Two separate insurance companies Allianz and Generali concluded a number of contracts with autorepair shops on the prices and other conditions that would apply for cars insured by these companies repaired by those shops. Moreover, Allianz and Generali also concluded similar contracts with the dealers who operated car repair shops. Finally, they concluded similar agreements with the association of car dealers. In this category of agreement, the prices for car repairs would increase with the number of insurances sold by the dealers. Allianz and Generali, therefore sought to link the number of insurances sold by the dealers to the remuneration for car repairs. This is obviously designed to increase or at least consolidate the market share of Allianz and Generali on the market for car insurances. Apart from this business strategy, the idea behind these agreements was that auto repair shops could start repairing immediately on the basis of the predetermined tariffs, something that is clearly a practical solution to get people back on the road as quickly as possible.
The Hungarian Competition Authority (referred to in the judgment as GVH; HCA hereafter), however, considered these agreements to restrict competition by object on the market for car insurance contracts and the market for car repair services within the meaning of the Hungarian equivalent of Article 101 TFEU. As it happens, this provision closely mimics Article 101 TFEU, in line with the trend towards a spontaneous harmonisation of competition law throughout the EU (see the preamble and explanatory memorandum to the Hungarian competition act referred to in paragraphs 3 and 5 of the judgement). Because the HCA considered that there was no effect on trade, Article 101 TFEU was not applied by the national competition authority. The prohibition decision issued by the HCA was challenged by Allianz, in particular on the grounds that the agreement did not restrict competition by object. the decision by the HCA was partially reversed and then restored upon appeal, against which Allianz appealed to the Hungarian Supreme Court. The Supreme Court then made a preliminary reference essentially asking whether the agreements at hand fall within the object category.
The distinction in the text of Article 101 TFEU between the object and effect (the object-effect dichotomy) of an agreement is as problematic as it is important to that provision. By and large, this dichotomy has allowed the Court and the Commission to confine their analysis of the object-category of agreements, often referred to as hard-core agreements, to a legalistic analysis that ignores many of the complexities of the markets involved. It can therefore be seen as a tool to reduce the enforcement costs of article 101 TFEU as it essentially rules out a definition of the market and an analysis of the position of the parties involved on that market. As a result, it provides the parties involved with a negatively defined legal certainty in that they know what the effect will be of having, say, a price fixing clause in an agreement. In order to avoid the per se applicability of Article 101 TFEU, parties will have to say away from such hard core restrictions. A very prominent application of this facts analysis can be seen in the Commission’s De Minimis Notice. Here, the Commission defines market shares below which agreements are considered not to have appreciable effects on competition. At the same time, in keeping with the object-effect dichotomy, the De Minimis Notice and its market share tests do not apply to restrictions by object.
Before answering the substantive competition law question in the Allianz case, the Court first addresses its jurisdiction as well as the admissibility of the preliminary reference. As to the jurisdiction, all parties agree that the preliminary questions concerning Hungarian competition law are admissible. Even though article 101 TFEU does not apply to the facts of the case, the fact that Hungarian competition law was aligned with EU competition law (see para. 21) brings the question within the Court’s jurisdiction. The interest in ensuring uniform interpretation of EU law is well recognised to result in this jurisdiction.
The Court’s reasoning leading to the admissibility of the request for a preliminary ruling is more interesting, as it relates to the amount of information the Court needs to answer the preliminary questions. Both the Hungarian government and the EFTA Surveillance Authority noted that the preliminary reference does not explain the economic and legal context of the agreements prohibited by the HCA decision (paras. 24, 25). In other words, they believe that more information should have been provided to allow the Court to rule on the classification of the agreements as restrictions by object. The minimal set of facts presented in the judgment would suffice for a traditional appraisal of the applicability of the object criterion. As we will see, this minimal set of facts leads to some reasoning on the part of the Court that is at best awkward and speculative. The Court addresses these issues by referring to its well established case law on the division of roles between national courts and the European Court of Justice (para. 29). This division essentially means that it is for the European Court to interpret EU law, whereas it is for the national Court to apply that interpretation to the facts of the case. As a result, the Court considers the preliminary reference admissible and rephrases it as follows:
By its question, the referring Court asks, in essence, whether Article 101(1) TFEU must be interpreted as meaning that agreements whereby car insurance companies come to bilateral arrangements, either with car dealers acting as car repair shops, or with an association representing the latter, concerning the hourly charge to be paid by the insurance company for repairs to vehicles insured by it, stipulating that that charge depends, inter alia, on the number and percentage of insurance contracts that the dealer has sold as intermediary for that company, can be considered a restriction of competition ‘by object’ within the meaning of that provision.
