Fines against Hard Core Cartels in Europe: The Myth of Overenforcement

Abstract

This article compares the level of fines actually imposed on cartel participants to the illicit gains captured by the firms and estimates a range of optimal restitution and dissuasive fines in each case. The results show that the fines imposed against cartels by the European Commission are, overall, moderate, regardless of the probability of detection. The article is based on a sample of sixty-four cartel decisions by the European Commission from 1975 to 2009 and a methodology that estimates optimal fines imposed on cartels on a case-by-case basis. /staff/connor/papers/Optimal_Deterrence.pdf. Connor concludes that on average, fines imposed upon international cartels range from 3.9% to 31.2% of the illegal profit realized by the undertakings. Other studies use case-by-case microeconomic analysis. Recent empirical studies of fines against cartels in Europe examine how fines are fixed in practice under European competition law, particularly under the 1998 Penalty Guidelines, 5 but not how they should have been determined so as to prevent the formation of cartels. 6Our article presents an empirical methodology based on easily computable microeconomic parameters, which allows us to estimate a range of optimal fines. More precisely, by analyzing sixty-four cartels condemned by the European Union over the period 1975-2009, we compare the sanction actually imposed to the illicit gain captured by the firms and estimate a range of restitution fines (amounting to the illegal profit, which corresponds to optimal fines given a 100% probability of detection) and dissuasive fines (which correspond to an optimal fine given that some cartels remain undetected). This comparative analysis allows us to assess over time the efficiency of the Commission's antitrust enforcement in preventing explicit collusion. We do not discuss comprehensively how fines are actually fixed in practice, as this article focuses on the ability of fines to deter the formation of cartels in Europe.

The Determination of Optimal Fines in Cartel Cases: The Myth of Underdeterrence

“…We have shown that the dissuasive benchmark measured by Combe and Monnier (2010) should be reduced by a factor no smaller than 10. This suggests that, considering the sample of 64 cartels studied by Combe and Monnier, 65% of the fines imposed by the European Commission in recent years are above the properly defined restitutive benchmark and 56% of them are above the properly defined dissuasive benchmark.…”

“…The following graph illustrates the basic intuitions behind the determination of the excess 10 See also Werden and Simon (1987), Gallo et al (1994) among others, and more recently Combe and Monnier (2010). 11 One must be careful in using such a tool as the parameters ε (price elasticity of demand), m (competitive markup and k (cartel price overcharge) are not independent: both m and k depend on ε (the price elasticity of demand is an important determinant of market power in a market), and a larger m will in general imply a lower k (if the competitive mark-up is already high, the cartel overcharge will tend to be lower).…”

The Determination of Optimal Fines in Cartel Cases: The Myth of Underdeterrence

“…The following graph illustrates the basic intuitions behind the determination of the excess 10 See also Werden and Simon (1987), Gallo et al (1994) among others, and more recently Combe and Monnier (2010). 11 One must be careful in using such a tool as the parameters ε (price elasticity of demand), m (competitive markup and k (cartel price overcharge) are not independent: both m and k depend on ε (the price elasticity of demand is an important determinant of market power in a market), and a larger m will in general imply a lower k (if the competitive mark-up is already high, the cartel overcharge will tend to be lower).…”

“…However, after collecting and examining a larger data set, Connor (2010) concludes that "...penalty guidelines aimed at optimally deterring cartels ought to be increased". Combe and Monnier (2010) consider 64 recently prosecuted cartel cases in the EU and contribute to filling the gap between evaluating cartel overcharges giving rise to illicit gains and defining and computing "optimal" restitutive and dissuasive fine benchmarks, in part directly from the court evidence and in part based on Connor's reviews of cartel overcharges. They use the latter benchmarks to gauge the actual fines inflicted to the 64 cartels they considered.…”

“…Indeed, the issue of cartel stability and the role of fine in deterring the formation of cartels must be analyzed in a dynamic framework. We will show that the proper and explicit consideration of such a dynamic framework, together with the determination of representative overcharge estimates properly corrected for model errors, estimation errors and publication bias, have a major impact on the determination of optimal dissuasive fines: the restitutive and dissuasive benchmarks measured by Combe and Monnier (2010) should be reduced by factors no smaller than two and ten respectively. This suggests that, considering the sample of cartels studied by Combe and Monnier, approximately two thirds of the fines imposed are above the properly defined restitutive benchmark and at least half of them are above the properly defined dissuasive benchmark.…”