Select Committee on European Union Thirty-Second Report


PART 4: THE SUBSTANTIVE TEST

98. Currently, the ECMR uses a 'dominance' test[41] to judge whether a merger is "compatible with the common market." Under Article 2 of the Merger Regulation, the Commission blocks those mergers that it determines would "create or strengthen a dominant position as a result of which competition would be significantly impeded in the common market or a substantial part of it". The Green Paper instigates a debate on whether it might be appropriate to substitute for this 'dominance test' a different test, namely, will the merger substantially lessen competition. This test, commonly known as the SLC test, is used in many other jurisdictions, including the USA, Canada, South Africa, Japan, Australia and New Zealand. Given that the Government's Enterprise Bill would replace the existing United Kingdom 'public interest' test with one based on SLC, it came as no surprise that the Government saw "real benefit in moving to such a test under the ECMR" (p 153).

99. The Commission has stressed that it is not "wedded to the current wording" and has "no prejudice in favour of one formulation or another." Commissioner Monti has said that he is not expecting to reach definite conclusions on the substantive test within the time available for this review as this is a subject he would like "further discuss[ed] with Member States before deciding what course of action to take".[42] When we met with the Commission, it emphasised that it was "looking into this issue carefully" and was doing so "with an open mind" (Q 402).

100. This section deals with the advantages and disadvantages of both tests, as well as the question of the proper role of efficiencies in merger assessment.[43] In it, we weigh up the array of arguments for and against changing the current substantive test.

Are we playing with semantics?

101. We were keen to discover what the differences were between the two tests. For some witnesses, it was a mere matter of semantics that distinguished the two tests. However, we received conflicting opinions on this issue.

NO REAL DIFFERENCES?

102. Some witnesses, such as Dr William Bishop, Chairman of Lexecon Limited, thought that there was "very little difference in practice between the two tests" (Q 57). He did not see "major differences" between SLC and dominance in the way that the latter had been developed and applied by the Commission (Q 68). The Government said that a change from dominance to SLC would not be "a fundamental change" to how the ECMR functioned (Q 411). The Confederation of British Industry (CBI) accepted that "real differences" between the tests were "difficult to pin down". Nonetheless, the CBI claimed that a move away from the dominance test would bring with it "certain risks": businesses were concerned that the rules on merger control might "significantly change" as a result of a change of substantive test (Q 24).

THE CURRENT ECMR TEST COULD BE CALLED 'DOMINANCE PLUS'

103. Several witnesses said the differences between the two tests were minimal because the Commission's current test was 'dominance plus'. They argued that the existing test, properly construed, involved two elements, namely (i) "creating or strengthening a dominant position" and (ii) "as a result of which effective competition would be significantly impeded in the common market or a substantial part of it." The Union of Industrial and Employers' Confederation of Europe (UNICE) claimed that the ECMR test as drafted required the Commission to demonstrate both the creation or strengthening of dominance and a significant impediment to competition in the common market before a merger could be prohibited (Q 306).[44] The CBI also argued that a switch was unnecessary as the concept of SLC was already incorporated into the dominance test (Q 40). The Commission too emphasised that, in the Regulation, dominance was "surrounded by other wording" which was "quite similar to the SLC test" (Q 399).

104. The National Consumer Council (NCC) argued that a move to the SLC test would remove "the distracting need to demonstrate dominance." Instead, the immediate focus of the authorities would be on the competitive conditions in the relevant market, which was "the key issue for consumers". Because of the order of the wording in Article 2, at present the Commission had to find a dominant position as "a pre-condition" before it could examine the competition effects of the proposed merger. Yet the notion of dominance did not, of itself, say "anything about the effects on competition that might take place" (pp 194-95).

REAL DIFFERENCES

105. There clearly are similarities between the wording of the two tests. However, some witnesses, such as the TUC, held that the two tests were "clearly different" and would lead to "different decisions being made" (QQ 132, 150). Professor John Kay saw the SLC test as "both different and preferable" to the dominance test (Q 174).

A MOVE TO SLC WOULD LEAD TO MORE INVESTIGATION

106. One area where many witnesses did perceive a difference between the two tests was in the number of mergers that might be subjected to serious scrutiny in Phase II. Witnesses were agreed that under SLC more cases would be taken to Phase II (and so possibly prohibited unless acceptable remedies could be found).[45] It was suggested that mergers involving companies in oligopolistic markets would be particularly affected (see below paragraphs 142-51). Witnesses disagreed over whether or not such an increase would be a good thing.

107. For some witnesses, an increase in the number of mergers prohibited or subjected to commitments was certainly not desirable. The Competition Law Association (CLA) thought there was "a risk" that under the SLC test there would be "a great deal more intervention" under "a far stricter policy". It feared that the Commission, "as it did with Article 81(1) and the concept of appreciability," would apply SLC with "a very wide scope" and as a result take many more cases to Phase II and possibly even prohibit "significantly more cases based on the SLC test than it could do on dominance" (Q 97). The Joint Working Party of the Bars and Law Societies of the UK (JWP) confirmed that it was a "genuine concern" that there would be a greater number of Phase II investigations (Q 104).

108. From industry, UNICE signed up to the proposition that SLC was "a broader test" and "stricter" than dominance. The American Chamber of Commerce (ACC) also saw SLC as "marginally less permissive" and thought that it would probably lead to "more transactions being prohibited or subjected to remedies". Both organisations were unsurprisingly against greater intervention by the Merger Task Force (MTF), with UNICE arguing that oligopolistic markets were not "wrong or bad per se". As a switch to SLC would probably lead to more intervention, the ACC deemed this "to a certain extent" to be "a political decision".[46]

109. Dr Derek Morris, the Chairman of the Competition Commission, was absolutely clear that a switch to the SLC test would lead to more cases being tackled by the MTF than under the dominance test. However, he considered that this was a good thing: "if there is a substantial lessening of competition, that is something that should be investigated" (Q 264, 276-77).

110. The Commission appreciated that "arguably" SLC could be "interpreted in a wider way than dominance". However, Commissioner Monti rejected the idea that "it is easy to make a link between the substantive test and the degree of rigour or of toughness in the outcome" (QQ 397, 399).

111. The evidence suggests that, despite the similarity between the wording of the two tests, there are differences between them. In particular, it would seem that the SLC test will lead to more investigation by the Commission and that this would particularly affect oligopolistic markets, where it might be difficult for the Commission to find dominance. We return to this issue below (paragraphs 142-151).

Arguments for retaining the dominance test

112. The CBI and Mr Nicholson, of Slaughter & May, put the case for retaining the dominance test as "if it is not broken, why fix it?" (QQ 21, 472) Taking the position that, due to the current test being dominance plus, the differences between the tests were not that great, they considered that the current test was already an effective instrument of merger control. Moreover, witnesses who wanted to retain the dominance test claimed that there were a number of risks involved in a switch to SLC.

