We reported on this blog about the General Court (GC)’s judgment in Servier on the day on which that judgment was handed down, noting in particular the rejection of the Commission’s novel approach to market definition.
In this post, we focus on the analysis applied in one of the parallel judgments issued to the generic companies which were party to the infringements of Article 101. While the finding of a ‘by object’ infringement was maintained in relation to the settlement agreements between Servier and other generic companies, the Krka judgment involved a full annulment of the relevant part of the Commission decision.
What were the Commission’s findings?
The case concerned a patent settlement and licence agreement between Servier and Krka in relation to Perindopril, a cardiovascular medicine used (primarily) to treat hypertension). The settlement agreement put an end to UK patent litigation between Servier and Krka in October 2006, following the grant to Krka of a marketing authorisation for Perindopril earlier in the year. At the time there was no generic Perindopril in Western European markets, but Krka had launched its generic Perindopril “at risk” in certain central and eastern European (CEE) markets. On the same day as the settlement agreement, Servier and Krka entered into a royalty-bearing licence agreement in relation to the CEE markets where Krka had already entered. This licence accorded Krka ‘sole’ licensing rights (i.e. while Servier itself also remained on the market, it committed not to grant licences to others).
In its decision, the Commission treated the combination of these agreements as a form of market sharing arrangement. According to this view, Krka obtained the licence in return for refraining from launching its generic product in a number of other European markets, notably the UK. The Commission considered that: (i) Servier and Krka were at least potential competitors at the time of entering into the agreements; and (ii) the sole licence which permitted Krka access to the CEE markets constituted an inducement to Krka to enter into the settlement agreement, under which it agreed to refrain from competing in the 18 restricted markets.
Findings of the GC
The GC rejected the Commission’s analysis that the mere conclusion on normal market conditions of a licence agreement could amount to an inducement, even if linked (whether legally and/or temporally) to a settlement agreement containing restrictive clauses. In the view of the GC, this would lead to a paradoxical outcome: the wider the scope of a licence agreement, the greater the inducement, and thus the easier it would be to find a restriction by object.
However, it accepted that a ‘side deal’ may amount to an inducement which renders any restrictive provisions in the linked settlement agreement unlawful where the side deal “serves as a vehicle for the transfer of value from the originator to the generic company”. In that case, the restrictions in the settlement are considered not to reflect the parties’ recognition of the validity of the patent but rather to be the result of a (significant) inducement. However, the GC went on to note that the Commission has to discharge its burden of proof that this is indeed so.
In this case, the settlement itself did restrict Krka’s market access in the UK and other Western European markets while relevant patents remained in force; it was also clear that the licence agreement counted as a ‘side deal’ as it was temporally linked (the two agreements had been concluded on the same day; by contrast, a later assignment of patent by Krka to Servier was held not to be ‘indissociable’ from the settlement agreement). A key question for the GC was therefore whether any value had in fact been transferred to Krka under the licence agreement. The licence granted by Servier was royalty-bearing at the rate of 3%. The GC emphasised that it was for the Commission to demonstrate that this rate was abnormally low, and thus not a commercial rate if it wished to prove that Krka had been induced to give up its other markets. The GC found that the Commission had not demonstrated that this was the case – in fact, the Decision acknowledged that the rate was fair, albeit relatively low.
The Decision was therefore annulled, insofar as it concerned restriction of competition by object.
The GC also reviewed the Commission’s conclusions that the agreement had an ant-competitive effect. It again disagreed with the Commission’s conclusions, finding that absent the agreements, Krka would not have entered the French, Dutch and UK markets, based on Krka’s assessment of the strength of the ‘947 patent. The GC noted that, at the time of entering into the agreements, Krka was the subject of an interim injunction in the UK and that it was “very unlikely” that without a licence agreement Krka would enter other markets “at risk”.
As with the cases concerning the other agreements between Servier and generic companies, the Court has sent an important message on the role of the concept of restrictions of competition ‘by effect’ under EU competition law.
The Court emphasised that where an agreement has been put into effect, it is not sufficient for the regulator to rely on the existence of ‘potential’ effects on competition. To do so effectively obliterates the distinction between object and effect restrictions. The GC conducted a careful analysis of past cases in which the consideration of effect restrictions has been limited to a consideration of potential effects, and distinguished this case from those earlier examples, which tended to concern preliminary references, or decisions in which no fine had been imposed.
Whatever the ultimate outcome of any further appeals to the CJEU on the treatment of patent settlement agreements as restrictions of competition by object in the other Servier cases, in Lundbeck (where hearings have taken place within the past week) or in Paroxetine, the approach of the GC to side deals in the Krka case is to be welcomed. The GC has dialled back the risk of concluding a licence agreement where there is a link to a settlement of litigation, provided the licence itself can be demonstrated to be on commercial conditions.
This sensible approach is likely to enhance the possibilities for companies in litigation to resolve those differences in a way which provides benefits to both parties, rather than requiring a total capitulation by one side or the other.