In December 2005 the Court of First Instance of the European Communities (CFI) delivered its long-awaited judgment on General Electric's appeal against the Commission's prohibition on its anticipated merger with Honeywell International. The 2001 Commission decision effectively blocked a deal, waved through by the US authorities and worth $42 billion, that would have resulted in the largest industrial combination in the world. To the Commission's relief, the CFI upheld the prohibition decision, albeit narrowly. The judgment contains some important lessons and has implications for merging parties whose deals must be assessed in Brussels.
The Commission's decision analyzed the numerous product markets in the aerospace and power systems industries affected by the transaction and rested on three theories of harm: horizontal effects, vertical effects and conglomerate effects.
The prohibition in GE/Honeywell is widely regarded as resulting mainly from the merger's vertical and conglomerate effects. Vertical effects (such as foreclosing competitors from the market) typically arise in mergers between companies that operate at different levels of the production or distribution chain (here, for instance, GE's engines and Honeywell's engine starters). Conglomerate effects may result from the aggregation of competitive strengths in different markets that are more or less closely related (typically portfolio power or the accumulation of technical advantages and of economic and financial strength in the merged entity). According to the Commission, the conglomerate effects in this case would have been the result of the combination of non-overlapping activities enabling or even facilitating GE's use, in particular, of its financial and commercial strength and of packaged offers concerning the newly expanded product portfolio (for example, bundled sales of GE's engines with Honeywell's avionics and non-avionics products).
On these crucial issues the CFI found that the Commission had committed serious errors of assessment. In particular, the Commission's evidence did not establish convincingly and with enough probability that GE would have extended the use of its financial and commercial muscle to Honeywell's markets thus forcing rivals out of them. Nor was the evidence sufficient to prove that the merged entity would have engaged in bundling practices. Further, the Commission failed to take into account the deterrent effect of Article 82 of the EC Treaty (prohibiting the abuse of dominance) when considering the future likelihood that GE would engage in practices such as bundling.
The CFI only affirmed the Commission's assessment in its analysis of the horizontal effects in those markets where a dominant position would have been created or strengthened. The correctness of this analysis was sufficient for the CFI to uphold the prohibition.
The judgment is important in several respects. First, it confirms that where the theory of harm is based on the merged entity's future conduct, the supporting evidence must be especially strong and convincing. This is particularly so in conglomerate and vertical effects cases, where the anticompetitive effects giving rise to the Commission concerns are not the immediate result of the merger, but will occur should the merged entity engage in certain conduct. In GE/Honeywell, to prove future behaviour the Commission mostly relied on past conduct patterns and on "pretty cutting-edge academic analysis", as some commentators have already put it. The CFI noted that evidence of past conduct may not be enough to establish to the requisite legal standard that the merged entity would engage in certain conduct in the future, though it acknowledged that to prove the ability and the incentive to behave in a given manner the Commission may rely on the merged entity's commercial interests and logic. However, the ruling clearly shows that even this can prove difficult, since the Commission may conclude that "it is likely that the merged entity will actually engage in the conduct foreseen" if "it is obvious that the commercial interests of an undertaking militate predominantly in favour of a given course of conduct". It would appear therefore that to satisfy this standard the Commission will have to base its conclusions on a robust economic analysis (that is, the empirical analysis must be detailed and based on an economic model that is well suited to the specific case).
As a result of the high evidentiary burden thus imposed on it, the Commission is likely to exercise further caution when blocking mergers in more complex cases, such as those presenting vertical and conglomerate issues. In turn, the need for a qualitatively high body of evidence, comprising robust economic analysis, will require a more intensive fact-finding investigation on the part of the Commission, resulting in a more cumbersome merger control process. It is likely that merging parties will be increasingly required to produce more and better evidence (including economic studies produced by specialized consultants) both in the pre-notification phase and during the actual investigation. Already the pre-notification phase can last several months. Similarly, third parties opposing a merger will find it difficult to assist the Commission in finding evidence to support its possible objections.
Second, regarding the need to consider the deterrent effect of Article 82 EC when assessing the likelihood of future anticompetitive conduct by the merged entity, the CFI judgment brings more rigour in an area where the European Court of Justice is seemingly less stringent. In the landmark ruling of February 15 2005 on the Tetra Lava/Sidel merger decision, the ECJ had dismissed the notion that, in assessing the likelihood of future behaviour in a conglomerate case, the Commission must consider whether a violation of Article 82 EC is likely, and ascertain that it would be penalized by the Commission and in the various member states. As a result of the need to consider Article 82 EC's deterrent effects, the complexity of the burden of proof imposed on the Commission has increased. This deterrence can override any finding (based on past experience, empirical analysis or company documents) that the merged entity would have the ability and economic incentive to engage in anticompetitive behaviour.
A third important lesson with serious implications for those considering challenging a Commission decision before the CFI is that to succeed the applicant must prove that the Commission's findings are wrong with respect to all the affected markets. GE argued that the various markets of the aerospace and avionics industries were closely interconnected and that, therefore, the elements of the reasoning of the decision reinforced one another. Accordingly, it would be artificial to analyze each aspect of the reasoning in isolation. In substance, GE submitted that if some of the grounds relied on by the Commission were unfounded, it was not for the Court to determine whether the well-founded grounds were sufficient to support the operative part of the decision. The CFI disagreed and held that, where some of the grounds (in this case, those on the horizontal effects) provide a sufficient legal basis for a decision, any errors in the other grounds of the decision do not affect its operative part. Thus, the greater the number of affected markets the more difficult it will be to overturn a merger decision.
The Commission welcomed the ruling as a vindication of its course of action in its most famous merger decision. On balance, though, it is questionable whether as a result of the reasoning expressed by the Court the judgment is really favourable to the Commission. It would appear that the already high standard of the burden of proof for the Commission has been further increased, particularly in the most complex cases.
Matteo F Bay is a partner and Javier Ruiz Calzado of counsel at Latham & Watkins in Brussels.