The time-framework within which remedies discussions take place under the Merger Regulation is a complex one. It also works differently during the 1st and 2nd phase proceedings. In the initial phase, the notifying parties have the possibility to submit proposed commitments within three weeks from the date of notification (Article 18(1) of Regulation 447/98). In the 2nd phase, commitments may be proposed up to three months after the date on which proceedings were initiated (Article 18(2) of Regulation 447/98). These deadlines for the submission of commitments are necessary in order to allow sufficient time for the Commission to conduct consultations with Member States and interested third parties and to finalise its own assessment of the case.

205. Article 18(2) contains a possibility for extending the three-month time limit in "exceptional" circumstances. In order to maintain respect for the deadlines and to guarantee a level playing field, this provision has to be interpreted narrowly. It may be applicable in cases where the notifying parties, due to external factors beyond their control, are unable to submit proposed remedies within the applicable deadlines [39]. The provision is not intended as a solution to the constraints that will be further elaborated below.

[39] See, for example, Commission decision of 13.10.1999 in case COMP/M.1439 - Telia/Telenor, OJ L40, 9.2.2001.

206. The above-mentioned deadlines in Article 18 aim to guarantee adequate time for good quality contributions by each of the three types of "actors" involved in the process, i.e. the Commission, the business environment (notifying & 3rd parties) and Member States. This specific relationship imposes unique constraints upon the organisation and procedural aspects of the EU merger control system. Indeed, these provisions were based on a desire to establish a balanced solution, with acceptable procedural guarantees for all of the interests involved, while at the same time keeping the overall time needed for the procedure to the minimum.

207. In the 2nd phase, from the viewpoint of the notifying parties, it may be seen as problematic that the three-month deadline will often occur 2-3 weeks, or in any case relatively shortly, after the Oral Hearing. This implies that the parties may have to prepare themselves for discussions about commitments at the same time as they are absorbing the contents of the Commission's statement pursuant to Article 18 ("the Statement of Objections") and preparing the Oral Hearing. Industry representatives have indicated that the procedure should allow for a clearer separation of the time when they should absorb and defend themselves against the Commission's objections, and the period when they should prepare for remedy discussions.

208. This, in itself legitimate, position of the notifying parties often leads to a situation where commitments are submitted on the last day of the three-month period. This, in turn, has the effect that a very limited amount of time is left for the Commission to conduct the necessary consultation with Member States and interested third parties, prior to preparing a draft decision for discussion in the Advisory Committee. In all but exceptional cases this means that the Commission will not be able to include a full analysis of the proposal in the draft that is sent to the Member States, unless the proposed commitments are clear and straightforward and/or have been submitted well in advance of the deadline [40].

[40] According to Article 19(5) the Member States are normally entitled to receive the preliminary draft decision two weeks in advance of the meeting of the Advisory Committee.

209. For the above reasons, the current procedural rules could be modified so as to provide more time to all involved for proper evaluation of how to address identified competition concerns. In order not to introduce any overly complex and potentially time-consuming procedures, it appears appropriate that the new provisions should operate solely at the initiative of the parties and then as a direct result of the law. At the same time, the system must be designed so as to provide incentives for submissions at a time that will not unduly compress the time available for Member States or the Commission.

210. Experience has also indicated that, as concerns 1st phase commitments, the three-week period within which remedy proposals have to be submitted may on occasion be too short for establishing the exact scope of the competition concerns at hand. Equally, the time that is left after the expiry of this deadline may place undue pressure on the assessment of the proposals that is being made. These factors can affect the notifying parties, who may face difficulties in formulating an appropriate proposal within the time available. Also for the Commission and Member States, the current system may entail disadvantages in terms of, inter alia, unduly compressing the time available to form an opinion on submitted proposals. It therefore appears appropriate to also provide for the possibility of a limited extension of time during the 1st phase, at the initiative of the parties.