THE LOSS OF JURISPRUDENCE

113. One argument frequently cited against a move away from the dominance test was based on the fact that, in the time since the ECMR had been adopted in 1989, the Commission and the Community courts had generated a considerable amount of case law on the dominance test. Some witnesses feared that this would be lost in a move to SLC and that a change of substantive test would give rise to legal uncertainty for merging parties (Q 308). After any change of test, it would take time for a body of precedent to develop. As European merger law has already undergone substantial change—first by the recognition that Articles 81 and 82 of the EC Treaty applied to concentrations,[47] and then by the adoption of the ECMR and subsequent amendments to it—lawyers and representatives from industry were, unsurprisingly, reluctant to see another change.

114. The CBI said that people knew where they stood with the dominance test, although it might not be "the perfect test" (Q 21). The uncertainty and unpredictability that would result from any change was the main concern of the CBI (QQ 30, 37, 38). The TUC also took this line, arguing that the dominance test should be retained "for the sake of certainty and clarity" (QQ 126, 131). The European Consumers' Organisation (Bureau Européan des Unions de Consommateurs—BEUC) also expressed some concern about the "confusion" there would be in forecasting results under the ECMR if the present case law were to become invalid.[48] Even advocates of the change to SLC, such as Dr Bishop and the ACC, acknowledged that associated with a switch there would be costs in terms of precedents being devalued and some uncertainty being created (QQ 57, 60, 369).

115. However, the Commission said that the effect of such a loss "should not be exaggerated" (paragraph 161). Some witnesses suggested that this short-term loss of certainty could be lessened by the Commission issuing clear guidelines on how it would interpret and apply the SLC test.

116. The JWP did not think that the loss of jurisprudence would matter "enormously" as long as the SLC test was "clearly described and set out in guidelines" (Q 100). It pointed out that the value of precedents was "limited because each case turns so much on its facts" (Q 109). The Office of Fair Trading (OFT) also explained that "much of the case law would still be relevant", for example, where it was to do with defining the relevant market and market power (Q 220).

117. Witnesses were against changing the substantive test because they were worried about uncertainty; they argued that the change would render it more difficult to forecast the outcomes of cases. That the dominance test has a long antecedence in EC law and there is a considerable body of jurisprudence on the meaning of 'dominant position' is a strong argument in favour of not changing the test. Once SLC was well established, however, there would be no reason why that test should be any less certain than dominance. The argument is about the transition costs of any change of test and whether the switch warrants the confusion of going from one to another. We note that the Government considered that, in the long term, the advantages of a change of test would "outweigh the loss of case history" in the short term (Q 427).

MANY DOMESTIC SYSTEMS OF MERGER CONTROL IN THE MEMBER STATES HAVE ADOPTED DOMINANCE AS THEIR SUBSTANTIVE TEST

118. Many witnesses claimed that a change to SLC in the ECMR might be controversial for the Commission, as many domestic systems of merger control in the Member States and the applicant countries applied dominance under their national law.[49] Moreover, some of these countries had only recently adopted dominance as their substantive test in order to mirror the test in the ECMR. The CBI argued that it would be unwise for the ECMR to change to SLC when the SLC test was not widely applied elsewhere in Europe (QQ 21, 30). Mr Van Gerven, of the ACC, thought that this was "very important". He believed that the Member States would block a move to SLC (Q 377). The concern about the disparity between EU legislation and that of its Member States was also shared by the International Chamber of Commerce (ICC) (p 190), the TUC (Q 131) and the BEUC.[50] Dr Bishop thought that the resulting lack of co-ordination between the Commission and those Member States who operated a dominance test would be the "main disadvantage" of switching to SLC (Q 60).

119. The JWP recognised the potential political difficulty of the Commission proposing a shift to SLC and appreciated that it might be not be easy to persuade the Council of the benefits of changing to SLC. However, this difficulty, in itself, did not "justify the retention of the dominance test" if the SLC test was a better test (p 34; Q 106).

120. The short-term discrepancy between regulation at a national and EU level obviously had not been viewed as a persuasive argument by the Government, who have proposed a change to SLC in its Enterprise Bill, despite the difference that would arise as a consequence between the United Kingdom on the one hand and several other Member States and the Community on the other. Although it would be "helpful", the Government did not consider it "crucial" to share the same substantive test as the other Member States and the Commission. So long as there were two tests operating internationally there would always be some differences between different jurisdictions. The Government considered themselves to be moving into the "mainstream" by switching to SLC and thereby aligning themselves with other jurisdictions. But of greater importance for the Government was the fact that they saw the SLC test as "the better test" (Q 427).

121. Changes to the ECMR require unanimity. It has been suggested, however, that if the Commission led the way in changing to SLC, Member States would follow. Furthermore, in its Green Paper, the Commission briefly mentions that, as well as Ireland and the United Kingdom planning to introduce the SLC test, "other Member States are interested in evaluating the possible advantages of moving to such a test."[51] Likewise, the OFT suggested that a number of the European the National Competition Authorities (NCAs) currently apply a dominance test were receptive to the idea of the SLC test, although OFT pointed out that it would be the governments who would make the decision whether to switch (Q 222). The extent of any trend towards SLC will be a relevant factor in the decision on the substantive test.

Arguments for switching to the slc test

'INTERNATIONAL' MERGERS AND THE CONVERGENCE OF SUBSTANTIVE TESTS

122. The increasing globalisation of markets has led to an increase in the number and scale of large cross-border mergers. This has meant that such mergers often have to be examined by numerous separate NCAs across the world. Consequently, international co-operation between competition authorities is increasing in importance (p 153). In view of the fact that the main merger control regimes outside Europe use the SLC test, some witnesses said that changing the substantive test in the ECMR could improve co-operation between competition authorities and the handling of international or global mergers that have substantial effects outside the EU. This was one of the Government's two main reasons for arguing that "a move to SLC would be beneficial in the ECMR context" (p 153). The JWP also considered that adopting the SLC test would be a "helpful move towards the international harmonisation of merger control systems" (p 34).

123. However, some witnesses pointed out that, even if the same substantive test were applied across the different jurisdictions, the way in which the test was applied could give different results. For the CBI, harmonising the legal test in the legislation would not "in and of itself lead to a more consistent approach" because, however the test was phrased, the regulators would retain "a broad margin of discretion". Differences in the way in which a substantive test could be applied meant that it would be necessary to go "much further than simply harmonising the basic legislation" to obtain convergence of international standards (Q 26).

124. The two different tests have been working side by side for some time and some witnesses observed that there had not been a great deal of difficulty between international regimes because of the different tests. When invited, the CBI was unable to name any cases where different jurisdictions using different tests had come to different answers that were attributable to the substantive test used. Even with the GE/Honeywell case as a possible example, Dr Bishop was amongst those who did not think the difference of substantive test between the US and EC had had any impact on the outcome of a case (Q 60).

