The Counterfactual Method in EU Competition Law: The Cornerstone of the Effects-Based Approach
24 Pages Posted: 12 Dec 2011
Date Written: December 11, 2011
In Book IX of his History of Rome (written around 25 BC) Titus Livy speculates about a hypothetical confrontation between Rome and Alexander the Great. What if Alexander had not died at the end of the Asian campaign but had returned to Europe to attack Rome? Livy argues that Rome and Carthage would have joined forces to crush the Macedonian army.
Livy’s musings represent an early example of the counterfactual method. This method can be used to assess the effects of an actual or a hypothetical event. The counterfactual describes the world in the absence of that event. If the event has already occurred, one needs to build an alternative past (this is what Livy does when he imagines Alexander’s return to Europe). If the event has not yet taken place, it is necessary to contemplate an alternative future.
The use of the counterfactual under EU competition law goes back to the seminal judgment of the Court of Justice in the Société Technique Minière case. However, until recently, this method was confined to the area of merger control. Under Articles 101 and 102 TFEU, the European Commission and the EU Courts initially adopted a “form-based” approach. This approach paid limited (if any) attention to the effects of the relevant agreement or conduct on competition and consumers. Because the aim of the counterfactual technique is to analyse the effects of a given event it had little relevance under the form-based approach.
In recent years the Commission has transitioned towards the effects-based approach. The modernisation of EU antitrust enforcement caused a renewed interest in the counterfactual technique. Counterfactuals are discussed in various Article 101 guidelines and in the Article 102 Guidance Paper. In June 2011 the Commission published a draft Guidance Paper on quantification of antitrust damages, which contains a detailed analysis of various counterfactuals.
In this paper, we examine the use of the counterfactual method in EU competition law. In our analysis, we distinguish between ex ante control (merger control and Article 101 self-assessment) and ex post scenarios (investigations under Articles 101 and 102, damages litigation).
In Section II, we examine ex ante counterfactuals. We conclude that these counterfactuals are relatively easy to establish because they are usually based on the status quo ante. For example, in merger control the Commission compares the hypothetical post-transaction world with the actual pre-transaction situation. However, the development of prospective analysis leads to more sophisticated counterfactuals which incorporate future events, such as the target’s bankruptcy in the absence of the transaction. This forward-looking counterfactual should be based on highly likely future events; the Commission and the parties should not contemplate hypothetical scenarios which cannot be predicted with a high degree of certainty.
In Section III, we consider ex post counterfactuals. These counterfactuals are much more difficult to build because they are intrinsically speculative and are always based on a hypothetical scenario. In its recent decisional practice under Article 102 TFEU the Commission attempts to demonstrate anticompetitive effects and consumer harm by relying on various comparators (e.g., by comparing prices that prevail in the affected market to average EU-wide and OECD-wide prices). We believe that this benchmarking does not amount to the “appropriate counterfactual” advocated by the Guidance Paper. The Commission should use more elaborate and robust techniques, including economic models that simulate the likely market outcome in the absence of the alleged infringement. We also criticise the Commission for relying on the concept of “infringement by object” to avoid using the counterfactual technique under Article 101 TFEU.
Keywords: competition, antitrust, damages, EU, counterfactual, effects-based analysis, foreclosure, anticompetitive agreements
JEL Classification: K00, K21, K40, L40
Suggested Citation: Suggested Citation