Monday, April 9th, 2007 by Juan Briones

Introduction

The Commission has launched, on 13 February 2007, a public consultation on draft Commission Guidelines on the assessment of non-horizontal mergers under Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings (the “Merger Regulation” and “the Guidelines” respectively).

Interested parties are invited to provide comments on this draft by 12 May 2007. These comments will be published on DG COMP’s web-site. Consequently, the comments submitted here below, do not contain any confidential information.

The publication of guidelines are most welcome, in so far as they clarify policy and allow companies and their advisors to assess ex-ante any potential competition issues their M&A activity may involve. In my view, the Guidelines correctly focus on the analysis of companies’ strategies and incentives to adopt one course of action or the other. The comments below will, however, focus on the points that are considered weaker or unclear, just because the purpose of submitting comments is obviously to identify areas where the Guidelines may be further clarified or improved. Therefore, the fact that most comments may be of a “negative” nature should not be taken to mean that the Guidelines are perceived as a negative document overall.

Principle of neutrality between vertical mergers and vertical restrainst.

1. There are two main comments of a general nature. Firstly, to the extent that the Guidelines focus on potential competition problems and not so much on situations where competition concerns can be excluded, the general impression is that vertical and conglomerate mergers are prone to involve competition issues. Although the Guidelines state that non horizontal mergers “are generally less likely to create competition concerns than horizontal mergers” because they do not entail loss of direct competition and may entail efficiencies, most of the guidelines deal with situations where a non-horizontal merger rises competition concerns, and very little effort is devoted to situations where high market shares do not involve a serious competition concern.
2. In its Guidelines on vertical restraints, however, the Commission stresses much more the economic rationale and efficiencies that may derive from any particular vertical restraint. If one agrees to the idea that a vertical merger is only an alternative to a vertical restraint, or a combination of vertical restraints, it is unclear why the attitude towards vertical mergers would be different to the attitude towards vertical restraints.
3. Thus, perhaps it would be advisable to refer in the Guidelines on vertical mergers to the analysis of vertical restraints, and explicitly indicate that vertical mergers are just an alternative to (a set of) vertical restraints. It is unclear in the current draft whether the Commission maintains a principle of neutrality with regard to vertical mergers and vertical restraints as alternative courses of action, and a clear statement to that respect would improve the transparency of the competition policy of the Commission.
4. The importance of recognising explicitly that the economic analysis of a vertical merger should not be different than the economic analysis of a vertical restraint is of consequence when considering the key policy aspect of efficiencies. The experience with the analysis of efficiencies in horizontal mergers (1) is, so far, rather disappointing. The Commission announced formally that it would take efficiencies into account in its analysis of concentrations, when adopting the guidelines on horizontal mergers, over 3 years ago. Since then, however, only in one instance (2) has the Commission recognized the existence of substantial and verifiable efficiencies arising from a merger, but these were not considered merger specific in so far as the Commission identified less restrictive means to achieve them (namely the creation of a production joint venture).

Is there really an efficiency defence in practice?

1. In the context of vertical mergers, in particular, it seems difficult to conceive any substantial and verifiable efficiency that cannot, in principle, be achieved by a vertical restraint or a combination of vertical restraints between the merging parties. In particular, the elimination of the double mark up, improved coordination between manufacturing and distributing, or the alignment of incentives between the upstream and the downstream firm (the three sources of efficiency identified in the Guidelines) can be achieved through any of resale price maintenance, exclusive territories, quantity forcing or non linear pricing.
2. Some of the vertical restraints may be considered less restrictive of competition than a vertical merger (such as absolute territorial protection or resale price maintenance), but further clarification in this respect would seem necessary.
3. Otherwise, although the Guidelines make express references to efficiencies, it is unclear whether those statements of principle will allow, in practice, to actually clear a case that raises issues of anticompetitive foreclosure on the basis of efficiencies. Perhaps the Guidelines make it logically impossible that efficiencies are actually taken into account in cases where they may play an important role. The requirement that efficiencies cannot be accomplished by less restrictive means than a merger. If applied in abstract, this requirement makes it hardly possible that efficiencies are really taken into account, both with respect to horizontal and with respect to (more clearly, perhaps) vertical mergers.
4. According to the Guidelines, efficiencies are taken into account to the extent that they may create incentives in the merged entity to act pro-competitively or to reduce prices (for instance, by means of new incentives to expand sales at both levels of the vertical chain due to the internalization of double mark-ups, point 13 of the guidelines, or directly by inducing a price reduction, see point 30 of the Guidelines).
5. In a framework where the anticompetitive effects of a merger are balanced against the pro-competitive effect arising from efficiencies, it makes sense to require that efficiencies be merger specific, but perhaps to require to that effect that, in addition, there are no alternatives that are less restrictive of competition, may be too stringent. Would that really matter? The point is that efficiencies are there (assuming they are proven to the requisite standard) and that they offset possible anticompetitive concerns, thus consumer welfare (and not total welfare) is not worsened by the merger. Is it coherent to object to the merger and dismiss efficiencies, just because it is possible to conceive, in theory, a less restrictive alternative to achieve those efficiencies?
6. The issue applies to horizontal mergers as well, but it appears to be even more acute with regard to vertical mergers. As the Guidelines stand, it is difficult to see any possibility in practice that an efficiency argument may proof successful. Perhaps further reflection is needed on the issue of efficiencies, and the Guidelines may offer a good opportunity to reassess the Commission’s approach to efficiencies under the merger regulation.