211. Upon adoption of the outlined amendments, it would also be necessary to adapt the Notice on remedies.

1. Amending the Applicable Procedure

212. A simple extension would most likely be disproportionate, as it would tend to penalise notifying parties who, under the current rules, come up early with appropriate proposals. A simple prolongation would in any case, unless the schedule of events and incentives is amended, risk to simply postpone the currently existing problem. What appears to be needed is a way to encourage timely submission of appropriate commitments ("best shot"), whilst allowing for a limited additional period that is often needed between the Oral Hearing and the deadline for submission of commitments. Another requirement of the system would of course be that it must be structured in such a way that the parties are not able to impinge on the time needed by Member States and the Commission to conduct an appropriate assessment.

213. It appears that a "stop the clock provision" could fulfil these criteria. Such a provision could, in a proportionate way, solve the problem of the current squeeze of the time available for the Commission and Member States when proposed commitments are submitted towards the end of the period allowed (or amended after this deadline).

214. However, in order to avoid introducing unacceptable inefficiency in the procedure or making it susceptible to abuse, such a provision would need to have some procedural safeguards. For the 2nd phase, these should include respect for the basic principles that a request for stopping the clock should be introduced within the original three-month deadline. Moreover, in order to maintain the efficiency of the procedure, the new provision should operate solely on the basis of a request from the parties. Furthermore, it should be built on the principle that the extension is available for a short finite period, for example 20-30 working days, and that the additional time is distributed in a proportionate way between all interested parties. In order to maintain the efficiency of the system as a whole it appears undesirable to provide for a longer period. It has also been considered whether the system should provide for the addition of a shorter period, to be equally distributed between the parties and the Commission (e.g. five working days each). Whilst the latter would provide for additional flexibility in the system (e.g. where it is likely that only a short additional period will be needed), it would however make the system significantly more difficult to manage (e.g. possibly less forward planning of Advisory Committees).

215. A "stop the clock provision" would appear to have a number of advantages in the 2nd phase procedure:

* It would strengthen the parties' procedural guarantees by allowing more time for discussion following the Oral Hearing, where the notifying parties and the Commission could elaborate on identified concerns and possible solutions;

* Assuming that the first proposal is made at a sufficiently early stage, it could restore time that has been lost as a result of the notifying parties making first sub-optimal proposal (e.g. where they genuinely, but ultimately wrongly, feel confident that they will be able to counter the Commission's objections);

* It would improve transparency of the procedure by allowing the Advisory Committee to discuss on the basis of a draft Article 8(2) decision evaluating the parties' final proposal (currently a draft Article 8(3) decision may be the only available document). Also in this respect, the notifying parties' procedural guarantees would be strengthened.

216. The "costs" of such a procedure would be limited. First, for the notifying parties, their own strategic choices would be decisive for whether the extra time would be used or not. For the markets, the possibility of an additional 20-30 working days would normally not have any significant impact. Second, for the Commission and Member States, it would mainly require some additional flexibility, as meeting schedules may need to be re-arranged, which in its turn would add some complexity to the logistics of Advisory Committee meetings (room, translations, interpreters etc). On the whole, it is expected that the positive effects would outweigh such minor inconveniences.

217. In practice, the system would provide notifying parties with a clear choice of options. The first such option would be to come up with a proposed commitment well in advance of the three-month deadline, allowing the results of the market test to be established before the end of the third month of the procedure. On the assumption that the result is positive, there would be no need to apply for the clock to be stopped, and sufficient time should be available for preparing a complete draft decision in view of the Advisory Committee. Should, on the other hand, the market test be negative, the parties would still have the option to request that the clock be stopped while additional discussions and market test are conducted. Under the proposal, the parties would then have an additional 10-15 working days to find an appropriate solution, after which the Commission would have the same amount of time to verify the viability of the final proposal and to prepare a complete draft decision in view of the Advisory Committee.

218. Alternatively, the parties may decide to propose commitments only at the very end of the three-month period. In such cases a proper market test may not be possible within the end of this period. Consequently, the parties may need to decide whether to ask for the clock to be stopped without knowing the full results of the market test. Should they decide to make such a request, the remainder of the procedure will follow that outlined in the preceding paragraph. In cases where the parties, on the other hand, decide not to request that the clock should be stopped, the continuation of the procedure will depend on the viability of that proposal. Thus, in cases where the late proposal turns out to satisfactorily remedy the identified concerns, the Commission may continue to be under some time pressure to prepare a complete draft clearance decision in good time before the meeting of the Advisory Committee. The risk of serious negative effects stemming from this should however rarely be significant, as successful remedies submitted towards the last day of the three-month period tend to be clear-cut in character. The same applies to unsuccessful remedies submitted towards the last day of the three-month period, where the Commission's draft decision would be built largely on its Statement of Objections, with some added comments on the proposal and its weaknesses.

