CPI EU News Column edited by Thibault Schrepel, Sam Sadden & Jan Roth (CPI) presents:
Cartel damages litigation is of increasing importance in the EU, evolving into a second main pillar of competition law enforcement besides the agencies’ activities. While at least in follow-on litigation, the violation of competition law as such is oftentimes relatively easy to establish, the showing and quantifying of economic harm can be very challenging. One of the typical hurdles consists in the so-called “passing-on defence,” i.e. the defendant’s argument that the claimant did not suffer lasting harm because it could shift a cartel-caused overcharge down the supply chain by raising its own prices. Passing-on can, however, also be invoked as a “sword” by an indirect purchaser who claims damages due to the passed-on overcharge.
Article 12 of the Directive 2014/104/EU (Cartel Damages Directive/CDD) stipulates that the passing-on defence is admissible, that the claimant shall nonetheless have the right to “claim and obtain compensation for loss of profits [also called “volume effect”] due to a full or partial passing-on of the overcharge,” and that “Member States shall ensure that the national courts have the power to estimate, in accordance with national procedures, the share of any overcharge that was passed on.” Article 16 of the Cartel Damages Directive takes up this last point by obliging the Commission to issue guidelines for national courts on how to estimate passing-on. Based on long and thorough preparatory work, including targeted stakeholder consultation and a study on the topic in 2016,2 the Commission has now published its “Guidelines for national courts on how to estimate the share of overcharge which was passed on to the indirect purchaser” (hereinafter: Passing-on Guidelines/POG). These Guidelines complement the Commission’s “Practical Guide” on quantifying harm in actions for damages based on breaches of Article 101 or 102 TFEU. While the Practical Guide focuses on the overcharge, the Passing-on Guidelines specifically address the passing-on of such overcharge.
1. Introduction and Economic Theory
The Guidelines’ main purpose is to set out the economic methods for determining passing-on, as well as guidance on identification and evaluation of relevant evidence, including disclosure requests, party statements and economic expert3 opinion. After an introduction explaining the passing-on effect in various constellations, sketching the rules on burden of proof (including the passing-on presumption in Article 14(2) CDD), as well as types and disclosure of evidence, the Guidelines turn, therefore, to the economic theory of passing-on. They set out and explain (including model case examples) key factors for the economic assessment, namely the respective nature of input costs and product demand, the intensity of competition in the customers’ markets, and a number of other elements, such as price adjustment costs or buyer power. As one principle of general importance, the POG state, in this section, that “it should be noted that an estimation of the total harm by simply subtracting the passing-on related price effect from the overcharge effect would lead to an underestimation of the harm suffered by the direct or indirect customer if the volume effect is not taken into account.”4 In a subsequent section, the POG even go so far as to contend that “[w]hen there is a passing-on related price effect, this necessarily implies, at least in theory, a volume effect.”5 Given this strong position, it would have been interesting to also read about the economic theory which uses determined overcharge as a lower bound for damages in cases involving volume-based effects.6
For quantifying passing-on and volume effects, the Guidelines suggest, while leaving room for other approaches (e.g. simulation), a three-step assessment which looks first at the overcharge effect, then at the passing-on related price effect, and finally at the passing-on related volume effect. One important part of all three steps is the quest for the counterfactual, i.e. the attempt to determine the (hypothetical) economic situation absent the infringement, because the delta between the claimant’s economic situation in the infringement-afflicted reality – including a potential passing-on – and in the infringement-free counterfactual yields a core component for assessing claimant’s harm.
For the determination of overcharge and passing-on related price effects, the POG favor a comparator-based approach, viz. a comparison between the economic parameters in the affected market and the same economic parameters in other real, unafflicted and comparable markets. This comparative exercise can be carried out either by looking at the affected market during infringement and non-infringement periods (before-during-after approach); or by comparing the affected market with different, but similar regional or product markets (cross-sectional approach); or by a combination of comparisons over time and comparisons across markets (difference-in-differences approach), which is the most reliable but also the most demanding method in terms of complexity and required data quality.
In spite of their support for comparator-based techniques, the POG remain mindful of the approach’s limitations: It can be hard to decide whether other markets are similar enough to serve as comparables; adjacent, comparable markets can be distorted by indirect infringement effects themselves; a continuation of supra-competitive price equilibria, time lags due to differing pricing terms on different levels of the chain of production and distribution can blur the delineation between infringement-affected and ‑unaffected periods in the market at issue. The adjustment of comparator data – in particular by way of regression analysis if data availability and data quality permit – is one way to mitigate these uncertainties. Other, pragmatic measures include a demanding threshold for statistical significance,7 safety discounts on the exactitude of econometric results in general, or a complementing consideration of qualitative evidence (e.g. records directly showing passing-on). Where available data do not support a comparator-based approach, other methods are available, for instance economic modelling or the passing-on rate approach which looks at the overall likelihood that a market participant passes on changes in input factors. Importantly, the POG underline that lack of sufficient data can also result from a court finding that extensive disclosure would be disproportionate given the facts of the case (e.g. the value of the claim in question).
So far, courts have – as the POG rightly point out – engaged only rarely in the quantification of volume-based effects, although there is much methodological overlap with the assessment of overcharge and price effects. The comparator-based approach is valid for volume-based effects as well, even though the relevant parameters take a specific form and include “(i) the observed quantity sold by the firm affected by the overcharge, (ii) the counterfactual volume sold and (iii) the price-cost margin that would have been achieved by the purchaser in the absence of the infringement.”8 Where comparative analysis fails, courts may consider qualitative evidence or – a test more specific to volume-based effects – a combination of observed price increases and the estimated price sensitivity/price elasticity of the relevant demand.