Competitors with similar degree of vertical integration or capable of replicating a similar degree of vertical integration in the post-merger scenario.

1. The second general remark refers to the absence of references in the Guidelines to whether the fact that other competitors have the same degree of vertical integration or a similar portfolio than the merged entity, is sufficient to dismiss concerns about input or customer foreclosure. From a general point of view, the fact that a number of competitors with a significant market share have integrated vertically before the parties to the concentration, will generally allow to dismiss any serious concerns about anticompetitive foreclosure (3), to the extent that these integrated competitors would be able to exert a sufficient competitive constraint on the merged entity.
2. Even if the notified concentration may possibly lead to the exit of certain non-integrated competitors in the upstream or downstream market, it cannot possibly lead to the exit of vertically integrated competitors. It seems safe to assume therefore, that the existence of significant competitors with a similar degree of vertical integration in the post-merger scenario is sufficient to dismiss any concerns on anticompetitive foreclosure (although it certainly may raise issues of coordinated effects) (4). The Guidelines will perhaps gain in clarity if this was explicitly stated. Such a statement would also help equilibrate the Guidelines in that they would provide more guidance on situations where concerns are not justified.
3. Furthermore, concerns on anticompetitive foreclosure could be dismissed when competitors do not have a similar degree of vertical integration, but could replicate that degree of vertical integration in the post merger scenario, i.e. there is a number of significant non-integrated competitors in both the upstream and downstream market. This would be, perhaps, the most significant effective and timely counter-strategy available to competitors under the threat of foreclosure. In this respect, it is noteworthy that competitors may replicate the same degree of vertical integration either through mergers or through a set of vertical restraints.

Incentives and risks to adopt anticompetitive conduct in vertical mergers.

1. With regard to conglomerate mergers, the Guidelines explicitly analyse that pure bundling and tying may entail losses for the merged company, and do not necessarily increase profits (point 105 of the Guidelines). However, there does not appear to be an equivalent assessment with regard to refusals to deal or price increases in relation to vertical mergers. Such strategies may also prove unprofitable, depending on the circumstances. The Commission has itself considered such a balancing trade off with regard to satellites and the (upstream) provision of launching services (5). The Commission concluded that in view of demand characteristics, restricting the provision of launching services only to customers of satellites would probably reduce the sales and profits of the merged entity in both the launching services market and in the satellite market.
2. There does not appear to be in the Guidelines an explicit recognition that strategies to foreclose may entail a risk to lose revenues, not just increased costs (see for instance point 104). This turns on the question of demand characteristics, as well as on the existence of alternative sources of supply, in line with the analysis of the Commission in Boeing/Hughes. A reference to this analysis may also help re-equilibrate the “negative” style of the Guidelines.

Illegality of future conduct.

1. Both in relation to vertical and conglomerate mergers, the Guidelines point out that anticompetitive foreclosure may require the adoption of certain course of conduct, and that in assessing the likelihood that the merged entity adopts such conduct, the Commission will take into account a number of factors, among which the legality of that conduct (point 44 of the Guidelines with respect to vertical mergers and point 70 and 108 with regard to conglomerate mergers).
2. The Commission cannot presume that a company will engage in unlawful conduct just because it is in a position to do so (sentence of the Court of First Instance in case T210-01 GE v Commission). On the other hand, anticompetitive foreclosure will necessarily require some course of action that will most certainly infringe article 82 (refusal to deal, discrimination between an affiliated company and third parties, etc…). In fact, and having in mind DG Competition Discussion Paper on the Application of Article 82 to Exclusionary Abuses, any conduct that is designed to foreclose rivals or has the effect of foreclosing rivals is considered an abuse.
3. The current wording of paragraphs 44 (vertical mergers) and 108 (conglomerate mergers) suggests that the test to assess the compatibility of a merger is based on the likelihood that the merged entity adopts a certain type of conduct, whereas in principle, the test would be whether the market structure in the post merger scenario allows for a sufficient degree of competition. Furthermore, such conduct should not be (i) a clear abuse by its nature, and (ii) be difficult to detect, if adopted. Otherwise, the merger would apparently be cleared.
4. The test is rather confusing when considered in abstract, but given the existing case law, it is difficult to see a clear cut solution to the paradoxical nature of the principle.

