The Pros and Cons of Rebates

The seminar, moderated by Professor Giorgio Monti, discussed the question of how the competitive assessment of rebates has evolved and how the more economic approach is reflected in the assessment of rebates.

2018 year's Pros and Cons seminar addressed the issue of Rebates.

Contributors (from back, left): Olli Kauppi, John Moore, Jorge Padilla, Lucas Peeperkorn, Alexis Walckiers, Johan Sahl. (Front, left) Corina Wahlin, Pinar Akman, Chiara Fumagalli, Doris Hildebrand, Karin Lunning, Giorgio Monti, Paul Nihoul.

Summary of the seminar

Prof Giorgio Monti of the European University Institute welcomed this year’s speakers and expressed that this year’s topic on rebates was both relevant and important to discuss. In the light of the latest legal developments in the area, he kicked off the debate by putting forth the following remarks. Is the outcome in the Intel case satisfactory? What is the As Efficient Competitor test (AEC test) and is it necessary to conduct it in order to reach a finding of abuse? As regards compliance Prof Monti noted that it is to this day still unclear whether the Court’s judgment in the Intel case answered the question of whether the conduct in question was abusive or not. Quoting the saying “justice delayed is justice denied” and posed the question of whether we are doing enough in order to have an administrative process that balances accuracy and administrability and that delivers justice that is somewhat speedier. In concluding, he reflected on the question of whether we are so concerned of over-enforcing that we are in fact under-enforcing.

The role of judges in economic law

Judge Paul Nihoul, General Court of the European Union, commenced this year’s seminar with a keynote speech. In his keynote speech, he commented on the role of judges in economic law by presenting the framework in which the courts have to operate. The Rule of law creates a strict framework that binds the courts. It is central to understand the concept of the Rule of law in order to understand how the EU-Courts reason and in order to understand the outcome in the Intel-case. This framework means that the judge has to abide by the Rule of law which can explained by an equation of three variables, formal aspects, jurisprudence and the inner morality of the law in terms of transparency, proportionality and the principles of legitimate expectations and material aspects. He went on to comment in what way these variables determined the outcome in the Intel-case. In concluding, he commented on the role of judges in developing the law through case law.

The As Efficient Competitor Test

Dr Jorge Padilla, Senior Managing Director at Compass Lexecon, spoke about the as efficient competitor test (“AEC test”) and the importance of considering what it is, its structure and implications. He highlighted the necessity of outlining who the “as efficient competitor” is since in some cases the incumbent may have significant advantages i.e. relating to switching costs, network effects, capacity constraints and vertical integration. Dr Padilla defined the “as efficient competitor” as someone who owns both the tangible and intangible assets needed to operate efficiently in the affected market. He went on to stress that even though as efficient competitors perhaps may not respond in the exact same way to their competitors conduct, they might still be able to respond by other means.

According to Dr Padilla, the AEC test should be seen as consisting of two distinct parts. It should be considered as being both a matching test, asking whether the competitor can offer a similar scheme, and as an attribution test, testing the ability to compete effectively by reducing their prices in other ways. By defining as anticompetitive the conduct of a dominant undertaking that cumulatively distorts the competitive process and reduces consumer welfare, Dr Padilla argued that the competitive process is distorted when an as efficient competitor cannot react profitably to the conduct in question. He also discussed the need to measure the contestable share and the associated difficulties, noting that the definition of the contestable share must follow from the theory of harm. In his concluding remarks, Dr Padilla commented on policy implications, noted that the AEC test, viewed as a replicability test, is indispensable to the assessment and cannot be separated from the concept of abuse.

The discussant Olli Kauppi, Chief Economist at the Finnish Competition and Consumer Authority, picked up from the point made by Dr Padilla that the AEC test could be viewed as consisting of two parts. He went on to state that, regarding distinction made between the two standards, harm to the competitive process and consumer harm, the latter should take precedence over the former in the matter of rebates. He thereafter discussed the issue of investigating all relevant circumstances and the potential dangers associated with the list in the Commission’s guidance since the guidance could be interpreted as too strict a guide. He cautioned against congruence and confirmation biases, since they could result in either giving too much, and perhaps even equal, weight to all of the factors that are included in the list, and not enough attention to the particular factors that are relevant in a specific case.

On the use of price-cost tests in loyalty rebates and exclusive dealing arrangements

Prof Chiara Fumagalli from Bocconi University tackled the second topic of the conference, “On the use of price-cost tests in loyalty rebates and exclusive dealing arrangements”. In her opening remarks, Prof Fumagalli mentioned that she would be reflecting on a paper co-written with Prof Massimo Motta reviewing the existing literature on exclusive practices with the aim to find guidance regarding the question of whether prices above costs should be a safe harbour for loyalty/exclusivity rebates. She mentioned that she had a slightly different take on the matter from Dr Padilla since she views price cost tests mostly as a test for profit sacrifice and went on to state that her conclusions in some aspects converged to those of Dr Padilla.