We see that the Court rephrases the question to concern only the interpretation of EU law, and not application to the facts of the case (as the Hungarian Supreme Court phrased it). This is in line with well-established case law of the European Court of Justice, but does highlight the difficulty in the watershed of the Court envisages between the interpretation of the law and its application to the facts, especially in relation to the effect-category, since this requires a factual analysis. As we will see this is particularly problematic in relation to the object-effect dichotomy that is central to article 101(1) TFEU.
The object-effect dichotomy
The Court goes on to answer the question by noting the conjunction ‘or’ between the object and effect in article 101 TFEU, and concludes that this results in the need to first consider the object of the agreement in the economic context in which it is to be applied (para. 33). It then holds as follows (paras. 34, 36):
Accordingly, where the anti-competitive object of the agreement is established it is not necessary to examine its effects on competition. Where, however, the analysis of the content of the agreement does not reveal a sufficient degree of harm to competition [bold and italics added], the effects of the agreement should then be considered and, for it to be caught by the prohibition, it is necessary to find that factors are present which show that competition has in fact been prevented, restricted or distorted to an appreciable extent […]
In order to determine whether an agreement involves a restriction of competition ‘by object’, regard must be had to the content of its provisions, its objectives and the economic and legal context of which it forms a part […]. When determining that context, it is also appropriate to take into consideration the nature of the goods or services affected, as well as the real conditions of the functioning and structure of the market or markets in question […]
These considerations are particularly interesting, notably the italicised part of paragraph 34. The Court starts out with the object effect dichotomy, and then states that an effects analysis is necessary where the analysis of the content of the agreement does not reveal a sufficient degree of harm to competition. In other words, determining whether an agreement falls within the object category requires an analysis that should reveal a sufficient degree of harm to competition. This immediately begs the question whether there is a possibility for an analysis that could result in a finding of an insufficient degree of harm to competition? The Court appears to allow for this when setting out the factors to be taken into account in this context analysis in paragraph 36. These factors include not only the objectives of the agreement, but also its economic and legal context. That context analysis in turn, requires the competition authority/ national judge applying article 101 TFEU to take into consideration the real conditions of the functioning and structure of the markets in question. To my mind, this analysis goes beyond the simple appreciably analysis required under Völk/Vervaecke. This case, as we may recall, involved the market sharing agreement (normally considered a hard core restriction or restriction by object) between two parties with a very minimal (<5%) market share only. As a result, the Court considered that agreement to fall outside the scope of article 101 TFEU because it had only in the significant effects on the market is concerned. I’m not really sure how far we can take paragraph 34 of Allianz. If this means that anyone applying article 101 TFEU would have to engage in a market analysis that goes beyond just ruling out insignificant market shares, the object-effect dichotomy is no longer relevant. More fundamentally, apart from the uncertainty concerning the interpretation of the object-category, we are left with considerable uncertainty as to the validity of the dichotomy itself. Tellingly, this uncertainty appears to also affect the Court itself.
