Pharma beware: EU consults on tighter rules for joint innovation

The proposed new draft texts for the EU R&D block exemption (Draft R&D BER) and R&D section in EU Horizontal Guidelines (Draft Guidelines) threaten to make worse an already unsatisfactory situation as regards the EU competition rules applicable to R&D agreements.


The proposed changes limit the scope of a block exemption that is already little loved or used and reduce the comfort available to agreements that fall outside its scope. Undertakings in the life sciences sector and other high innovation areas are the most likely to be affected. They have until 26 April to make their views known to the European Commission (EC) in the context of the ongoing consultation on the drafts.

Once final, the new texts will replace the existing R&D block exemption (Existing R&D BER) and Horizontal Guidelines (Existing Guidelines), which expire on 1 January 2023. The EC consultation on the drafts runs to 26 April.

There seems to be no good justification for tightening the rules in this area. Most R&D agreements bring together parties with complementary skills and technologies. They are, as a result, strongly pro-competitive. A recent study conducted by the EC found no examples of formal enforcement action being taken during the ten year period covered (2011 to 2020) against a pure R&D agreement by any EEA authority.

If the new texts are implemented, businesses will find it increasingly difficult to self-assess the competition risk associated with R&D projects. Quite why the EC wishes to push up the cost of R&D and innovation in this way is unclear.


Between November 2019 and February 2020, the EC held an initial public consultation on its current rules for horizontal agreements. The feedback indicated that most respondents rarely or never used the Existing R&D BER and twice as many thought it reduced legal certainty as found it helpful. In its subsequent Evaluation paper the EC concluded that respondents considered reported the Existing R&D BER “[does] not provide sufficient legal certainty” (at page 45).

To lawyers and businesses accustomed to dealing with R&D collaborations, this was not news. The safe harbour offered by the Existing R&D BER is subject to a number of conditions, including:

  • that all parties to the joint R&D have “full access” to both
    • the final results of the R&D; and
    • the other parties’ pre-existing know-how (if necessary to exploit the final results); and
  • a 25% market share threshold in the “technology market” and the downstream product market.

Such stringent conditions might be appropriate in cases where the parties to a collaboration were competing head-to-head before entering into an agreement and could innovate just as effectively apart as well as together. But the vast majority of R&D arrangements are not of this type. Parties typically collaborate because doing so brings together their complementary technologies, skills and/or ability to access cash, allowing them to achieve more together than apart. In such cases, the cooperation will typically be procompetitive, whether or not meets the pre-conditions of the Existing R&D BER.

The Existing Guidelines recognize this in a number of places. They state clearly (at para 130) that: “[if] the parties are not able to carry out the necessary R&D independently … the R&D agreement will not normally have any restrictive effects”.

Similarly, they recognise that in some circumstances, granting all parties full access to R&D results may not be commercially feasible, in which case granting exclusive access to one party or the other will not prevent the deal from being lawful and pro-competitive (para 140). This will be the case where: “exclusive access rights are economically indispensable in view of the market, risk and scale of the investment required”.

Perhaps most importantly, they make it clear that R&D agreements will only constitute an automatic (by object) infringement of EU competition rules where: “they do not truly concern joint R&D, but serve as a tool to engage in a disguised cartel [through] price fixing, output limitation or market allocations” (para 128).

For most R&D arrangements this comfort is sufficient to relieve them from the need to rely on – and therefore comply with the specific requirements of – the Existing R&D BER.

The proposed changes

The proposed changes in the Draft R&D BER are a disappointment. It retains all the major weaknesses of the Existing R&D BER, including the requirements of ‘full access’ to results and pre-existing know-how and 25% market share threshold. To these, it adds a further layer of complexity and uncertainty. Where the parties compete “in innovation”, three or more competing R&D efforts that are “comparable” must exist before the block exemption will apply (Art 6(3)). Precisely how parties should count the number of – doubtless confidential – third party innovation efforts, or assess whether they are comparable, is unclear. This proposed change is perhaps influenced by recent merger decisions in relation to pipeline products in the pharma sector, where divestments have been required where the number of independent competitors are reduced to fewer than three. This appears misguided: requirements to divest in cases such as GSK/Novartis or Hospira/Pfizer related to pipeline products rather than early-stage R&D. In those cases the contemplated acquisitions clearly resulted in a reduction of independent products; R&D collaborations by contrast are intended to increase innovation.

More problematic are the proposed changes in the Draft Guidelines. First, the comfort offered to agreements between parties unable to carry out the R&D independently has been materially qualified. The Draft Guidelines (para 72) now state only that: “R&D cooperation between not competing undertakings generally does not give rise to restrictive horizontal effects on competition” (emphasis added).

This raises the spectre of R&D agreements giving rise to restrictive vertical effects, without offering guidance on how to identify or avoid such effects.

Second, the recognition that, if exclusive access rights are commercially indispensable, R&D agreements may be lawful despite not giving all parties “full access” to prior rights and results – and therefore falling outside the block exemption – has been deleted altogether.

Third, the Draft Guidelines indicate that R&D agreements may operate as automatic (by object) infringements of EU competition rules in a much wider range of circumstances (para 70), including where they: “(a) prevent or delay the market entry of products or technologies, (b) coordinate the characteristics of products or technologies which are not covered by the R&D agreement or (c) limit the potential of a jointly developed product or technology when they bring such a product or technology individually to the market.”

The third by object category, agreements that “limit the potential of a jointly developed product or technology” is particularly troubling given that it is entirely unclear what this might mean.


The combined effect of the proposed changes in the drafts is to limit the scope of the block exemption safe harbour and reduce the comfort available to R&D agreements that fall outside that safe harbour. This will increase the cost of risk assessments in relation to such agreements. Access to robust and well-informed external legal advice will increasingly be required – but it will come at a cost.

It is well recognised that innovation is a key driver of economic growth (including by the EC itself, as evident in DG Research and Innovation’s Strategic Plan for 2020-2024), and R&D is of course a key driver of innovation. Quite why the EC thinks it is appropriate to hinder R&D in Europe in this way is unclear.