As a first point, Prof Fumagalli stated that economic theory does not predict that the incumbent should suffer losses for predation to occur, rather the crucial point is that the rival’s profits need to be limited and the performance needs to be undermined for exclusion to occur. Prof went on to elaborate on the theory of predation based on scale economies and concluded that economic theory does not allow us to conclude that what distinguishes predatory from non-predatory prices is that the incumbent is suffering losses. As Prof Fumagalli put it, price-cost tests are used in order to avoid the risk of chilling competition. According to Prof Fumagalli, it is therefore more important to spell out the theory of harm in predation cases and she went on to state that price-cost tests are just a piece of evidence that complement the theory of harm.

In her concluding remarks, Prof Fumagalli stated that what distinguishes predation from exclusive dealing or loyalty rebates is not whether the incumbent incurs losses. Furthermore, it is not the presence of the incumbent’s losses that necessarily distinguish between abusive and non-abusive conduct. According to Prof Fumagalli the literature suggests that the anti-competitive potential of exclusive dealing and loyalty rebates is stronger than that of predation thus concluding that it is reasonable to treat the former practices differently from the latter. Prof Fumagalli concluded that even though above-cost pricing is not an appropriate safe harbour, price-cost tests are still relevant and informative for the assessment of exclusive dealing and loyalty rebates because their outcome is a piece of evidence that must go hand-in-hand with the theory of harm.

Picking up from one of the main points made by Prof Fumagalli that although it is imperfect, a price-cost test can be an important addition to a coherent theory of harm, John Moore from the French Competition Authority (FCA), he ventured to complement the theories presented by Prof Fumagalli with experience drawn from practice. He began by commenting the implications of the Intel ruling stating that it clarifies that the Commission cannot adopt a form-based approach when dealing with exclusive dealing and loyalty rebates. He then went on to present the FCA’s latest experience and challenges in applying the price-cost test in its recent loyalty/retroactive rebates cases. One of the challenges faced is measurement issues with complications arising in particular when the test does not need to be applied to all consumers and when dealing with individualised rebates. Commenting on the distinction made by Prof Fumagalli between exclusive dealing contracts and loyalty rebates, he noted that in practice contracts may have features of both.

In concluding, he echoed the point made by Prof Fumagalli in that the price-cost test is an important piece of evidence that needs to complement a coherent theory of harm. However, the French Court of Appeal does not impose an obligation to apply it in particular when other evidence supports a finding of exclusion and in some cases it may be deemed irrelevant, e.g. if exclusivity contracts are applied.

The More Economic Approach and Internal Coherence in the Case Law: Are Rebates Special?

Prof Pinar Akman, Director, Centre for Business Law and Practice and Professor of Law at the University of Leeds, discussed the case law on rebates relating to coherence of the law under Article 102 TFEU, the Intel judgement, and the step towards a more effects-based approach in abuse of dominance cases.

As regards the European Court of Justice’s judgment in Intel, Prof Akman pointed out that it is difficult to place the case within the rest of the case law, partly because the court said so little about the legal assessment on rebates. Prof Akman argued that the court gave a procedural answer to the substantive questions of the case and questioned how the legal test of whether a conduct is abusive will depend on what the dominant undertaking has argued during the administrative procedure. According to Prof Akman, it is at least important to note what the court did not do, as what it did do. For example, the court did not overrule Hoffmann-La Roche, nor did it decide on whether rebates are per se abusive, if evidence of effects are required to establish abuse or if the AEC test is necessary. She noted that the court did not refer to the customers’ freedom to choose. In her concluding remarks, Prof Akman questioned what conclusions can be drawn from the Intel-judgment as regards the assessment of rebates, do they have to be analysed in the framework for exclusive dealing or predatory pricing? In view of the Courts’ role in developing competition law through their case law, she concluded with her reflections on the challenges Courts face in providing clear precedents regarding the application of Article 102 TFEU.

Alexis Walckiers, Chief Economist of the Belgian Competition Authority, began by stating that he very much agreed with Prof Akman on the fundamentals. He went on to state that he wished to present a more positive view of the outcome of the Intel judgement, by focusing on some useful messages from the judgment. According to Mr Walckiers, the Intel case reminds us of the importance of competition on the merits and that competition on the merits can lead to exclusion, for example through innovation. He went on to examine what could be inferred from the judgment regarding the role of a less efficient competitor as well as the importance of the counterfactual. In his view, the Intel judgment clarifies Hoffman-La Roche in that there is a rebuttable presumption of harm. According to Mr Walckiers, the fact that the Court in the Intel judgment does not say too much in not necessarily bad provided it sets out clear guiding principles for an assessment. He went on to argue that the AEC test is helpful in identifying anticompetitive behaviour and concluded by elaborating on the benefits of conducting the AEC test.

Conditional pricing and the AEC test – a happy marriage or an awkward couple?

Lucas Peeperkorn, European Commission, began by setting out some key elements of the judgements in Post Danmark I and Intel and stressed that the goal of European competition law is to protect consumers. He stated that there exists a rebuttable presumption of illegality regarding exclusivity rebates, and that it is generally not possible to foreclose efficient rivals whenever the price is above cost. He then went on to present some unresolved issues in relation to these key elements, in particular whether above-cost pricing can also be abusive and, as a corollary, the usefulness of the AEC test in the assessment of rebates.