We see this where the Court attempts, despite the division of tasks it had just reaffirmed, to apply its interpretation of article 101 TFEU to the facts of the case. Concerning, for example, the linking of car insurance to the remuneration for car repairs, the Court notes that these agreements aim to maintain or increase the market shares of the insurance companies (para. 44). Here the Court goes on what can only be described as a bizarre counterfactual analysis. The Court considers that if there was horizontal agreement or concerted practice between Allianz and Generali, this agreement would be treated as a restriction by object (para. 45). This is nothing but speculative and completely unhelpful, as neither the GVH’s decision nor the preliminary reference make any mention of such a horizontal agreement. The Court then holds that even if there is no agreement or concerted practice between Allianz and Generali, a context analysis would have to reveal whether the vertical agreements are ‘sufficiently injurious to competition as to amount to restriction of competition by object’ (para. 46). Again, this appears to introduce an element of effects-analysis in the object category. Concerning the Court’s observation in paragraph 44 that the agreements aim to maintain or increase the market share, the Court seems to have achieved Robert Bork’s Antitrust Paradox. The maintenance and preferably increase of market shares is the essence of entrepreneurship competition law should protect. In the absence of dominance – the judgment contains no references at all to such a situation – this aim cannot by itself lead to the applicability of the object category.
In my opinion, the Court’s application of this analysis to the facts of the case in paragraph 47 is a spectacular misconception of competition law. As was observed above, the insurance companies gave car dealers acting as insurance intermediaries an incentive to sell their insurances by increasing the tariffs for repairs. This is probably contrary to Hungarian consumer protection law that requires such intermediaries to be independent from the insurance companies. In essence the Court considers that the agreements could qualify as a restriction by object when the agreements would go against the Hungarian law that requires car dealers acting as intermediaries to be independent from the insurance companies. Is the Court seriously arguing that the infringement of consumer protection legislation qualifies an agreement as a restriction by object?
The competition law analysis, fortunately, becomes slightly more orthodox in paragraph 48 where the Court holds that the agreements will also amount to restriction by object whenever the agreements will eliminate or seriously weaken competition on the market. This analysis of foreclosure requires that the structure of the market the existence of alternative distribution channels and their respective importance as well as the market power of companies concerned are taken into account (the network or bundle effect). This, is truly nothing short of a fully-fledged effects analysis starting with a thorough market definition as well as an analysis of the parties position on that market. It looks like the analysis required in Delimitis, for example. It is therefore only slightly more orthodox in that it again conflates the effects analysis and the object category.
Paragraphs 49 and 50 of the judgment depart from a similar hypothetical situation, this time relating to the price harmonisation by means of the decision taken by the association of car repair shops. If there is such a horizontal price fixing decision by an association of undertakings, the agreement concluded between the association and the insurance companies also falls under the object category. This appears to mean that anyone concluding an agreement with the association (or members thereof – the Court is not clear on this) ‘that voluntarily confirmed’ that decision, is a party to a restriction by object. The word ‘voluntarily’ is probably and hopefully simply the result of a poor translation, as the Dutch en German version contain the word ‘bewust’ and ‘bewusst ‘, which is translated as ‘knowingly’. It is understandable that an agreement that knowingly confirms a restriction of competition suffers the same fate as the original restriction. The French version, however, refers to ‘volontairement’, which could point to both intention and the absence of compulsion being the standard.
What, then, is the upshot of Allianz? In my opinion any judgment that only results in more uncertainty as to the correct interpretation of any provision is by definition a bad judgment. In this case, mixing up the object and effect categories in trying to answer a preliminary reference that simply does not contain the necessary legal and economic facts has resulted in a judgment that only complicates what could have been a perfectly clear dichotomy: Firstly, market sharing, horizontal and vertical price fixing is prohibited per se, irrespective of the effects this may or may not have had. Secondly, agreements outside the object box only fall under Article 101(1) when they produce appreciable negative effects for competition. Whatever we may think of the exact classification of specific types of agreement as restrictions in the object or effect category (should vertical resale price maintenance qualify as a per se / hard core / object restriction?), there is merit to at least being clear about this classification and the underlying dichotomy. This judgment provides no clarity on either. It only results in impractical competition law.