Under what circumstances may an incumbent associate with a competitor in a public tender? The ICA provides useful guidance in the field of plasma derivatives.

In January 2018, the Italian Competition Authority (AGCM, or “ICA”) launched an investigation for a possible breach of Article 101 TFEU against two international manufacturers of blood-based pharmaceuticals – Kedrion S.p.A (“Kedrion”) and Grifols Italia S.p.A (“Grifols”) – for participating as a joint venture in a public tender (issued by a contracting authority acting on behalf of a group Italian regions, “Intercent-ER”) to collect blood and plasma, to produce plasma derivatives and to deliver these to the Italian National Health System (NHS). However, in December 2018 the ICA closed the proceedings without a finding of infringement on the grounds that the evidence and information collected during the investigation were insufficient to substantiate its initial concerns.

It is indeed interesting to illustrate the factual findings as well as the initial and final reasoning of the ICA which led to this decision, as the matter of joint participation of competitors in public tenders always raises uncertainties on compatibility with competition law. The peculiarities of the pharmaceutical industry and of markets related to public health increases the complexity of the framework and makes this case particularly explicative of the do’s and don’ts for pharmaceutical companies in the context of public tenders.

The investigation was prompted by competitors’ complaints – namely, Baxter Manufacturing S.p.A and Baxalta Italy S.r.l. (jointly referred to as “Baxter”, part of the US group Shire) and CSL Behring S.p.A (or “CLSB”, part of the Australian group CSL) – raising concerns that the temporary association between Kedrion and Grifols to participate in the Intercent-ER tender was overabundant and anti-competitive, also pointing to a wider commercial collaboration between the two firms at an international level. In the decision to open a formal probe, the ICA initially supported this possible conclusion based on the preliminary findings described below.

The regulatory and economic context

In Italy, plasma derivatives are always included in the “A” and “H” classes of drugs, whose cost for patients is covered by the NHS under the national regulatory framework for pharmaceuticals. As in most European countries, the collection of human blood and plasma is mandatorily gratuitous as the supply of blood/plasma from donors on a voluntary basis is associated to greater safety for health.

Further, according to the applicable sectoral legislation (Law No. 219 of 21 October 2005),  the blood collected on the Italian territory is property of the State (namely, of the “Regione” where the collection occurred) and, once collected, is then entrusted to highly specialized companies for the production of plasma products and delivery to the NHS for a consideration. The necessity for public tenders to assign the service begins here. The annual value of these services in Italy amounts to approximately EUR 150 million.

Law No. 219/2005 removed a previous statutory requisite that all processing and production plants of plasma derivatives had to be localized in Italy by substituting it with a requisite that such plants must be localized in any EU Member State where the collection and supply of blood/plasma is not organized on a lucrative basis. However, the ministerial decree enabling the new requisite to become effective was issued in 2014 (Ministerial Decree of 5 April 2014) and this delay caused Kedrion to become a monopoly in the national market for the collection, processing and production of plasma derivatives. However, the 2014 Ministerial Decree introduced procedures to release new authorizations based on the less restrictive requisite, which allowed new undertakings to enter the market (namely, Baxter, CLSB and Octopharm Italy).

Following these authorization proceedings, four groups of Regions organized four tender procedures to collect blood and plasma from authorized laboratories, to produce plasma derivatives and to deliver these to the National Health System (NHS). Two of these tenders were awarded in 2016 and 2017 – for a duration of five years (plus possible extensions) – to, respectively, CSLB (for EUR 200.250 million) and to a temporary joint venture between Kedrion and Grifols (for EUR 224 million). The ICA’s probe relates to this second tender (the “RIPP tender”).

The first tender (hereinafter, the “NAIP tender”) was annulled by an Italian first-instance court following legal actions filed by Kedrion and Grifols while the RIPP tender was challenged by Baxter and CSLB, but the court’s decision was still pending when the ICA opened the probe.

A third tender (for a similar duration and for EUR 278.250 million, the “PLANET tender”) was launched in November 2017, but was still to be awarded at that time. A fourth tender was planned to be issued within 2018.