219. For the 1st phase, the new procedure could largely follow that set out above. It would thus be up to the parties whether they, within a limited period after the end of the three-week period, felt that there would be merits in asking for the clock to be stopped for a short and finite period, in order to discuss a new or substantially revised proposal. Unlike the situation in the 2nd phase there would however appear to be reasons for allowing the Commission discretion as to whether to accept a request for the clock to be stopped. The reason for this is that it would be procedurally inefficient to use more than the current six weeks for 1st phase cases where the Commission does not see any possibility of adopting an authorisation decision, even on the basis of the new or substantially revised proposal. In such cases, the Commission should retain the power to adopt an Article 6(1)c decision at the end of the six-week period.

220. The Commission invites interested parties to indicate their experience and views on the outlined proposal as well as generally on the functioning of the procedure for 1st and 2nd phase commitments.

221. In addition to the above-discussed issues on the timing of the procedure in cases involving commitments, some commentators have suggested that a modification should be introduced whereby the Commission should take a more active role in identifying the measures that it deems necessary in order not to oppose a notified concentration. Views are also invited as to the desirability of such an amendment, and in particular as to the legal effects of such a "suggestion" or the absence thereof (as in the case of prohibitions where no appropriate remedies can be identified).

G. Article 8(4)

222. In case that a concentration has already been implemented, Article 8(4) allows the Commission to require the separation of assets brought together, the cessation of joint control or any other action that may be appropriate to restore conditions of effective competition. Such decisions, which are subject to the procedural protection under Article 18, may be adopted in pursuant to Article 8(3) or by separate decision.

223. The main scenario where an application of Article 8(4) may be necessary is when a concentration has been implemented in contravention of Article 7(1). Other possible scenarios may include public bids, which are not subject to Article 7(1), and cases that have come under the Commission's jurisdiction by application of Article 22. Finally, it may be appropriate to apply this provision in situations where the parties commit a serious breach of conditions or obligations that have enabled the Commission to approve the concentration under Article 6(1)b or 8(2). In the latter scenario, the breach would entail automatic illegality (in the case of breach of a condition) or the possibility to revoke the clearance decision (in the case of an obligation, see Article 8(5)). There may, however, be situations where reasons of legal certainty would necessitate recording these findings and at the same time setting out the type of action that would be appropriate to restore conditions of effective competition.

224. It has been suggested that Article 8(1), by stating that all proceedings initiated pursuant to Article 6(1)c shall be closed by means of a decision as provided in Article 8(2) to (5), would limit the applicability of these provisions to cases where the concentration has been notified and a second phase proceeding has been initiated. The Commission does not share this interpretation, as it would run counter to the very purpose and wording of Article 8, in particular of Article 8(4), which refers to "a concentration", without limiting the powers of the Commission to cases of "notified" concentrations (language used elsewhere in the Merger Regulation; see e.g. Articles 6, 8(2), 9(1), etc) and empowers the Commission to act "by separate decision", a possibility that would be pointless if the Commission could only act in cases of concentrations prohibited under Article 8(3). Nevertheless, the Commission invites comments on whether further improvement of the wording of Article 8(4) might be useful.

