On 30 January 2020 the ECJ issued its preliminary ruling in Case C-307/18 which is likely to have both general and sector-specific implications for antitrust practice. The case concerns certain pay-for-delay agreements between GSK as the manufacturer of an originator drug called paroxetine and its (potential) generic competitors on the UK market IVAX, GUK and Alpharma. A pay-for-delay agreement is a patent dispute settlement by which the generic manufacturer acknowledges the patent of the originator pharmaceutical company and undertakes not to sell its generic product for a specific period of time, normally against payment. In the recent years such arrangements have been subject to an increased scrutiny by competition authorities due to their capability to restrict market entry and limit consumers’ choice.
In Lundbeck (T-472/13) the General Court ruled that arrangements intended to delay the market entry of potential competitors against payment may be regarded as restrictions of competition by object. This would be particularly the case where, among others, such ‘reverse’ payments correspond approximately to the expected profit from potential market entry and there is a clear link between the payment and the competitors’ commitment to stay out of the market.
In the more recent case of Servier (T-691/14) the GC did not seem to attribute much importance to the actual ratio between the reverse payment and the expected profit. Rather, it put the emphasis on whether such reverse payment covered (legal and other) costs that were inherent to a patent settlement or went beyond that and thus served as an incentive for non-competing. Research, development and production costs are not inherent to the patent dispute according to the GC. Their compensation may incentivize the generic company to refrain from entering the market unless such costs are of insignificant amount.
Against this background, the ECJ’s view on several points of law was highly anticipated.
In Paroxetine the ECJ confirmed its long-standing practice that Article 101 TFEU applies to practices impairing both actual and potential competition. In the case of pay-for-delay agreements, it must be determined based on the legal and economic context whether, absent such agreement, there would have existed real and concrete possibilities for the generic company to enter the market. The ECJ outlined some of the relevant factors in that assessment:
- the preparatory actions taken by the potential competitor prior to entering into the agreement and its readiness to challenge the patent and to take the risk, upon entering the market, of being subject to infringement proceedings brought by the patent holder;
- the subjective perception of the established operator – the very existence of a value transfer between undertakings in consideration for not competing is an indication that the party established on the market perceives the other as a potential competitor, whereas the larger the value transferred, the stronger the indication;
- the absence of unsurmountable barriers to the entry of the new operator – the Court found that the existence of a patent is not, as such, an unsurmountable barrier since it is not an obstacle neither for the issuance of marketing authorization for a generic drug, nor for challenging that patent’s validity.
According to the Court, for a potential competition to be established there is no need to demonstrate with certainty that the potential competitor will enter the market concerned and that it will be capable of retaining its place. Even a prospective market entry may exercise competitive pressure over an established operator.
The concept of ‘by object’ restriction
In Paroxetine the ECJ recalls that the concept of restriction of competition ‘by object’ must be interpreted strictly and can be applied only to some concerted practices which reveal a sufficient degree of harm to competition, thus there is no need to assess their effects.
An interesting point of law discussed by the Court is the role of pro-competitive effects in the ‘by object’ analysis. The Court confirmed that where undertakings rely on the pro-competitive effects of their agreement, those effects must be duly considered for the qualification of the agreement as a restriction ‘by object’. Such pro-competitive effects according to the Court, however, must be demonstrated, relevant, specifically related to the agreement at issue and “sufficiently significant, so that they justify a reasonable doubt as to [the agreement’s anticompetitive object]” (para. 107). The first three conditions mean that any pro-competitive effects evoked by the parties must not be purely hypothetical or theoretical, whereas the last condition (“sufficiently significant to cast a reasonable doubt”) seems to be the threshold of “pro-competitiveness” that such effects must meet to prevent a finding of a ‘by object’ restriction.
In Paroxetine the pro-competitive effects of the pay-for-delay agreement were minimal, i.e. there was a slight reduction in prices. The effects were also “probably uncertain” since, absent the settlement agreement, the actual reduction of the prices would have been much greater if the generic company was successful in the patent dispute. However, it is not clear if the ECJ would have dismissed the pro-competitive effects of the agreement at issue as irrelevant for the ‘by object’ analysis, if these effects were significant, but still uncertain due to the underlying patent dispute.
Pay-for-delay agreements – a restriction ‘by object’?
It is a basic premise of competition law that each economic operator must determine independently the policy which it intends to adopt on the market and that agreements whereby competitors deliberately substitute practical cooperation between them for the risks of competition can be characterised as restrictions ‘by object’.
According to the ECJ, an arrangement under which a generic company undertakes to abandon its patent invalidation claims and envisaged market entry against certain transfer of value is not, as such, sufficient to classify as a ‘restriction by object’. An anticompetitive object would exist where the net gain from the transfer of value is that significant that it cannot be justified by the existence of any quid pro quo, but rather by the incentive that it creates for the potential competitor to refrain from entering the market. The assessment should take into account all pecuniary and non-pecuniary transfers of value, including indirect transfers such as awarding to the potential competitor a distribution contract under specific terms etc.
Further, the finding of an anticompetitive object according to the Court may not be rebutted by the fact that the agreement is limited to the scope and the remaining period of validity of the patent.
Restriction ‘by effect’
If no anticompetitive object is established, the competition authority should assess whether the agreement has appreciable potential or real adverse effects on competition. It is a settled practice that in such case the analysis of the counter-factual, i.e. the competitive environment absent the agreement at issue, is of crucial importance. In its decision the ECJ ruled that the counter-factual analysis in the case of pay-for-delay agreements does not necessarily require any definitive finding on the outcome of the patent proceedings, although the latter is a factor “among many to be taken into consideration in order to determine how the market will probably operate and be structured if the agreement concerned is not concluded” (para. 120).
As the outcome of patent proceedings is uncertain and not within the competition authorities’ competence, the Court’s finding seems reasonable and practical from procedural standpoint. It may appear, however, that pro-competitive effects relied by the parties would often risk being dismissed as uncertain on the grounds that they may have been even greater, absent the agreement at issue and provided that a particular party wins the case. This raises the question whether the merits of the underlying dispute should not be considered an important part of the counter-factual analysis in cases where the pro-competitive effects of the agreement are not “minimal” as in Paroxetine.
Another topic discussed by the ECJ is the definition of the relevant market in cases concerning pay-for-delay arrangements. In a situation of a disputed process patent that impedes the market entry of generic drugs, such generic versions must be also taken into consideration, if the manufacturers concerned are in a position to enter within a short period on the market with sufficient strength to constitute a serious counterbalance to the established operator.
The Paroxetine case is surely one with potentially large implications for antitrust practice. It sheds light over the widely discussed dichotomy between ‘by object’ and ‘by effect’ restrictions of competition. It would be particularly interesting to see if the ECJ will confirm and further build on its view in the pending Budapest Bank case (C-228/18). In the same time, the ECJ’s ruling provides some practical guidance for pharmaceutical companies negotiating patent settlement agreements.
* The content of this article is not a legal advice and should not be relied upon as such.