The RIPP tender and the initial concerns of the ICA with respect to the Kedrion/Grifols JV

The ICA first considered that Kedrion represented the national incumbent in the sector of blood pharma-derivatives, enjoying a monopoly position for more than thirty years in the specific services for the collection of plasma from blood donors and delivery of plasma products to the NHS. Grifols, in turn, is the national subsidiary of one of the main global operators in the segment of plasma products.

Further, the ICA pointed to commercial and industrial relations between Kedrion and Grifols in the US (e.g., the reciprocal transfer of production and processing plants from one to the other) and to an allegedly unjustified waiver from Kedrion to legally challenge an authorization obtained by Grifols to operate in Italy, despite Kedrion persisting in pursuing the same action against the other competitors.

The Italian Competition Authority conducted a preliminary examination of the tender’s rules and requisites with regard to the capabilities of the various participants. An indispensable requirement for participation under the RIPP tender’s rules was that the undertaking be at least capable of supplying a certain minimum quantity of certain plasma products, which reflected specific statutory provisions and represented most of the demand for such products. Moreover, the tender’s rules required participants to indicate their capability to provide additional ancillary plasma products, which however did not constitute an indispensable condition to participate in the tender but rather granted additional points on the technical evaluation of the offer.

All the five authorized providers of the tendered services autonomously satisfied the indispensable requirements to participate to the tender. In addition, all five competitors were able to provide one or two of the non-indispensable products in the required quantity. However, Kedrion was capable of supplying all but one of these ancillary products and Grifols was the only participant able to provide the other one. Octopharm was able to supply one ancillary product that only Kedrion was also able to supply but did not submit any offer in the RIPP tender in the end (nor to the previous NAIP tender). Hence, the Kedrion/Grifols JV was the sole tender’s bidder capable of supplying two of the ancillary products on an exclusive basis (one available to Grifols and the other one to Kedrion).

This led the ICA to the concern that while Kedrion and Grifols could have participated autonomously in the tender and (particularly Kedrion) would nonetheless have enjoyed a technical competitive advantage over the competitors thanks to the availability of one or more valuable ancillary products, by participating through the JV they were able to further increase this competitive advantage and thus maintain a higher bidding price. Indeed, the Kedrion/Grifols JV was awarded the tender despite the fact that its economic offer was much greater than that of the second-best competitor (EUR 118/kg v. EUR 90/kg).

This said, the ICA considered that the rationale behind allowing undertakings to temporarily associate and join forces to participate in a public tender rests in favoring more competition by enabling undertakings, which otherwise would have not participated, to compete effectively. However, in light of the economic/industrial and legal/administrative barriers to enter the specific sector in question as well as the broader collaboration arrangement between Kedrion and Grifols, ICA considered that it was possible that Kedrion and Grifols constituted the JV as a means to eliminate direct competition between them, particularly on the economic part of the offer (which Kedrion and Grifols would have had to lower absent the JV). In this way, according to the ICA’s initial accusation, they were able to gain a competitive advantage that was incompatible with the pro-competitive rationale of temporary joint ventures.

Accordingly, ICA also emphasized that even if such temporary associations are expressly allowed by law and, according to case-law, a per se prohibition of overabundant JVs to participate in public tenders does not exist, the instrumental exploitation of statutory rights for anti-competitive aims may nonetheless constitute an infringement of competition law, and namely of Article 101 TFEU (and the national equivalent) on the prohibition of anti-competitive agreements. The relevant market was identified in the single RIPP tender based on a consolidated national and EU case-law, though ICA did not rule out the possibility that the anti-competitive cooperation could interest a broader market for the type of services concerned.

The final ICA decision to close the probe for insufficient evidence

After roughly a year from the opening of the probe, following a thorough factual investigation and economic analysis on the tenders’ rules and information collected and having heard the arguments of the parties, the ICA overturned its preliminary findings.

The ICA first pointed out that the NAIP tender, which was awarded to CSLB but initially annulled by a national court following challenges by Kedrion and Grifols, was then finally validated by an appeal judgement of the last-instance court. Further, the Planet tender, which was pending at the time of the opening of the probe, was then awarded to Baxter in July 2018 (though provisionally, as it was also subject to a challenge pending before national courts).