H. Enforcement Provisions

225. During the consultation for the 2000 Report, some Member States suggested that the effectiveness of the Merger Regulation's enforcement provisions should be reviewed. These provisions relate to two main issues; powers of investigation (Articles 11-13) and penalties (Articles 14-15). When the Merger Regulation was adopted, these provisions were modelled closely on the pre-existing provisions on the same subjects in Regulation No 17. The Commission's modernisation proposal for Articles 81 and 82 introduces a number of amendments to guarantee the protection of competition. To the extent that these adjustments relate to issues of similar importance to the merger control procedure, it is considered appropriate to introduce the same amendments in the Merger Regulation. Following the order of the modernisation proposal, the following items could be amended and/or introduced into the Merger Regulation:

* Sector inquiries. Under the Merger Regulation, the Commission should be empowered to conduct inquiries not only on the basis of a specific notification, but also in order to do more general studies, including post-merger studies. Such studies can provide valuable insight into the effects of past decisions and thereby be used for fine-tuning future policy.

* Clarification that companies remain responsible for the correctness of information provided by their duly authorised representatives (lawyers).

* Power to take statements. The merger control procedure, with its short deadlines, is particularly well suited for allowing oral submissions to be recorded and used as evidence. Indeed, it is felt that this may considerably enhance the efficiency of fact-finding in merger investigations.

* Inspections. Under the Merger Regulation, inspections have so far only been used exceptionally. Still, there would not seem to be any reason not to increase the potential effectiveness of these provisions along the same lines as for Articles 81 and 82.

* Investigations by national competition authorities. In the context of Articles 81 and 82, this provision will be modified to allow one Member State to conduct investigations on its territory on behalf of the competition authority of another Member State. This amendment has no equivalent function under the Merger Regulation, which, contrary to Articles 81 and 82, is based on a system of exclusive jurisdiction.

* Fines. It would be appropriate also in the context of the Merger Regulation to switch to a percentage-based calculation of fines for breach of procedural rules (up to 1% of annual turnover). Moreover, it would be appropriate to add the violation of failing to comply with an obligation imposed by decision pursuant to Article 6(2) of the Merger Regulation to the list in Article 14(2)(a) of that Regulation.

* Periodic penalty payments. It would be appropriate also in the context of the Merger Regulation to switch to a percentage-based calculation of periodic penalty payments (up to 5% of average daily turnover). Again, it would be appropriate to add the violation of failing to comply with an obligation imposed by decision pursuant to Article 6(2) of the Merger Regulation to the list in Article 15(2)(a) of that Regulation.

* In addition, the short and fixed time periods of the Merger Regulation, would at least in certain circumstances justify the Commission to be empowered to adopt a decision requiring certain information to be provided, without first having made an unsanctioned request. Under such a provision the Commission would be allowed to directly use Article 11(5) of the Merger Regulation, for example, when there are reasons to assume that the respondent will not provide a full reply in a timely manner.

226. The Commission invites interested parties to indicate their experience and views on the proposed modifications outlined above.

I. Filing Fees

227. The 2000 Report indicated that the introduction of filing fees for merger notifications would be further evaluated. Such fees are currently applied by several Member States (the UK, Germany, Austria, Spain, Ireland and Greece), as well as by a number of candidate countries and other non-EU jurisdictions, including the US [41]. In those jurisdictions filing fees are often justified as a means of ensuring that the merging parties bear the costs associated with merger control.

[41] A comparison of the size of filing fees in different jurisdictions can be found in Annex I, Section B.

228. Studies of other jurisdictions reveal that there are two main options for the introduction of filing fees. Such fees can be either of a fixed amount for each notification, or they can be proportional to the complexity of the notified transaction. Within each model, the fees may fall to be paid by one or more of the involved parties. Regardless of the preferred structure, a basic principle for the setting of fees should be to de-link them from the budget of the competition authority, so as to not create any perceived bias.

229. Accordingly, the Commission seeks comments on the appropriateness of including an enabling provision in the Merger Regulation, allowing the Commission to introduce filing fees by means of a subsequent Commission Regulation, if and when it might regard the introduction of such fees as justified.

230. It should moreover be noted that the Commission participates actively in the ongoing discussion at the international level about the effects of the introduction of filing fees by numerous competition authorities worldwide, a development that is a cause of some concern to industry. The Commission looks favourably on the fact that the issue of filing fees in merger cases is being discussed internationally, and could see some benefit to be derived from a possible future international harmonisation of approach to the issue.