The AEC test, Mr Peeperkorn argued, can be neither a necessary nor a sufficient condition for the finding of an abusive pricing practice as its relevance relates directly to the theory of harm. If the theory of harm is exploitative, as is the case with price discrimination, or concerns a conduct that is the result of collusive conduct, for instance when retroactive rebates are used to share the market, prices may very well be above cost and applying the AEC test does not make sense. On the contrary, if the theory of harm is about exclusion of competitors, in principle applying the AEC test does make sense. However, he cautioned that when conducting the AEC test it is necessary to consider the reliability of the data. Furthermore, he pointed out than when carried out, the AEC test should only be used as a part of a more elaborate effects-based analysis. In concluding, Mr Peeperkorn addressed the practical problems that arise when identifying the existence of a non-contestable part of demand and went on to state that the dominant firm should have strong incentives to identify the contestable part in order to enact an efficient rebate scheme.

The discussant Prof Doris Hildebrand, Founder and Managing Partner of European Economic & Marketing Consultants, brought additional perspectives to the topic by addressing the significance of the treaty change made to Article 3 TFEU with the Treaty of Lisbon. In particular, she stressed that the addendum that the goal of the EU is to strive for a highly competitive “social” market economy relates to the concepts of fairness and equality. According to Prof Hildebrand, these two concepts form the basis for an initial evaluation of whether a dominant firm is acting in a way that is consistent with how a non-dominant firm would behave by posing the question of whether a non-dominant firm would behave in the same manner. If this is not the case then there is reason to suspect that the motive of the dominant firm is either to exclude or to exploit. She went on to stress that each case must be analysed on a case-by-case basis by first defining markets and thereafter, assessing dominance and whether or not the conduct in question is likely to give rise to negative effects given the legal and economic context. She concluded by stressing that it is helpful to have the principles of equality and fairness in mind when evaluating whether a conduct is abusive or not.

A Transactions Cost Analysis of Rebates and Non-Linear Pricing

Due to technical issues Prof Joshua D. Wright, George Mason University Antonin Scalia Law School, was unable to participate was therefore not able to hold a presentation on the paper he had prepared together with Prof Bruce Kobayashi on transactions cost analysis of rebates and non-linear pricing.

The discussant Johan Sahl, Deputy Head of the Market abuse unit at the Swedish Competition Authority (SCA), presented the Swedish competition authority’s perspective on rebates. Mr Sahl made reference to the SCA’s prioritisation policy stating that as regards abuse of dominance cases the Authority prioritises investigations into suspected exclusionary conduct. Mr Sahl made a distinction between the AEC principle and the AEC test in the context of loyalty rebates and went on to state that the AEC principle provides guidance on all price-based abuse. He then elaborated on the role of the AEC test in the assessment of loyalty rebates. Mr Sahl then presented the cases on rebates that the SCA has dealt with after the Commission’s Guidance Paper on Article 102 TFEU came into force as well as the cases the SCA dealt with after the Intel judgment. Mr Sahl presented the benefits and challenges of applying the AEC test as seen through the SCA’s experience and concluded by stating that the AEC test is a useful tool in combination with other investigative measures.

Concluding remarks

In his concluding remarks, Prof Monti recalled the wording of Article 102 TFEU noting that no reference is made to a specific welfare standard, other than a conduct’s incompatibility with the internal market. This must be taken to mean that the EU-courts have a lot of flexibility when interpreting whether a conduct is anticompetitive. According to Prof Monti, in some judgments it can be tentatively ambiguous as to which welfare standard has been adopted, consumer welfare or the protection of the competitive process. Taking this into account and in light of the Intel, Prof Monti concluded that this gives rise to the question of the goal of Article 102 TFEU, which all the more seems to be to protect consumer welfare.

Prof Monti went on to touch upon the question raised about whether Article 102 has an object category and whether it is shrinking in light of recent case law. Prof Monti expressed that he was not sure if there has ever been an object category under article 102 TFEU. In light of the Intel judgment, Prof Monti went on to reflect upon what the legal significance will be of an argument that the contested rebate did not have an impact on the market in consideration of events and market developments occurring after the period during which the conduct took place. Prof Monti then went on to raise the issue of what competition authorities should do and underlined that agencies should more carefully try to match cases with the theory of harm in order to take more robust decisions. In this context, he noted that there was a consensus that some form of the AEC test as a preliminary filter is an appropriate mechanism for prioritising cases.

Prof Monti also raised some institutional concerns regarding the decentralization of EU competition law enforcement. He pointed out that the EU-court renders judgments not only for the Commission, but also for national courts and national competition authorities. It therefore has a responsibility to bring forward reasonable and meaningful standards that can be applied in a fairly robust and fair way across a union with diverse capacities. In connection to this he raised the question of whether the growing decentralisation should affect the way in which we construct the legal test.