125. Discussions between US and EU case handlers have frequently focused on the definition of the relevant product market. Where the authorities have formed different views on market definition (e.g. American Home Products/Monsanto, Hoffmann-La Roche/Boehringer Mannheim) the reasons for the divergence have been explained by other factors and are rarely attributable to different substantive tests. In a relatively small number of instances, differences in analytical approach and substantive conclusions have led the EU and US authorities to seek different remedies in respect of the same transaction. For example, in Glaxo/Wellcome, the US agencies required complete divestiture of Wellcome's R&D activities in a certain field, whereas the EU conditioned approval on the license of only one of the parties products. Up to GE/Honeywell, the most notable example of the Commission and its US counterpart failing to reach agreement occurred in Boeing/McDonnell Douglas, where the Department of Justice viewed the transaction as benign and concluded that no remedy was necessary, while the Commission challenged the transaction and ultimately required a package of remedies to address their competition concerns.

126. Derek Morris agreed that problems caused by the difference of substantive tests had been "relatively rare." Nonetheless, he stressed that in individual cases the distinction could be "significant". Moreover, he saw this as "an increasing problem" with the situations in which differences would occur becoming "more prevalent". Consequently, he thought that harmonising the substantive test with that of other countries was "certainly an important reason" for the Commission to switch to a SLC test (QQ 264, 278).

127. The OFT also argued that this potential "transatlantic harmony in terms of language" would be one of the benefits of switching. It too was keen, however, to point out that differences in how some mergers had been handled were "relatively rarely" to do with differences in the substantive test; the test had not been "a significant factor" in the exceptional cases where there had been disagreements. The OFT suggested that there was already "considerable" convergence between the way in which mergers were handled in Brussels and the US (for example, in the methodological analysis of market definition, market power, and barriers to entry). The major differences were procedural (QQ 223, 236).

128. Whereas these witnesses presumed that, for international convergence to occur, the ECMR would have to move to SLC, the CBI and UNICE hoped for international convergence on the dominance test (QQ 31, 303, 317).

129. As far as Professor Kay was concerned, these considerations of international convergence were secondary. He considered "the only reason" for having convergence of tests was if one were "converging towards a good policy" (Q 187).

THE SLC TEST IS MORE SOLIDLY ROOTED IN ECONOMICS

130. A number of witnesses claimed that the SLC test was directly grounded in economic analysis and the impact of a merger on competition in a way that the concept of dominance was not. Thus, the Government argued, the SLC test was "fundamentally better adapted to merger control" (p 153). For Professor Kay, the dominance test focused too much on "the structure of an industry"; it was not flexible enough to deal with differences between industries and did not focus enough on maintaining competition (Q 183).

131. For the JWP, the SLC test was "a less artificial notion than dominance," which was "a legal construct" and was not "rooted in economic theory" (p 34). The ACC and NCC also believed that the adoption of the SLC test would result in a more economics based approach in the assessment of mergers (p 129; Q 372). Dr Bishop thought the SLC test was better suited to the economic analysis required in merger cases, but only "to a marginal degree" (Q 58). The CBI, however, said that, although it recognised that it had been argued that SLC was a more economics based test, it did not agree (Q 40). On the basis that the current test was 'dominance plus', UNICE also rejected the argument that SLC a was a more economics based test (Q 306).[52] However, the Commission did concede that 'dominance' was "not directly an economic term" (Q 402).

THE SLC TEST HAS A COMMON-SENSE RESONANCE

132. Many witnesses explained that the SLC test was a more straightforward test than the current dominance test, which could be interpreted as a two-stage 'dominance plus' test (see above paragraphs 103-04). The regulator, rather than having first to establish that a merger would create or strengthen a dominant position and then to examine whether that would result in a significant impediment to effective competition, could just analyse whether a merger would significantly lessen competition.

133. John Vickers, the Director General of the Office of Fair Trading, said that, if the prime purpose of merger policy was to safeguard against competitive market processes being undermined, the SLC test was better. "A great attraction" of the SLC test was its "directness" (Q 218). The NCC agreed that having to demonstrate dominance was a distraction. The authorities should focus directly "on the competitive conditions in the relevant market," which was "the key issue for consumers" (pp 194-95). BEUC considered that the SLC test would "fit in more easily with the analytical process" of examining a merger.[53] Derek Morris concurred that, from the regulators' point of view, the SLC test was superior "just in terms of the very words and concepts" contained within it (Q 264).

134. Dr Bishop also took the position that the SLC was a better test than dominance and was linked more directly in its language to the way in which regulators needed to examine mergers. There was an "artificiality" in talking about dominance (Q 57). John Vickers agreed that the SLC test would bring greater clarity to the substantive test in the ECMR. Even if the two substantive tests did mean the same thing, it was unfortunate that that was not clear to everybody (Q 218). However, the Commission pointed out that although SLC probably had "some conceptual advantages", it was not completely transparent as the word 'substantial' in 'substantial lessening of competition' did not have an automatic definition and there would still be borderline cases (Q 402).

135. The Committee shared the concern that proving dominance might be a distraction for regulators, who are primarily concerned with maintaining the competitive conditions in the relevant market. The SLC test might more closely mirror the Commission's analysis of mergers and could assist them in focussing on the competitive impact of the merger in question. However, the question remains whether the issue is one of substance or just language; that is, whether to some extent the dominance test as applied by the Commission is indistinguishable in effect from to the SLC test.

COULD THE COMMISSION ALREADY BE APPLYING THE SLC TEST?

136. It has been suggested that a move to SLC would not significantly change either the way in which the Commission went about its competition analysis or the outcome of the ECMR process in the great majority of cases. Nonetheless, some witnesses argued that the Regulation should be amended so that the substantive test set out in it and the analytical method applied in practice by the Commission more closely equate.

137. The Government developed this argument and said that it could be used as evidence that the risks of a change to the substantive test in the ECMR were "in practice relatively limited". They reported that some stakeholders had held that the Commission's flexible use of dominance meant that the test in operation was "in practice already very close to SLC and that a change would lead to the legal text reflecting enforcement practice more precisely" (p 153). The JWP agreed with this argument saying that "most cases would be decided in the same way, whether the test was based on dominance or the substantial lessening of competition" (p 34).