231. Comments on this topic, including relevant experience of filing fees in other jurisdictions, are invited.

J. Due Process and "Checks and Balances"

232. One of the most successful and praised features of the Merger Regulation has been, since its conception, its capacity to deliver results within a very tight time schedule, while maintaining in the process a high degree of efficiency and transparency.

Nature and characteristics of the procedure

233. Merger control systems around the world combine in various valid ways the administrative and/or judicial elements in the procedure. To a varying extent, some merger control systems also involve a political element, for example in the shape of a test based not solely on competition grounds or, more directly, through intervention by Government or Ministers. It must be made clear that the procedure established by the Merger Regulation and its implementing regulation is administrative in nature and not judicial. [42] The apparent dual function of the Commission as investigator and decision making body is inherent in the structure of such administrative procedure and has not been found to contravene Article 6 of the European Convention on Human Rights. Most importantly, the Commission has been held to be subject to "effective" judicial review by an independent and impartial judge. [43]

[42] According to case law the Commission is not a tribunal in the sense of Article 6 (1) of the ECHR: Joined Cases 100-3/80 Musique diffusion Francaise v. Commission [1983] ECR 1825.

[43] Joined Cases 100-3/80 op.cit., Case T- T-348/94 Enso Espanola v. Commission [1998], ECR II-1875.

234. The formal administrative procedure incorporates an array of due process rights in favour of parties, affirmed and refined by a significant body of EC competition case-law. Transparency is a paramount value throughout the process and is safeguarded by features such as the adoption of a reasoned decision when an in-depth investigation is launched, communication of objections to parties, access to file, rights to refute the objections in writing as well as to an oral hearing, and full reasoning of final decisions.

235. On a more general level, transparency as to the Commission's policy is further strengthened by the practice of publishing Guidance Notices on important jurisdictional and substantive issues. Furthermore, publicity is given to the Commission's treatment of individual cases, e.g. on notifications, on decisions to open in-depth proceedings and on the full text of final decisions.

236. Other Commission departments and Member States provide internal and external "checks and balances". Second phase decisions, including prohibitions, must be adopted by the Commission deliberating in college. These working methods reflect the nature and structure of the Commission in its function as guardian of the Treaties and implementing authority of the Merger Regulation. In this light, it is understandable that the structure and timeframe of the procedure must also allow space for the organisational constraints imposed by the Commission's very working methods, i.e., consultations between services and with 15 Member States, translations and interpretation, logistics of meetings, of hearings and of decision finalisation (preparation of college vote, signature etc).

237. The following paragraphs illustrate briefly the main due process and "checks and balances" elements incorporated in Commission's merger assessment procedure.

Due process

238. It should be noted, in the first place, that a well-established practice of pre-notification contacts and a quite detailed form CO allow the Commission to depart upon a formal procedure immediately after notification and contribute to the effectiveness of the ensuing process. The Commission may exercise further fact-finding powers which directly involve the notifying parties and other market players, throughout both the first and second phase. The Merger Regulation subjects requests for information as well as investigations to a series of formal procedural requirements (Articles 11-13). Information acquired may be used only for the purposes of the relevant proceedings (Art. 17 (1)).

239. Reasoned decisions under Article 6(1)(c) to instigate in-depth proceedings are communicated to the parties and will, in practice, allow them to accurately forecast the scale and scope of the remainder of the procedure.

240. The second phase of the procedure may take on a more adversarial character than the first phase, and this is where the due process requirements are most solidly and explicitly built-into the system.

241. In specific terms, parties have the following rights:

* At the end of the investigation (roughly two months before the final decision) the Commission must duly communicate to the parties a statement of objections laying down clearly and in detail all its objections so that the parties know exactly the points they need to address in their defence. The statement of objections will also form the basis on which the parties may start preparations to formulate appropriate remedy proposals.

* After communication of the statement of objections, parties may request to have access to the Commission's file in order to verify the results of the investigation and examine for themselves the sources upon which the Commission based its objections (subject to the legitimate interests of preservation of business secrets and commercially sensitive information).

