The EU Court has upheld a Commission decision that was generally considered an important test of the Commission’s current methodology of analysing exclusionary conduct by dominant companies.

The judgment concerned the appeal brought before the General Court of the EU (the “Court”) by Tomra Systems ASA, a supplier of reverse vending machines that facilitate the collection of empty drink containers (“RVM”), against the decision of 26 March 2006 of the European Commission (the “Commission”). In that decision, the Commission found that Tomra had abused its dominant position in the RVM market in five countries by means of exclusivity agreements, individualised quantity commitments and individualised retroactive rebate systems. The Commission found that the conduct of Tomra led to the foreclosure, and in some instances the elimination, of Tomra’s competitors to the detriment of consumers. For this, it imposed on Tomra a fine of 24 million EUR, which consisted of a basic fine of 16 million EUR, increased by 10% for each full year of the infringement in the period 1998-2002. (the “Relevant Period”).

In its appeal, Tomra sought the annulment of the Commission decision or in the alternative the annulment or the substantial reduction of the fine. The Court’s judgment, rendered on 9 September 2010, dismissed the appeal in its entirety and upheld the Commission decision (Case T-155/06).

Tomra put forward several arguments contesting the relevance and the accuracy of the evidence used by the Commission, the finding by the Commission of an exclusionary strategy and of an unlawful conduct as such and the Commission’s assessment of the exclusionary effects of such conduct.

Rebates based on individual customer requirements

In relation to the assessment of the individualised quantity commitments, the individualised rebate schemes and their exclusionary effect, Tomra argued that it was not in a position to estimate the requirements of its customers accurately in order to determine the exclusionary effect of the commitments. In support of this, Tomra pointed out that customers did not inform it of their total RVM requirements in the contract period, while, as the Commission also acknowledged in its decision, demand for RVMs was irregular and non-recurring. Such a task was made even more difficult as none of the countries where the alleged infringement took place had introduced mandatory deposit systems during the Relevant Period. Finally, Tomra argued that quantity commitments and objectives only rarely coincided with customers’ actual purchases and that in fact, actual purchases were consistently higher than contracted quantities.

The Court dismissed these arguments and corroborated the Commission’s conclusion that Tomra had sufficient information to estimate accurately the quantity requirements of each of its customers. This information could come from the customer itself, past purchases, Tomra’s own market research and other transparent factors (e.g. the number of outlets of a customer where a RVM could be installed). As far as the inconsistency between the requirements and the contracted volumes were concerned, the Court observed that the evidence provided by Tomra demonstrated that actual purchasing volumes were in most cases slightly above the volumes provided in the quantity requirements. Such finding confirmed, in the Court’s view, the premise of the Commission decision.

Rebates with foreclosing effects

Tomra also argued that the Commission was wrong to find that the conduct of Tomra was capable of foreclosing competition. In this respect, Tomra maintained first that the Commission was wrong to find that exclusivity agreements, individualised quantity commitments and individualised retroactive rebates were automatically (per se) unlawful; second, that the Commission failed to consider whether the contestable part of the RVM market was sufficiently large to enable equally efficient competitors to remain on the market; and third, the Commission’s assessment of the foreclosing effects of the retroactive rebates was based on incorrect and misleading evidence and assumptions.

The Court, repeating the settled jurisprudence to date, noted that the concept of abuse is an objective concept that relates to the conduct of a dominant company, “which is such as to influence the structure of a market where, as a result of the very presence of that undertaking, the degree of competition is weakened” and which has the effect of hindering the degree of competition. It also recalled that a dominant company that ties customers even at their own request (e.g. by non-binding clauses, obligation or promise voluntarily undertaken by the customer) abuses its market position under Article 102 TFEU. In terms of rebate schemes, if these are based on volume they are less likely to be deemed unlawful provided there is an economically justified countervailing advantage (e.g. passing-on of the reduction of costs for producing larger quantities of the product). However, in doing so, all relevant circumstances need to be examined.