138. The ACC explained that the way in which the Commission was now construing dominance had moved towards what the US did under its SLC test namely, not so much studying the market structure and applying a structural test based on comparing market shares, but analysing instead what the effect of the merger would be likely to be by examining how it would impact on the conduct of the undertakings involved: "In reality therefore the two tests have moved together" (Q 371). The ACC also explained how the Commission had had to "stretch" the interpretation of dominance in order to carry out investigations into the cases it wanted to examine (QQ 371, 376). The dominance test was originally addressed to transactions involving firms that were individually dominant. The Commission had since expanded the test to cover situations of 'collective dominance'. Even witnesses who argued against a change of substantive test, such as the CBI and the TUC, were concerned that the concept of dominance was being over-stretched by the Commission (p 2; Q 153). John Vickers explained that this had led to a "frustrating uncertainty" about exactly what was meant by the phrase collective dominance (QQ 219, 224).

139. Mr Götz Drauz, the Director of the Merger Task Force, admitted that the Commission had "developed and exploited progressively" its interpretation of dominance since 1989, when there was a presumption that dominance could only occur in cases where an individual company had more than 50 per cent of the market. The Commission had since found dominance in market shares as low as 27 per cent (Q 399).[54] For Dr Bishop, the uncertainty over collective dominance and the way in which the Commission had moved the goalposts for the dominance test was "one of the least satisfactory matters" of the ECMR and one that had led to a "great lack of clarity" (QQ 54, 55).

140. Derek Morris agreed that the Commission was probably applying SLC, but dressed up in the language of dominance. That was "very confusing" for the parties and introduced "an element of randomness" into a test that should be as predictable as possible (Q 265).

141. The benefits associated with the Commission's practice mirroring the wording of the substantive test in the ECMR might not outweigh the disadvantages of a change. If the Commission were to adopt SLC merely as a reflection of its current enforcement policy, any impact might be restricted to the wording of the Regulation and not its application. However, that might be too conservative an assessment; there remain other arguments for change. For its part, although the Government said that a move to SLC would "in all probability not alter the way that the Commission goes about its job in the majority of cases," it remained convinced that adopting SLC "could bring real benefits in certain types of cases" (Q 411). We examine these scenarios below.

THE SLC TEST IS BETTER THAN THE USE OF 'COLLECTIVE DOMINANCE' FOR DEALING WITH THE PROBLEM OF CO-ORDINATED EFFECTS

142. It was argued that the SLC test was better for examining mergers that did not give a company a market share that could be judged to make it individually dominant, but which reduced the number of companies operating in a market that was highly concentrated (i.e., that held few competitors). Changing the ECMR for this reason would only affect a small percentage of mergers, but, as shown by the Airtours/First Choice case, these can be very significant.

143. The Committee heard much criticism of the approach taken by the Commission to cases of collective dominance. It was said that, in applying the dominance test, the Commission had failed to articulate a consistent framework for assessing co-ordinated effects. Dr Bishop was particularly concerned:

"It is not natural to speak of dominance when you have a leading firm—that is, say, 40 per cent—and what you are really worried about is that the second firm will merge with the fourth firm, and the total concentration will rise. You have to go through the mental contortions of saying that they are dominant, and you do that through saying that there is such a thing as a collectively dominant entity. It would be much simpler to analyse the likely outcome directly, in terms of whether quantities will be further restricted and prices rise, and not have to go through these mental contortions about 'dominance' and describing a 'dominant' entity, which is a rather artificial construct." (Q 60).

144. Others witnesses also expressed concern about the concept of collective dominance in the ECMR context. The CLA thought that it was a "slightly artificial concept" (Q 97). The Government agreed, considering that the SLC test avoided "the need to shoehorn assessment of a significant competition detriment into the artificial construct" of collective dominance (Q 430). Professor Kay agreed that the dominance test resulted in the MTF "trying to shoehorn" the decisions they wanted to make into a framework that was "not really appropriate for them" (Q 180). He argued that

"it would be a much better structure if one could get to the right decisions for the right reasons even if it does not necessarily change the decisions, and it certainly serves better in future the task of giving people more confidence in the process and more ability to assess what the outcome is likely to be" (Q 192).

145. Many witnesses considered that, if adopting SLC meant losing the case law concerned with collective dominance, this would not be a cause for concern. Derek Morris was sceptical of the value of the existing case law (QQ 278-79). Professor Kay said the loss would not be a problem as he did not feel that there was "a body of case law that had defined what [collective dominance] was or applied it in a coherent way" (Q 191). John Vickers also drew attention to these "ambiguities and uncertainties" that had arisen, pointing out that there remained "open questions" on collective dominance "given the case law to date" (QQ 219-20). It was suggested that moving to SLC would enable the Commission to establish a clear framework for evaluating collective dominance and co-ordinated effects.

146. The Commission's application of the concept of 'collective dominance' in merger cases was recently examined by the Court of First Instance (CFI) in the appeal over the Commission's decision to prohibit the merger between Airtours and First Choice. The ruling by the CFI in the Airtours/First Choice case clarified the assessment of collective dominance under the ECMR and the burden of proof required of the Commission to prohibit a merger in highly concentrated markets.[55] In the Government's view, although the judgment did "not undermine the concept of collective dominance", it did "make clear the very high level of proof" that the Commission was required to meet in order to block a merger on the grounds of collective dominance (Q 430).

147. Witnesses not only complained about the lack of clarity over the concept of collective dominance; they also argued that there were three types of case with which collective dominance was incapable of dealing. Each involved companies competing less vigorously post-merger without necessarily colluding tacitly and without the creation or strengthening of a dominant position.

148. The first type of case, mentioned by the Government (Q 429) and Derek Morris, involved small companies introducing new innovations into a well-established market. Dr Morris painted the picture of:

"a set of incumbent firms in the market, perhaps a market in which there has not been any great innovation, and a new entrant appears with new technology, new ideas, new products, and it really threatens the existing companies, so one of them buys it out. It may be a very small company. It may just involve buying this particular computer specialist or whatever. It is almost impossible to apply the concept of dominance, and yet it could very effectively and substantially lessen the competition that would appear in that market. Indeed, you could have a series of such innovative entrants emerging and the incumbents just picking them off one at a time." (Q 264)

149. In the second type of case, which was given as an example by the Government (Q 429), there might be one very significant player in the market (with a market share of say 75 per cent) and a merger between the two remaining, more minor players (who would occupy the other 25 per cent of the market). In this situation, where the second and third largest firms merge,[56] it would seem that the dominance test would not allow the Commission to intervene, despite the substantial lessening of competition. Commissioner Monti appreciated the potential difficulty for dealing with such a case under the dominance test (Q 401). Mr Van Gerven, of the ACC, suggested that perhaps the Commission might be able to raise competition questions about such cases but it would not do so "using the right analysis" and would have to lower still further the test for collective dominance (Q 376).