What happened to the brand portfolio effects?

1. The words “portfolio effect” only appear once in the Guidelines. However, the Commission has developed in its case law a “theory” of portfolio effects which appears to be confusing at present, due to the conflicting points of view which have been adopted in different decisions in the past. The portfolio theory was originally developed with respect to the combination of a broad portfolio of brands in alcoholic beverages (6).
2. The Commission considered at the time that a large portfolio of brands conferred to the merged entity a number of competitive advantages that would result in the weakening of all remaining competitors and their progressive marginalisation in the markets. The approach was subsequently refined, and the Commission established a rule of thumb by which it would not consider the existence of market power deriving from a portfolio of brands unless the markets where the merged entity would enjoy a market share of 40% or more, would together represent at least 35% of the aggregated sales of the merging parties (7). The Commission also considered in this last decision that it had found no evidence that secondary brands within the portfolio would allow to strategically act with the aim to marginalise residual competitors.
3. The Commission has more recently carried out a much more rigorous analysis of brand portfolio effects or conglomerate effects, taking into account, in addition, the possibility that economic efficiencies are created through combining brands in different markets (8). The Commission analyzed in this case possible anticompetitive foreclosure in terms of (i) P&G bundling together its range of products or (ii) P&G using its strategic position as category manager.
4. As regards the bundling together of products, the Commission considered two possibilities (i) P&G could make the sale of its leading brands conditional on the purchase by the distributor of other weaker or secondary brands (“pure bundling”), or (ii) P&G could encourage the purchase of its secondary brands through discounts on a range of products to distributors or the carrying out of joint promotions (“mixed bundling”).
5. The Commission considered, in line with recent microeconomic theory, that an anticompetitive effect is particularly likely to occur if the products in question are complementary in the strictly economic sense of the term, which was not he case between P&G and Gillette line of products. Furthermore, the Commission took into account the following factors indicating that there were not concerns on foreclosure: (i) range discounts were directly related to economies in production and in particular, transport and distribution (ii) there was little sense in cross-promotional activity across different categories of products (iii) there were other competitors with similar product ranges, who offered a viable alternative (iv) there was a sufficient customer countervailing power, as reflected by the use of private label and delisting of leading brands.The factors described above are of a fairly general nature, while sufficiently concrete, and they could perhaps be explicitly mentioned in the Guidelines as a set of circumstances in which a conglomerate merger would not raise competition concerns on grounds of anticompetitive foreclosure.

Product market competition and financial structure.

1. There is a consolidated body of economic literature (9) that has found a relation between the financial structure of firms and its competitive behaviour in product markets. In particular, firms may foreclosure financially weaker rivals by engaging in financial predation. Even if the Commission has not had the opportunity to examine financial predation in a concrete case, perhaps it would be useful to include a reference to financial predation in the Guidelines._______________________________________________________________________

1. “Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings”, OJ C 31, 5.2.2004.
2. Case M.4000 INCO / FALCONBRIDGE.
3. As the Commission acknowledged in case M.3653 - SIEMENS/VA TECH.
4. The Commission analysed such a situation in case M.3535 - VAN DRIE / SCHILS. Case M.1879 - BOEING / HUGHES.
5. Case M.1879 - BOEING / HUGHES.
6. Cases M.938 - GUINNESS / GRAND METROPOLITAN and M.2268 - PERNOD RICARD / DIAGEO / SEAGRAM SPIRITS.
7. Case M.3779 - PERNOD RICARD / ALLIED DOMECQ.
8. Case M.3732 - PROCTER & GAMBLE / GILLETTE, which for its detailed assessment of competition, the rigour of the economic analysis, and its relatively recent date (15/07/2005) could be regarded as one of the most important Decisions as far as conglomerate effects are concerned, and a reference to this decision could perhaps be included in the Guidelines.
9. See for instante, Aghion, Dewatripont and Rey (1998) Ceston, Fumagalli (2001).

This entry was posted on Monday, April 9th, 2007 at 3:47 pm and is filed under e-konomica articles. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

Previous: Excessive prices - VISA - Multilateral interchange fees |