The Court held that in the case at hand, the Commission met all its obligations in the assessment of the rebates and even went beyond the requirements by examining the actual effects of Tomra’s practices. The Court dismissed other factors such as the technical superiority evoked by Tomra, as it held that it was not relevant for determining an abuse (but rather relevant for determining the market position of Tomra); or the fact that the customers were professional buyers who could compare Tomra’s RVMs with those of competitors (Tomra’s conduct was clearly designed to provide an incentive to its larger customers not to purchase from competitors).

Tomra also contested the finding of foreclosure by arguing that even if all the agreements in question had foreclosure effects, its competitors would still be free to seek other customers. According to Tomra, the question would be whether a competitor could profitably remain on the market by serving only the “contestable” part of the market. In its view, the Commission ought to have determined the size of minimum profitability required to operate on the relevant market. According to Tomra, had such an analysis been undertaken, it would have proven that equally efficient competitors could not have been foreclosed from the market.

The Court dismissed all these arguments and noted that it is right to consider that by foreclosing a significant part of the market a dominant company restricts the entry to at least one or some competitors, thus reducing the intensity of competition as a whole. Such foreclosure cannot be justified by proving that the remaining “contestable” share of the market would be sufficient to accommodate a limited number of competitors. The Court noted that it is not for the dominant undertaking to dictate how many competitors should be allowed to compete for the contestable portion. In this particular case the Court found that in any event Tomra’s conduct foreclosed 40% of the total demand in the relevant market.

Finally, Tomra argued that its rebates, even though retroactive (i.e. applied in the totality of the purchases of a customer over a defined period of time) were leading to positive prices and as such were not necessarily capable of being exclusionary. In other words, a competitor could still offer a competitive price to the same customer and make a profit. The Court dismissed this argument as irrelevant and noted that there are a series of other considerations to determine the anticompetitive character of retroactive rebates in the Commission’s decision, such as the fact that the incentive to obtain supplies exclusively from Tomra was particularly strong in Tomra’s offered rebates, the rebate schemes in question were individual to each customer and that Tomra offered retroactive rebates to its largest customers with the aim to induce loyalty.

The fine

Tomra requested the annulment or the reduction of the fine claiming that it was disproportionate to the seriousness of the infringement or discriminatory. In this respect it argued that the fine imposed by the Commission represented almost 8% of its worldwide turnover in 2005 which is the “highest ever percentage of worldwide turnover of a company fined for violating competition rules” and that it was disproportionate considering that the relevant countries for the infringement represented only a small portion of is annual turnover (less than 34% in the EEA as a whole). It noted that in comparison with the Microsoft case that concerned a “very serious” infringement, the Commission’s fine represented only 1.5% of its worldwide turnover or in the Astra Zeneca case the relevant fine concerned only 3% of its turnover.

The Court dismissed these arguments. It found that the Commission’s practice in earlier decisions cannot serve as a legal framework for the setting of the fines in competition matters and ruled that the Commission was within its powers to impose such a fine based on Tomra’s worldwide turnover considering the seriousness of the infringement and all the relevant factors.


This is the first Court ruling following the adoption by the Commission of the Guidance document on exclusionary conduct by dominant companies. If the Court had contested the Commission’s methodology in analysing Tomra’s exclusionary conduct, the usefulness of that Guidance would have been reduced.

Although the judgment of the Court in the Tomra case does not add new elements in the assessment of rebates, it does provide additional guidance and clarification on the elements that must be considered by a dominant company when it offers rebates on its customers. In this particular judgment the Court did not expressly refer to the Commission’s discretion with respect to assessments of economic nature. However, the judgment is characterised by the absence of any assessment of the relevant economic arguments, and this in a case that relied heavily on economic theories. The Court’s judgment reminds us that dominance cases, and essentially those related to rebates are rather complex cases involving legal and economic elements that the Court will rather avoid analysing in depth.

Tomra can appeal before the Court of Justice the judgment of the General Court strictly on points of law within two months and ten days from the date of the ruling.