150. The third case again demonstrates how a reduction in competition could occur in an oligopoly without either tacit collusion taking place or the creation or strengthening of a dominant position. Professor Kay and Derek Morris both mentioned the Abbey National/Lloyds TSB merger in the United Kingdom as a proposed merger that could have caused problems for the Commission applying the dominance test. The merger would have increased the market share of Lloyds TSB from 22 per cent to 27 per cent and the share of the four largest banks from 72 per cent to 77 per cent. The Competition Commission had prohibited the merger but reported that it would have been "extremely difficult" for it to have done so under the dominance test, even though there was clearly a substantial lessening of competition. Professor Kay thought this was such a problem that it warranted a change of substantive test in the ECMR (QQ 179, 182, 264).

151. Dr Bishop agreed that the main problem with the dominance test was its capacity to deal with such oligopolies. There was no "robust method" of predicting a post-merger outcome in most oligopolies and this would "remain a question of judgement for the foreseeable future" as judgement could "never be done away with". Nonetheless, the SLC test would be preferable here (QQ 63, 64). The JWP were quite clear that "the 'problem' of oligopolistic markets could be better addressed under a SLC standard" (p 34). The Government also viewed SLC as "particularly well-suited to tackling oligopolistic markets" (p 153).

SEPARATING THE SUBSTANTIVE TEST FROM ARTICLE 82

152. Dominance is also a precondition for Article 82 of the EC Treaty and assessing abuses of market power—a separate matter from mergers—under Article 82 and in national competition law in almost all of the Member States (including in the UK Competition Act). When the ECMR was first established, one of the reasons the Committee was in favour of the dominance test was on account of its "familiar terminology" (1989 Report, p 21). At the time, the Committee was concerned about the legal uncertainty if the wording in the ECMR test were different from (what was then) Article 86 EC. In 1989, the dominance test had the advantage for legal practitioners that the jurisprudence of the European Court of Justice in interpreting this word would apply to the Regulation. After over a decade and some 2000 decisions made under the ECMR, it is necessary to reconsider whether or not this link between the language of Article 86 (now 82) and the substantive test in the ECMR is still beneficial.

153. The Committee heard sound arguments for completely detaching the test under the ECMR from the dominance standard under Article 82.[57] Witnesses explained that the purpose of merger control was to ensure that the structure of the market is not rendered less competitive after a merger than before it rather than to prevent future abuse of a dominant position. They argued that there would be an advantage in the fact that the adoption of the SLC test would effectively de-link the ECMR and Article 82. The Government saw real benefits in introducing a "clear separation" between the competition test applied in the ECMR and the concept of dominance under Article 82:

"Under Article 82, a party that holds a dominant position is under a special obligation as to its future commercial conduct. This obligation applies to parties found to be either singly or collectively dominant. The increasingly broad interpretation of the concept of collective dominance under the existing ECMR is, therefore, placing numbers of undertakings under a special responsibility under Article 82 even if they themselves hold a relatively small market share, since they have been found to be in a position of collective dominance. A move to SLC under the ECMR would prevent this situation being aggravated any further, since the direct read-across to Article 82 would be removed" (p 153).

154. The JWP agreed that merger control required a prospective consideration of the structure of a market, post-merger, whereas Article 82 sought to identify those acts or omissions of dominant firms that deviated from the standards required arising from the special responsibilities which Article 82 imposed upon them. The JWP therefore considered that it would be "beneficial" if the jurisprudence and decisional practice in relation to these two matters were left to develop separately from one another. This was an increasingly acute problem as the Commission, in some of its decisions under the ECMR, appeared to have given a different meaning to dominance and had reached a finding of dominance which it would have been unlikely to have reached under Article 82. The JWP concluded that eliminating that confusion between the two tests by a move to SLC would be helpful. (p 34; Q 100).

155. Many other witnesses, including John Vickers (Q 218), agreed with this analysis. The ACC considered the impact of the ECMR decisions on the dominance case law under Article 82 to be a "real drawback" and saw this as one of "the most compelling arguments" for a change to SLC (QQ 371, 379). Dr Bishop agreed that this problem of "cross-contamination" between the ECMR and Article 82 was the "main disadvantage" of the dominance test (Q 57).[58] Professor Kay saw the current link between the two pieces of legislation as limiting the way in which the MTF could operate and interpret dominance (Q 181).

THE NEED FOR GUIDELINES ON THE SUBSTANTIVE TEST IN THE ECMR

156. Whether witnesses were for or against SLC, they were unanimous in calling on the Commission to issue, without delay, guidelines on how it interpreted the substantive test in the ECMR. The CBI said that it was "high time that the Commission issued a notice on its substantive analysis" and wanted the Commission to provide "very detailed" guidelines (Q 39). The JWP considered such guidelines to be "essential". If the Commission were to move to SLC, guidelines would be even more important as everyone would need to know in advance precisely what that test meant (p 34; Q 106). The JWP was concerned about the amount of discretion the Commission would otherwise have when it came to interpreting the term 'substantial' (Q 98). Dr Bishop and the OFT both agreed that the substantive test should only change when the Commission had issued guidelines (QQ 60, 220).

157. It was suggested that the Commission could draw on the guidelines on the SLC test that other jurisdictions had already produced. John Vickers considered that "a great deal" could be learned from the various guidelines issued by regulators around the world on the application of the test. He thought that there should be and would be "considerable" international convergence in terms of the guidelines issued on this topic (QQ 220, 232).[59]

158. Commissioner Monti promised that the Commission would issue a notice on market power in merger analysis irrespective of whether or not it proposed a change in the substantive test. He was confident that this notice would provide "a very important set of clarifications" (Q 402).

Conclusions on the Substantive Test

159. The debate about the substantive test raises questions of fundamental importance to the Community's competition regime: the decision to change the test should not be taken lightly. We considered the arguments advanced in favour of both dominance and SLC. There are similarities between the tests; both require making predictions and in many cases their application may give the same result. However, there is broad agreement that the SLC test is in many ways better in theory and that, if now one were drafting the Regulation from scratch, it is probable that the SLC test would be adopted.[60]

160. The Regulation has, however, been in force for more than a decade and witnesses were concerned about the implications of a change, in terms of the uncertainty it would cause and the unpredictability of how the new test would be applied. A move in Europe to SLC could certainly have short-term disruptive consequences. There is, therefore, a tension between what is convenient in the short term and a sensible long-term objective. Nevertheless, we believe that the SLC test should be the substantive test in the ECMR for the following reasons:

· If a merger might significantly impede competition in a market, the transaction should be examined in detail. It should not be a pre-requisite for the Commission first to prove that the merger would also create or strengthen a dominant position. Moving to the SLC test would allow the Commission to analyse mergers solely on how they lessen competition. The Commission's application of the present test stresses dominance, may cause the Commission to "stretch" the wording of the Regulation and runs the risk that significant impediments to competition may not be investigated.

· The SLC test comprehends the dominance test, but this does not apply vice versa. In particular, the recent Airtours/First Choice case may have serious implications for the dominance test and the ability of the Commission to vet mergers that substantially lessen competition in oligopolistic markets.