In two annexes, the Guidelines provide more details on some of the economic concepts suggested for assessing passing-on, as well as a helpful glossary of key terms.
A Few Comments
The Passing-on Guidelines do not bind national or EU courts, but it is very likely they will impact case-law on both levels, not least because the extensive references to Member State case-law make them a repository on the Union-wide state of the law in this area.
Not all parts of the POG are equally insightful. For instance, the reader may wonder what additional information to gather from the visualization of the before-during-after approach:
Overall, though, the Guidelines provide a helpful overview on key economic concepts in the field. The POG’s – moderate – penchant for the comparator-based approach and, arguably, for non-econometric, qualitative evidence are certainly appropriate for many cases. Nonetheless, courts should also remember the Guidelines’ statement that “there is no technique that could be singled out as the one that would in all cases be more appropriate than others.”9 In fact, the Guidelines themselves contain strong passages on a pragmatic approach to selecting, combining and adjusting methods and evidence as the circumstances of the case request.
Importantly, this communication of the Commission is far from covering all relevant aspects of passing-on. Claim cumulation within or across Member State jurisdictions, for instance, is merely touched upon and non-price effects are explicitly excluded from the assessment. Neither do the POG sufficiently address the relation between available qualitative and quantitative data, disclosure and prima facie evidence, although (the threshold for) the acceptance of prima facie evidence on the economic effects of a cartel in the absence of reliable data, the ensuing shift of the burden of proof, and the preferability of this mechanism over broad disclosure10 particularly matters in the passing-on context. The Guidelines are right to stress the importance of, hitherto somewhat neglected,11 volume-based effects. In their consideration of such effects they strongly focus, however, on the one or two levels immediately beneath the cartelizing undertakings in the chain of production and distribution. Just as the preparatory 2016 study,12 the POG thereby omit the volume-based effects on further undertakings which, for instance, supply the cartelizers or sell complementary goods.13 Obviously, measuring and evidentiary difficulties may frequently forbid cartel damages litigation based on such effects. At least the theoretical framework should be broad enough, however, to encompass all major “ripples of harm”14 a cartel can send through the economy. With regard to overcharge and price effects, the analysis is largely based on a single pass-on sale scenario, omitting the particularities e.g. of transfer of use contracts (renting, leasing, etc.) and of scenarios in which part of the (potential) passing-on lies, at the time of the damages litigation, still in the future (problem: consideration of lost/gained chances?). All in all, courts will, therefore, remain in a pioneering role regarding certain aspects of passing-on. For others, however, clearer guidance is now at their hands.
1 Chair for Economic Law, Chairman Center for Intellectual Property and Competition Law (CIPCO), University of Zurich; Fellow Max Planck Institute for Innovation and Competition, Munich; Of-Counsel, Schellenberg Wittmer, Zurich. For more information, see: https://www.ius.uzh.ch/de/staff/professorships/alphabetical/picht/person.html.
2 EU Commission (2016), Study on the Passing-On of Overcharges, para. 20, http://ec.europa.eu/competition/publications/reports/KD0216916ENN.pdf.
3 Subsequent passages of the Guidelines provide some more details on the handling of economic experts by courts, including interesting language on court-induced collaboration between parties’ experts and on parties’ right to be heard regarding court-appointed experts. Not much is said, however, on the concept of a neutral expert consensually named by the parties, cf. on this Kruse, Wege einer ökonomischen Beweisführung in Kartellverfahren, NZKart 2017, 432.
4 Para. 58 POG.
5 Para. 134 POG.
6 On this concept, see for instance Hellwig, Private Damage Claims and the Passing-On Defense in Horizontal Price-Fixing Cases: An Economist’s Perspective (September 2006), MPI Collective Goods Preprint No. 2006/22.
7 In this respect, the POG seem to champion a 95 percent confidence interval, allowing for a 5 percent risk of Type 1 errors (erroneous assumption of passing-on), see para. 105 POG.
8 Para. 139 POG.
9 Para. 155 POG.
10 Examples from the German case-law on the topic are Federal Court of Justice, 28.6.2011, KZR 75/10; Kammergericht, 1.10.2009, 2 U 10/03 Kart; Higher Regional Court Karlsruhe, 31.7.2013, 6 U 51/12 Kart.; 31.7.2013, 6 U 204/15; Regional Court Dortmund, 21.12.2016, 80 O 90/14 [Kart].
11 Schwalbe, Lucrum Cessans und Schäden durch Kartelle bei Zulieferern, Herstellern von Komplementärgütern sowie weiteren Parteien, NZKart 2017, 157; Hellwig, Private Damage Claims and the Passing-On Defense in Horizontal Price-Fixing Cases: An Economist’s Perspective (September 2006), MPI Collective Goods Preprint No. 2006/22.
12 EU Commission (2016), Study on the Passing-On of Overcharges, para. 20.
13 On such effects, see Schwalbe, Lucrum Cessans und Schäden durch Kartelle bei Zulieferern, Herstellern von Komplementärgütern sowie weiteren Parteien, NZKart 2017, 157; Han, Schinkel, Tuinstra (2009), The Overcharge as a Measure for Antitrust Damages, Amsterdam Center for Law & Economics Working Paper, 2008-08; E. Bueren, F. Smuda (2014), A Primer on Damages of Cartel Suppliers – Determinants, Standing US vs. EU and Econometric Estimation, ZEW Discussion Paper No. 13-063. Constitutive, and showing the force of theoretical vis-à-vis comparator-based approaches: Verboven, van Dijk, Cartel Damages Claims and the Passing-on Defense, The Journal of Industrial Economics, Vol. 57, No. 3 (Sep., 2009), pp. 457-491.
14 Associated General Contractors v. California State Council of Carpenters, 459 U. S. 519, 534, 103 SCt. 897, 907.