Having said that, the ICA made an accurate comparative analysis of the three tenders’ rules and of their consequences for competition. In this regard, it emphasized the significant differences in the tenders’ strategies and conditions, particularly with respect to the weight (in terms of points) assigned to the economic component of the offers compared to the technical one. Indeed, while the NAIP tender (which was organized under the older EU legislative framework for public procurement) gave significantly more weight to the economic part of the offer (90 points out of 100), the subsequent RIPP and PLANET tenders (organized under the new EU legislative framework) gave more weight to the technical component of the offer rather than to the economic one (in a ratio of 60/40 and 70/30, respectively). Moreover, with respect to the points assigned to the technical offer, the ICA emphasized that the RIPP tender gave more premium points (5) to the ancillary products available to each participant rather than to the PLANET tender for the same products (from 2 to 3).

The ICA also considered that in the RIPP tender certain parameters to calculate the score linked to the bidding price further demoted the relevance of price cuts compared to the PLANET tender: in the latter’s design, discounts from the base price awarded more points to the bidder than in the RIPP tender’s design and therefore the economic offer was overall less relevant in the RIPP tender than in the PLANET tender (despite the latter assigning 70 points to the technical offer v. 60 points assigned by the RIPP tender).

Notably, Intercent-ER, which was heard on this issue, confirmed that it deliberately designed the RIPP tender’s rules with a view to strongly prizing the technical offer (particularly with respect to the range of available ancillary products) rather than the economic offer.

Against this background, the ICA finally concluded that there was insufficient evidence to find a breach of Article 101 TFEU (or the national equivalent) in connection with the constitution of the Kedrion/Grifols JV in the context of the RIPP Tender, based on the following key considerations:

  • Overabundant temporary associations of competitors to participate in a public tender are not prohibited per se and therefore a case-by-case analysis of all the elements of the economic and legal context is required;
  • The analysis of the competitive context and of the RIPP tender’s rules and conditions revealed that at least one of the two competitors in the JV (namely, Grifols) was certainly precluded from competing effectively on its own; and induced the other one (Kedrion) to legitimately strengthen its offer by associating with Grifols;
  • Indeed, the economic and technical analysis of a hypothetical autonomous offer by Grifols showed that, being able to only provide one ancillary product, it would have not been able to compensate the low technical value of its offer with an aggressive economic offer unless it lowered the price to an “unsustainable” level (namely, a maximum of EUR 66.63/Kg, which was lower than any other competing offer in the tender; it would have impeded Grifols gaining a minimum margin; and, further, it would have constituted an “abnormal” reduction exceeding 50% of the auction base, i.e. EUR 140/Kg);
  • As to a hypothetical autonomous offer by Kedrion, the analysis showed that by participating alone Kedrion would have been able to submit the best offer in terms of range and quality of available products, but not as complete as in association with Grifols (which granted an additional valuable ancillary product) and not at competitive economic terms compared to the competitors’ offers (particularly relevant to this conclusion was the fact that Kedrion’s internal documents revealed that it deemed uneconomic to price below EUR 118.1/Kg, but by associating with Grifols it was then able to bid at EUR 117.8/Kg);
  • Hence, the evidence gathered during the investigation showed that the constitution of the JV reflected an objective need of the two competitors to formulate an offer that was effectively able to compete with those of other existing competitors, one of which (CSL) had just been awarded the NAIP tender and the other one (Baxter) would have been awarded the subsequent Planet tender (and thus no risk of eliminating competition was detected on the market as effect of the JV).

Conclusive Remarks

This decision is particularly relevant as it illustrates and provides valuable guidance to operators on how even an incumbent undertaking can lawfully associate with a competitor to participate in a public tender. This is also possible in the case where the incumbent is hypothetically able to participate alone if, by virtue of the association with the competitor, it will gain the ability to improve its technical offer and the price/quality ratio without eliminating competition in the overall market for the products/services concerned.

Seguici su