The EC Fining Policy for Violations of Competition Law: An Empirical Review of the Commission Decisional Practice and the Community Courts' Judgments

Abstract

Fines represent the principal tool in the European Commission's enforcement of EC competition law. Unlike in the United States where there is a formidable congeries of weapons against undertakings which breach anti-trust law, there are no criminal penalties, such as imprisonment for individuals in the EC. Moreover, private enforcement of EC competition law is still minimal. Thus, fines represent the main tool to remedy and deter violations of competition law. The European Court of Justice indicated in Musique Diffusion France (Pioneer), that the underlying rationale for the imposition of fines is to ensure the implementation of Community competition policy. The meting out of fines, therefore, serves two objectives (i) the suppression of illegal activity and (ii) the prevention of recidivism. During the first three decades in which the Commission imposed fines for breaches of EC competition law, the Commission was criticized for the obfuscation surrounding how it determined a given fine. During this period, there were no guidelines providing a reference point from which the Commission could impose fines leading to a lack of transparency in the fining process. There was thus a tendency to litigate before the courts in the expectation that the fine would be reduced. In addition, fines were generally fixed at such a low level that it was questionable whether they had any deterrent effect. There has been a recent evolution in Commission fining policy, however. First, the promulgation of both the Commission Guidelines on fines in 1998, which aims to make decisions over fines more transparent and impartial. Second, the toughening of the fines, which is particularly evident when one notes the condign fines of 462 million euros and 497 million euros imposed on Hoffman-La-Roche and Microsoft. Third, the development of the leniency notice, which provides an incentive for cartel members to admit to their anti-competitive conduct. Since the adoption of the 1998 Guidelines, the majority of the fines imposed by the Commission have been for cartel activity. The Commission has, however, shown an increasingly heavy-handed approach towards other infringements of Article 81 EC and abuses of a dominant position under Article 82 EC. Yet, it is not quite sure that these evolutions have reached their objectives as both the constituent elements of the 1998 Guidelines and the fining decisions, which are based on the 1998 Guidelines, are vague. This has left much room for conjecture as to how the Commission reached the final fine. The corollary of this is that there has been, as in the period preceding the 1998 Guidelines, a steady yet significant number of parties litigating before the courts. It is also still open to debate whether the fines imposed by the Commission are stringent enough. The Microsoft decision bears testimony to this. This raises the issue of whether the EC should not turn to other forms of penalties, such as criminal penalties. This path is already being followed in some Member States (e.g., UK), but seems unlikely to be followed in the EC. The Commission can neither impose fines nor criminal sanctions on individuals in light of the wording of Article 81 EC. On the other hand, Article 83(1) EC stipulates that the Council "give effect to the principles set out in Articles 81 and 82". This could be interpreted as encompassing sanctions on individuals as the effect of this would be to enhance the deterrent effect of the cartel prohibition. Article 23(5) of Regulation 1/2003 states, however, that decisions are not to be of a criminal law nature. The main purpose of the article is to provide a detailed analysis of the parameters taken into account by the Commission when imposing a fine, as well as the parameters used by the Court of First Instance when reviewing fines imposed by the Commission. In order to do this, we have reviewed all the Commission decisions and CFI judgments dealing with fines, which have been adopted since the publication of the 1998 guidelines. For each Commission and CFI judgment, we have identified the factors that have been taken into account to determine/review the fines imposed for infringements of EC competition law. The results of our analysis are summarized into two tables (one for the Commission decisions and one for the CFI judgments), which allow the reader to find for each case the factors that have been taken into account to determine/review the fines. This is the empirical side of the paper. While most of the papers analyzing the fining policy of the Commission discuss factors, such as the gravity or duration of an infringement, the presence of various mitigating circumstances, in a rather general or theoretical fashion, this paper provides precise data as to the elements that are most/least likely to be considered in the determination/review of fines. Thus, our table on the Commission decisions will, for instance, allow the reader to know in which cases, a cartel member was the leader and/or imposed coercive measures on other cartel members and to what extent this was considered as an aggravating circumstance. In turn, our table on the CFI judgments will allow the reader to identify the various reasons why, in a given case, the fine imposed by the Commission was reduced before the CFI. Another aim of the paper is to give a critical look at the Commission decisions imposing fines to see whether the reasoning on which there are based is coherent. As will be seen, it is often difficult to understand the logic of the fines imposed by the Commission. Identical factual scenarios will be treated differently, while different factual scenarios will be offer the same treatment. By contrast, we will not deal with theoretical issues, such as the optimal level of the fines or whether criminalization of competition law violations is desirable as there is abundant literature on this.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 51 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.