· Businesses are understandably concerned that the SLC test would lead to more mergers going to Phase II investigation and possibly being prohibited. It is, however, the Committee's view that any merger that substantially lessens competition should be investigated and the SLC test provides a better basis on which to assess such cases than the dominance test.

· A move to SLC would establish a clear separation between the language contained in the substantive test of the Merger Regulation and the threshold for assessing abuses of market power under Article 82 of the EC Treaty. It would allow the jurisprudence and decisional practice of the Commission in relation to these two matters to develop separately from one another and result in greater clarity.

· There would be costs and risks associated with a change of substantive test in the ECMR but these, we believe, would be manageable. Any change should be accompanied by the Commission issuing draft guidelines on how it proposed to interpret and apply the SLC test. It is imperative that such guidelines are discussed and put in place ahead of any change taking practical effect. The argument about the uncertainty that would result from adopting a different substantive test—that parties would not be able to anticipate how the Commission might apply the test—should then lose much of its force. Even if there is no change of test, the Committee is firmly of the opinion that the Commission should, without delay, issue guidelines on how it interprets the substantive test in the ECMR. This would help to alleviate the current uncertainty as to how the dominance test as applied by the Commission differs from SLC.

· International convergence of merger laws is a desirable objective. As a first move in this direction, we would like to see the harmonisation of substantive tests. Although there might still be differences in its practical application, this would represent a major step. It would improve co-operation between competition authorities and the handling of international or global mergers.

161. Having weighed all the evidence and the arguments on either side, we consider that the SLC test should be the substantive test in the ECMR. A move to SLC would give greater clarity, transparency and predictability.

Merger-specific efficiencies

162. Article 2(1)(b) of the ECMR provides, inter alia, that, when determining whether or not a merger is compatible with the common market, the Commission should take into account "the development of technical and economic progress provided that it is to consumers' advantage and does not form an obstacle to competition". The Commission, however, has never approved a merger that created or strengthened a dominant position on those grounds. If the Commission deems that a merger would create or strengthen a dominant position that would result in a significant impediment to competition, the merger is blocked.

163. Some witnesses suggested that the creation or strengthening of a dominant position was not always unacceptable when set against the attempts to achieve efficiencies for consumers. In some instances, a merger that might create or strengthen a dominant position could, at the same time, produce benefits for consumers in the form of either lower prices or greater choice or quality. It was proposed that, if these efficiencies—such as economies of scale or innovation through expensive research and development—were merger specific, they might occasionally outweigh the disadvantages to consumers associated with the reduction in competition. In these very exceptional cases, the ECMR should be able to take account of these efficiencies and allow the merger to go ahead. Such an amendment would create a 'defence' for those mergers that might lessen competition but which would yield demonstrable economic efficiencies.[61] This could be achieved by introducing a possible second stage to the ECMR appraisal process. Those mergers which were judged to fail the initial assessment of competitive effect (i.e., that create or strengthen a dominant position—or, following a switch, substantially reduce competition) would still have a possibility of exemption and authorisation. This would allow the competition detriment to be balanced or set off against other benefits.

164. Witnesses from industry were very keen for the Commission to give explicit reference to an efficiencies defence in the ECMR (pp 8; 130). They argued that efficiencies, such as innovation and consumer benefits, should be balanced against any anti-competitive effects of a merger. Noting that the US was more prepared to make such a balance, the Union of Industrial and Employers' Confederation of Europe (UNICE) even suggested that a change of Commission policy in this respect "could improve constructive co-ordination with the United States" (Q 293; p107). Arguments in favour of considering efficiencies also came from the European Parliament's Committee on Economic and Monetary Affairs, who said that, "for mergers in innovatory sectors, a balance should be sought between the need for economies of scale in the research and development sector […] and maintaining reasonable competition".[62] Dr Bishop also thought that "in principle" an efficiencies defence should be established (Q 76).

165. Professor John Kay, however, was against introducing an efficiencies defence into the ECMR. It would necessitate the Merger Task Force (MTF) making the kind of speculative cost-benefit analyses which he considered impossible to do with any degree of certainty and he was sceptical about the ability of economists to predict what the future structure of the industry and market would be. Although Professor Kay believed that there were pro-competitive mergers, he considered that it would be possible to take account of all the relevant arguments involved in such cases by applying the SLC test, without the need for an explicit efficiencies defence (Q 188). Mr McMahon, of Mytravel Group, agreed that efficiencies should not stand apart as a separate aspect of merger appraisals (Q 478).

166. The Commission explained that "one of the main streams of thought behind the merger review" was "to give more explicit and more predictable recognition to efficiency considerations." It said that if a merger brought efficiencies, and it could be expected that these efficiencies would be passed on to consumers, this was something to be taken into account. It believed that clarifying this issue would help companies to perceive its procedure as more predictable. (QQ 396, 398)

CAN AN EFFICIENCIES DEFENCE BE INCORPORATED INTO THE DOMINANCE TEST?

167. The Commission requested views on the role of efficiencies in merger control, independently of a discussion of the two tests. However, the Green Paper highlighted claims that the dominance test might not allow for the proper consideration of efficiencies that could arise from mergers (paragraph 40).

168. Indeed, several of our witnesses who had supported the introduction of the SLC test considered that SLC would make it easier to adopt an explicit rule on efficiencies (pp 34, 195). In this respect, they thought that the SLC test was more solidly rooted in economics than the dominance test (see above paragraphs 130-131), and would provide "a more explicit basis" for assessing efficiencies arising from mergers.[63]

169. Not all the witnesses agreed on this point, however. Some claimed that as the current test was 'dominance plus' (see above paragraphs 103-04), this already enabled the Commission to take efficiencies into account. The "second limb" of the existing substantive test (i.e., "as a result of which effective competition would be significantly impeded in the common market or a substantial part of it") made the consideration of efficiencies possible (QQ 319, 378). Other arguments were used to suggest that the Commission already had a legal basis to take account of efficiencies (pp 50, 107).

170. The Confederation of British Industry (CBI) also raised a more general point. The ECMR was based on Article 81 of the EC Treaty, so Article 81(3) arguments (about efficiencies being verifiable and measurable) should apply to it by analogy in any event (p 8). The Commission also mentioned Article 81(3) and, perhaps conscious of Article 2(1)(b) of the ECMR, did not think that there was "an objection in principle" to efficiencies being taken into consideration in the ECMR dominance test. They too stressed that the dominance test was 'dominance plus' and "on that basis" felt "more confident about looking in a more prospective and dynamic way at efficiencies" (QQ 397-99).

WOULD AN EFFICIENCIES DEFENCE BE PARTICULARLY IMPORTANT IF THE ECMR CHANGED TO SLC?