* In their written response to the Commission's objections, parties have the right to request a formal oral hearing, where they can develop their defence to the statement of objections. The competent authorities of the Member States attend the hearing. The importance of Member State's opinion, particularly with a view to the following procedural step (i.e., Advisory Committee on concentrations) is clearly recognised by parties, who regularly contact the Member State authorities on a bilateral basis concerning the case. Parties on whom the Commission proposes to impose a fine or periodic penalty are also entitled to the same defence rights.

242. Third parties who can demonstrate a sufficient interest (such as members of the board and representatives of the employees of the companies concerned (Art.18 (4), customers, suppliers and competitors), may also ask to be heard and submit their observations in writing.

243. One important category of third parties that only rarely make their views known in the context of ongoing merger control procedures is consumers (or their representative organisations). Although not strictly an issue falling within the ambit of the review of the Merger Regulation, the Commission would welcome any suggestions as to what assistance it could lend in order to encourage and facilitate consumer groups and organisations to more actively make their views known in relation to mergers falling within the scope of the Regulation and affecting their interests.

244. The Commission is open to hearing the views of employees, including during phase I of a merger investigation, to the extent possible given the severe time constraints inherent in such investigations. In the light of this, the Commission would welcome suggestions as to how employees or their representatives might be enabled to more effectively express these views, in particular regarding the likely market impact of a proposed merger.

245. Oral hearings are organised and conducted by a Hearing Officer (HO) whose role in safeguarding the right to be heard throughout the procedure has been strengthened with a new decision on his mandate [44]. His report of the hearing is made directly to the competent Commissioner. It covers due process issues generally, including disclosure of documents and access to file, time limits for replying to the Statement of Objections, the proper conduct of the oral hearing and potentially observations on the further progress of the proceedings (i.e., withdrawal of objections). His final report (which is communicated to Member States, attached to the final draft decision and published with the final decision in the OJ), will consider whether the draft decision deals only with objections in respect of which the parties have been afforded the opportunity of making known their views, as well as the objectivity of any inquiry concerning the competition impact of commitments. According to the new rules, the HO is attached directly to the Commissioner responsible for competition, is appointed through a transparent procedure and does not have to be a Commission official.

[44] Commission Decision of 23.05.01 on the terms of reference of hearing officers in certain competition proceedings, OJ L 162/21, 19.06.2001.

"Checks and balances" in the decision making process

246. Member States provide a most important element of external control to the procedure. Their role is more formalised in the second phase. Whereas in the first phase they are entitled to receive copies of the notification and of most important documents (including commitments) upon which they may comment, in the second phase they are also convened in the Advisory Committee on concentrations, which consists of representatives of the relevant authorities. A set of documents including a preliminary draft decision and any proposed commitments are sent to the Member States beforehand. If parties submit new undertakings, the Committee may have to be consulted again. The fact that the Commission attaches significant importance to the Committee's opinion is verifiable, as the opinion is published.

247. Other competent units within DG Competition as well as other departments of the Commission have a role in the process and their input is sought throughout the entire procedure, from the fact-finding and investigation stages through to the drafting and reasoning of the decision. The Legal Service plays a particular role as it oversees the legal aspects of the case, both as to substantive and procedural issues. Its prior approval is required before any act having legal effects can be proposed and adopted. When merger decisions are challenged, the Legal Service will be called upon to defend the Commission before the European Courts. DG Financial & Economic Affairs, and DG Enterprise or any other appropriate sectorial Directorate General, as well as the internal economists of DG Competition, provide technical, economic and industrial expertise into the analysis that is made in merger cases. They are also called upon to comment on preliminary opinions formed at various stages of the procedure, both in 1st and 2nd phase, by the team responsible for the investigation.

Modifications to the current procedural framework

248. Improvements to the current merger assessment procedure must be possible but, obviously, they should fit into the limits imposed by the present Treaty system and by the Commission's working methods, as outlined above. An important principle behind any modifications should be that they must aim at allowing the actors involved to make better quality contributions at the appropriate moment and the appropriate stage in the procedure. Any proposals must also acknowledge the possibility that concessions in one direction might mean restrictions in another. Any modifications concerning the commitments procedure (see above), should be seen in this light. Indeed, the Commission's proposal concerning 2nd phase proceeding allows concessions in terms of a very tight time schedule in order to enhance transparency and participation of all the actors involved in the discussion of commitments.