171. A separate argument was founded on the proposition that a test based on SLC would be likely to prohibit more mergers than the dominance test (see above paragraphs 106-11). The Joint Working Party of the Bars and Law Societies of the UK (JWP) said that if the ECMR incorporated the SLC test it would be particularly important for it to adopt an explicit efficiencies defence in order to compensate for the higher number of mergers that it envisaged would be prohibited under the new test (p 34). The Competition Law Association (CLA) agreed that if the SLC test were adopted an explicit efficiency defence would be "one of the most important things" that would need to change in the ECMR, in order to reflect "the lower SLC hurdle" (p 50; QQ 97-98).

172. John Vickers, the Director General of the Office of Fair Trading, acknowledged that there were some who would argue that the efficiencies defence would or could be treated differently in the context of the two substantive tests but he described this as "rather arcane and difficult territory" (Q 225).

DOES THE COMMISSION HAVE AN EFFICIENCIES 'OFFENCE'?

173. The American Chamber of Commerce (ACC) mentioned "the widespread perception" that occasionally the Commission had de facto applied an efficiencies offence. That is, in situations where the merger of two companies could generate economies of scale and scope making it more difficult for other companies to compete, the Commission would say that this, in itself, was an offence and a reason to prohibit the merger (p 130).[64] Indeed, the National Consumer Council (NCC) considered that the Green Paper implicitly acknowledged that, "on rare occasions", the dominance test might "actually lead to the rejection of a merger case that stands to benefit consumers" (p 195).

174. The Commission, however, insisted that it did not have an efficiencies offence. Rather, when companies become more efficient, and therefore stronger competitors, the Commission considered this to be "a positive point for competition" (Q 396).

WHAT CONDITIONS MUST BE MET FOR EFFICIENCIES TO BE CONSIDERED?

175. Although the Government saw merit in the Commission clarifying its approach to this subject, they urged "a relatively cautious approach" because they considered that efficiencies would "only be relevant to the assessment of mergers in a small number of cases" (p 156).

176. The JWP argued that any efficiencies defence "should be drawn narrowly". It should be available to the parties only where the efficiencies were merger-specific (i.e., where the efficiencies could be achieved only as a result of the merger) (p 34). The view that the Commission should consider only those efficiencies which came about as a consequence of the proposed merger was widely held (e.g. pp 50, 156). Other criteria were also commonly judged to be important for the Commission to consider. Dr Derek Morris, the Chairman of the Competition Commission, insisted that for efficiencies to be considered they should be "significant, reasonably imminent, reasonably certain." The most forceful opinion, however, was that the Commission should take account of efficiencies only where they would demonstrably be passed on to the consumer. It was argued that if the significant efficiency gains accrued only to shareholders, there would be no reason for them to be passed on to consumers. Dr Morris stressed this point:

"A merger that clearly has substantial synergistic gains but creates a very substantial degree of market power such that there really would be no particular incentive or pressure to pass those benefits on, would have benefits that would be unlikely to count" (QQ 267, 275).

177. The European Parliament's Committee on Economic and Monetary Affairs agreed that, when evaluating efficiencies, attention should focus especially on the consumer.[65] Judging whether or not the efficiencies would be passed on would demand very close examination by a Competition Authority.

178. The Commission confirmed that under its proposed efficiencies test the parties would have to show that the efficiencies would be passed on to the consumer. The Commission admitted that, once it had ruled that a merger would create or strengthen a dominant position, it would be difficult to demonstrate that the remaining elements of competition were such that they would "lead to a transfer of wealth or benefits to the consumer." The Commission, therefore, thought that in future any efficiency defence would be used "rather exceptionally" (QQ 396, 400).

179. Consumer groups expressed a concern that, when considering efficiency gains that accrued to the consumer, the Commission should not extend these benefits to include other social concerns. The NCC believed that the criteria should be "specifically focused on the interests of consumers rather than the wider economy or social concerns". Furthermore, it asserted that "very strong evidence" should be necessary to approve a merger in such circumstances (p 196). The European Consumers' Organisation (Bureau Européan des Unions de Consommateurs—BEUC) was concerned that an efficiencies test "would introduce 'industrial policy' considerations into key decisions." That could "politicise" the decision-making process, "much to the detriment of expediency, transparency and consistency".[66]

AT WHAT STAGE IN THE REVIEW PROCESS SHOULD THE COMMISSION CONSIDER EFFICIENCIES?

180. Witnesses were clear that an explicit efficiency defence could only take place after the Commission had found that a merger could give rise to a competition problem. If a merger had been judged to have failed the substantive test, there could be explicit provision in the ECMR to consider whether there were any offsetting benefits from efficiencies when the Commission was considering remedies.

181. John Vickers saw "a lot of merit" in a system where, if there were benefits flowing through to customers in the relevant market, those could be taken into account at the remedies stage. However, he considered it to be "quite hard to think of mergers which would substantially lessen competition and bring such efficiency gains which moreover flowed through to customers in the relevant market." Nonetheless, he did not consider it to be impossible and consideration of efficiencies could be a "useful" part of the remedies stage (Q 226).

CONSULTATION WITH CONSUMER GROUPS

182. We considered whether it would be possible for the Commission to judge better whether, and if so to what extent, the proposed efficiencies of mergers would be passed on to consumers by consulting more with consumer groups.[67] The Government attached "importance to facilitating the active involvement of consumers' representatives in competition enforcement". They welcomed "more active steps to engage consumers' representatives in ECMR cases", although they stressed that "any input should be on competition issues alone and not the broader social implications of any concentration" (p 150). The NCC and BEUC both called for the MTF to notify consumer groups directly of mergers and invite these organisations to comment on each case.[68] The NCC even went as far as suggesting that the MTF should create an Office for the Consumer in European Mergers in which a separate post would be established—"the Consumer Participation Officer." This person "could be regarded as comparable to the Hearing Office" and would have responsibility for notifying consumer groups and facilitating their access to non-confidential aspects of the file (pp 197-98).

THE NEED FOR GUIDELINES ON EFFICIENCIES

183. Witnesses wanted the Commission to provide guidelines on all of these areas of uncertainty. The JWP argued the need for the Commission to clarify the position of efficiencies in merger analysis. The Commission should "provide guidance both on efficiencies as a 'defence' and as on 'offence'" (p 34). Witnesses from industry also called for clarification on the scope of the efficiency defence that would lay out both when efficiencies would be considered and what the standard of proof required by the Commission would be (pp 8, 107, 190).

184. As noted above (paragraph 158), the Commission undertook, "by the end of this year", to produce draft interpretative guidelines on its practice relating to the application of the substantive test. The guidelines would embrace the analysis of market power and "necessarily" would have to include "an important chapter" on how the Commission would look at efficiencies and take such considerations into account (QQ 389, 396).