249. The Commission would openly consider any other substantiated suggestions in this spirit.

250. The current system of judicial review of merger decisions is considered by certain commentators to be unsatisfactory and ineffective. The length of EC court proceedings is deemed to discourage litigation and therefore to eliminate the constraint posed upon administrative action by the threat of judicial control. On 6 December 2000 the Court of First Instance adopted various amendments to its Rules of Procedure with a view to expediting proceedings in various fields which could include merger control [45]. As these amendments entered into force on 1 February 2001, it is however still too early to evaluate their effects. It is noted at this point that although the reform of judicial procedures is outside the competence of the Commission and certainly outside the scope of this revision exercise, the Commission would welcome any further reform undertaken by the European Courts to expedite appeals. This would moreover have the additional benefit of further enriching judicial precedent.

[45] OJ L 322 of 19 December 2000

251. There has in any event been both a not insignificant amount of litigation in merger cases and, more importantly, a number of appeals against prohibition decisions. Indeed, out of 15 prohibition decisions 7 have been challenged by the notifying parties, i.e. in 47% of the cases the Commission decision was (or is currently being) closely scrutinised by the Court of First Instance [46].

[46] As of 1 October 2001, RTL/VERONICA/ENDEMOL, GENCOR/LONRHO, KESKO/TUKO, BERTELSMANN/KIRCH/PREMIERE, AIRTOURS/FIRST CHOICE, WORLDCOM MCI/SPRINT AND GENERAL ELECTRIC/HONEYWELL.

252. Certain commentators have pointed to elements in other merger control systems which they regard as more capable effectively guaranteeing systematic judicial review. Under the US system, for instance, competition authorities must initiate proceedings before a federal court in order to block a merger. Experience shows, however, that many cases which are settled in the US on the basis of agreed remedies are not subject to court review. Nor are prohibition decisions in the US necessarily subject to judicial review, as merging parties may abandon their merger plans at the time when the competition authorities file suit. [47]

[47] In the period 1996-2001, the FTC filed a complaint for a preliminary injunction to block the implementation of the merger in 12 cases. Seven of those cases went to trial and one was settled prior to trial. In four cases the parties abandoned their transaction after a complaint was filed (in another four cases, the parties abandoned after Commission had voted to file such a complaint or after learning of Staff's intention to recommend with Commission to seek a Preliminary Injunction). Thus, overall, the number of abandoned transactions over the period (8) was as high as the number of cases argued before a judge (7). In the period 1996-2000, the Deparment of Justice filed an action in district Court in relation to 80 cases, whilst 126 cases were restructured or abandoned prior to filing a complant as a result of an announced challenge.

253. In conclusion, and quite independently of the fact that the reform of judicial procedures falls outside the scope of the review of the Merger Regulation, the Commission does not believe that the current system of judicial review fails to provide adequate judicial protection to companies whose merger plans are challenged under the Regulation. Nevertheless, within these limitations, the Commission would openly consider what assistance it might provide in support of any substantiated suggestions aiming at enhancing judicial review of its decisions in merger cases. Equally, the Commission would invite substantiated reflections on the relative merits of the overall procedural system provided for in the Merger Regulation as compared to merger control procedural systems employed in other jurisdictions.

V. INVITATION TO COMMENT

254. Interested parties are not only invited to submit their views on any or all of the specific issues raised throughout this paper, but also on any other matters of relevance to the implementation of merger control in Europe.

255. Comments should reach the Commission by not later than 31 March 2002 and can be transmitted in the following ways:

By mail to:

European Commission Competition Directorate-General - Merger Review - Green Paper -

B-1049 Brussels Belgium

By electronic mail:

mtfmergerreview@cec.eu.int

256. Extra copies of this paper can be found at

http://europa.eu.int/comm/competition/mergers/review/

ANNEXES

ANNEX I : THRESHOLDS RELATED ISSUES

ANNEX II : Article 9 - Article 22

The annexes are available in English only, together with the original version of the Green Paper at the following address:

http://europa.eu.int/comm/competition/mergers/review/