CONCLUSIONS ON EFFICIENCIES

185. The Committee is extremely sceptical whether it is necessary or desirable to introduce into the text of the ECMR provisions setting out an explicit efficiency defence. If there were a significant reduction in competition in a market, there would be very little incentive or reason for the company to pass on any efficiencies to the consumer. The ECMR is about maintaining competition and the Committee is keen to ensure that that object is achieved.

186. The Committee is also concerned about how the Commission could ensure that the efficiencies proposed by the parties were in fact passed on to the consumer after the merger, without regulating prices or controlling capacity. It is difficult to envisage an acceptable and effective system of imposing sanctions where projected efficiencies do not materialise and it would be hard for the regulator to defend the public interest. This is a particular problem as the vast majority of mergers have been shown not to deliver value.[69] Accordingly, the Committee recommends that the ECMR should not include an explicit efficiencies defence.

187. If, however, in a specific case an efficiencies defence were to be put forwarded, the Commission should be encouraged, when considering possible efficiency gains and remedies, to seek advice from consumer groups. The Commission should also issue guidelines to consumer groups on the most effective ways of becoming involved in merger cases.


41   This is the common short-hand name for the substantive test and is used by the Commission in the Green Paper. Some witnesses questioned whether it accurately reflected the wording of the test and the procedure by which the Commission examines mergers. We outline this discussion below (paragraphs 103-04). Back

42   Paragraph 169 of the Green Paper. See also Commissioner Monti's speech "Review of the EC Merger Regulation-Roadmap for the reform project", given at the Conference on Reform of European Merger Control, in the British Chamber of Commerce, Brussels, on 4 June 2002. Back

43   There is a debate as to whether mergers that fail the substantive test in the ECMR (i.e., that create or strengthen a dominant position as a result of which competition would be significantly impeded) should be permitted if, as a result of the merger, benefits (or 'efficiencies') would accrue to the customer. We explore this issue below (paragraphs 162-87). Back

44   In practice, it is not clear whether both elements of the test must be satisfied before the Commission may prohibit a transaction or whether the second element (i.e., significantly impeding effective competition) is merely a description of the consequences of the first. Back

45   The GE/Honeywell case was cleared in the US on SLC and prohibited in Europe on the dominance test. However, witnesses referred to this as an exception that was not comparable to other cases (see below paragraph 124-25). Back

46   QQ 306, 310-15, 323, 372-73. Back

47   See paragraph 5 above. Back

48   BEUC/X/016/2002. Back

49   For information on which accession countries use which substantive test see the Government's evidence (p 169). Back

50   BEUC/X/016/2002. Back

51   Footnote to paragraph 161 of the Green Paper. Back

52   For an explanation of this argument see above paragraphs 103-04. Back

53   BEUC/X/016/2002. Back

54   This is probably a reference to Carrefour/Promodes, where the Commission's decision found dominance at less than 30 per cent. Back

55   Case T-342/99, judgment of 6 June 2002. The Court of First Instance annulled the European Commission's decision to block the acquisition of the United Kingdom's tour operator First Choice by competitor Airtours, on the basis that the Commission had made a series of fundamental errors of assessment. The CFI concluded "that the Decision, far from basing its prospective analysis on cogent evidence, is vitiated by a series of errors of assessment as to factors fundamental to any assessment of whether a collective dominant position might be created". The ruling does not call into question the theoretical idea of collective dominance it merely expounds strict criteria that the Commission will have to satisfy in order to defend a finding of collective dominance. The Court stated that three conditions must be met if there is to be a finding of collective dominance:

  1. Each member of the alleged dominant oligopoly must have the ability to know how the other members are behaving in order to monitor whether or not they are adopting a common policy. According to the Court, this required "sufficient market transparency for all members of the dominant oligopoly to be aware, sufficiently precisely and quickly, of the ways in which the other members' market conduct is evolving".
  2. The situation of tacit collusion must be "sustainable over time, that is to say there must be an incentive not to depart from the common policy on the market".
  3. The Court stated that "the Commission must also establish that the foreseeable reaction of current and future competitors, as well as of consumers, would not jeopardise the results expected from the common policy".

In short, tacit collusion must be feasible and it must be sustainable, both internally (within the dominant oligopoly) and externally (the competitive pressures from fringe firms and customers must be sufficiently weak).

Equally important, the judgment stresses that by merely relying on assumptions without establishing "cogent evidence", the Commission failed to prove "to the requisite legal standard that the concentration would give rise to a collective dominant position".  Back

56   This was typified in the Heinz/Beech-Nut case in the US, which was mentioned by the OFT, the Commission, Dr Bishop and Mr Burnside, of Linklaters (QQ 58-59, 222, 376, 401). Back

57   If the current test is interpreted as a two-stage 'dominance plus' test (see above paragraphs 103-04), it could be argued that some detachment is afforded by the second element of the test (i.e., "as a result of which effective competition would be significantly impeded"). But no witness mentioned this as having an effect. Back

58   UNICE was alone in not considering this to be an "important" point (Q 316). Back

59   The Office of Fair Trading, in the context of the Enterprise Bill, is currently working on draft guidelines to explain how it considers the SLC test will be applied in the United Kingdom. Back

60   Derek Morris: "almost all economists and almost all lawyers in principle agree that the substantial lessening of competition test is the better test. To get such widespread agreement across both professions is truly wondrous to behold and extremely rare. […] I do not think if we were designing the test today almost anyone would go for the dominance test." (Q 264). Dr Bishop said: "If we started with a blank piece of paper today, I think we would start with substantial lessening of competition and it would not be controversial" (Q 57). Back

61   The Government's Enterprise Bill suggests a limited efficiencies defence in United Kingdom domestic law. This appears in Clause 29 (relevant consumer benefits) as first published for the House of Lords (HL 92). The efficiencies that are acceptable in this bill and so qualify as relevant customer benefits are "lower prices, higher quality of goods or services in any market in the United Kingdom […] or greater innovation in relation to such goods or services." The benefits must also accrue "within a reasonable period [of the merger]" and be "unlikely to accrue without the [merger taking place]". Back

62   A5-0217/2002. Back

63   BEUC/X/016/2002. Back

64   The case most frequently cited (e.g. by the ACC, p 130) to support this assertion is GE/HoneywellBack

65   A5-0217/2002. Back

66   BEUC/X/016/2002. Back

67   In the Green Paper, the Commission declares that it is open to ideas encouraging or facilitating the involvement of consumer groups (paragraph 243). Back

68   BEUC/X/016/2002. Back

69   cf. World Class Transactions: Insights into Creating Shareholder Value Through Mergers and Acquisitions, KMPG Transaction Services, 2001. In its 1999 survey, KMPG found that only 17% of the companies in the survey created value as a result of the merger. In 2001, KMPG found that 30% of deals added value. Back


 
previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2002