Competition: The EU and UK extend the scope of merger control

For several years, law circles have been concerned over a perceived lack of enforcement in the face of increasing market concentration. Until recently, the main result appeared to be a marked increase in intervention against US tech giants including Google and Facebook. However, recent developments in both EU and UK merger control suggest entities in other innovation-heavy sectors – including biotech and pharma – are likely to find themselves increasingly in the crosshairs, sometimes even where their links to the jurisdictions appear slight.

In the UK, the Competition and Markets Authority (CMA) has been flexing its merger control muscles as it develops its new role in a landscape remade by Brexit, most recently by imposing a fine of over £50m on Facebook for procedural infringements. Part of that process has been establishing a record for intervening in mergers with little if any jurisdictional link to the UK in cases such as Roche/Spark and Sabre/Farelogix (more on these below). the Competition Appeal Tribunal (CAT)’s judgment in May upholding the CMA’s jurisdiction in Spark/Farelogix means this process is likely to continue. And the CMA’s focus on innovation issues means deals in the tech, pharma and biotech space will be a particular focus.  

The Illumina/Grail deal

Illumina develops and commercialises next generation sequencing (NGS) systems for genetic and genomic analysis as well as the services and consumables needed to operate them. It is a world leader in the field, with – according to an earlier finding by the CMA – a global share of DNA sequencing systems in excess of 80 per cent. On 20 September 2020, Illumina announced its acquisition of the US oncology company Grail for $8bn.

Grail has developed tests intended to allow the early detection of around 50 cancer types via blood samples in patients that are not showing symptoms. It is still in the early stages of commercialising these products. Grail began limited commercial activities in the US in April 2021, but it still has no EU sales. Grail’s tests are based on genomic sequencing and rely on NGS systems. Indeed, Grail started life as an Illumina business unit before being spun out in 2016. 

On 22 July, the EC opened an in-depth phase II investigation of the deal on the basis it raised substantial concerns about the impact of the transaction on the development and supply of NGS-based cancer detection tests. In particular, existing and potential competitors to Grail in the cancer detection test space also rely on Illumina’s sequencing technology. Thus, it is concerned Illumina may cut short innovation efforts that might result in alternatives to Grail’s technology. It is rare for the EC to oppose a merger between non-competitors and an even rarer example of an innovation theory based on threats to future competition in a vertical context. This would therefore appear to be an example of the EC’s increasing interest in pursuing non-standard theories of harm involving effects on innovation.   

The jurisdictional issues

The primary basis for jurisdiction under the EUMR – and most national merger control laws in Europe – are turnover thresholds which the parties must fulfil, typically including some minimum turnover for the target in the relevant jurisdiction. Since, in this case, Grail had no sales in any EU Member State, the acquisition was not originally caught by the EUMR or national merger control rules in the EU. As a result, the parties were not automatically obliged to notify the deal to the EU for clearance prior to completion.

However, Article 22 of the EUMR allows member states to request the EC initiates a merger review even where the EU turnover thresholds are not met. In this case, six EEA member states – France, Belgium, Greece, the Netherlands, Iceland, and Norway – submitted requests asking the EC to take jurisdiction over the Illumina/Grail deal. The EC accepted the requests and opened an initial phase I investigation on 19 April 2021. On 22 July, it opened an in-depth phase II investigation, indicating it had substantial concerns in relation to the merger.

Article 22 dates back to the original version of the EUMR which entered into force in 1990. It was designed to allow those Member States that in 1990 had no national merger control regimes – particularly the Netherlands – to correct that deficit by referring potentially problematic mergers to the EC. As a result, Article 22 became known as the ‘Dutch clause.’

Over time, the EC had, in its own words, developed a practice of “discouraging” Article 22 requests from Member States that did not have jurisdiction over the deal under their own merger control rules. This reflected both concerns regarding legal certainty and experience indicating such deals were not generally likely to have a significant impact on EU competition.

In recent years, concerns had arisen that companies might play a significant competitive role despite having little or no sales or turnover, as a result of their R&D and innovation potential. As a result, in April 2021, the EC issued the new guidance that overturned that previous practice. This is the first case in which the new policy in that guidance has been implemented. Illumina has, in other words, found itself in the unfortunate position of being the test case for a new policy, and one in place after it initiated the transaction. 

Gun jumping

The opening of the EC investigation automatically triggered Article 7 of the EUMR, the standstill obligation which prohibits parties from implementing a merger pending EC clearance. This created a serious problem for Illumina. Its deal to acquire Grail would expire on 20 December 2021. Once an in-depth investigation had been opened, EC clearance before that date was unlikely, and after a ‘stop-the-clock’ suspension in October, the deadline has extended into February 2022. 

Illumina therefore faced a choice between completing the deal in the face of the standstill obligation, or risking its collapse. Illumina adopted a dual strategy to deal with the situation. First, it challenged the EC’s jurisdiction over the transaction in the EU courts. Second, it completed the acquisition while offering undertakings to hold the two businesses separate.

Unfortunately for Illumina, the EC did not regard those undertakings as sufficient. On 20 September, it issued a Statement of Objections (SO) to Illumina and Grail informing them that it considered them to be in breach of the Article 7 standstill obligation and intended to adopt interim measures. On 29 October, those interim measures were adopted. This is the first time that interim measures have been imposed under the EUMR. They require, among others, that:

• Grail is run by independent hold separate management acting exclusively in Grail’s interests;
• Illumina supplies any additional funds necessary for the operation and development of Grail;
• Interactions between the parties take place on an arm’s length basis with no confidential information being shared outside the ordinary course of business; and
• Grail work on alternatives to the transaction.

The SO also raises the possibility of fines. ‘Gun jumping’, the partial implementation of deals before clearance, has recently become an enforcement priority for the EC. For example, in 2018 it imposed fines of €125m on the telecoms company Altice, which were upheld by the EU General Court in September this year.

The new Article 22 guidance

The new guidance is a response to long-standing EC concerns that the turnover-based thresholds of the EUMR were failing to catch problematic transactions. A particular focus in this regard are deals in digital, pharma and others sectors where innovation is central to competition. In digital markets, free services frequently launch with the aim of building up a significant user base and/or commercially valuable data inventories, before seeking to monetise the business. In pharma, innovators may have strong competitive potential, even before their R&D activities are complete. Such sectors are seen as vulnerable to so-called ‘killer acquisitions’ in which emerging competitors are acquired before they grow large enough to trigger notification requirements. 

The EC’s solution adopted is to actively encourage Article 22 requests (also known-as referrals) from Member States authorities (NCAs). The guidance suggests that the EC does not intend to passively wait for transactions of interest to be brought to its attention. Instead, it will seek to “identify concentrations that may constitute potential candidates for a referral” in close cooperation with Member State authorities. It seems the EC envisages something similar to the CMA’s existing Mergers Intelligence Committee, which actively monitors merger activity to allow the CMA to ‘call in’ transactions of interest.

Consistent with that CMA model, the guidance indicates that transactions will remain at risk of referral and review post-closing. This risk is potentially open ended, at least until after the “material facts about the concentration have been made public”. Although the EC will not “generally” consider a referral “appropriate” where more than six months has passed once that has occurred.

As the guidance makes clear, the legal threshold that must be satisfied before a referral can be made is extremely low. The transaction must (i) affect trade between Member States and (ii) threaten to significantly affect competition in the referring Member State. These criteria will be satisfied if the transaction may have “some discernible influence” on inter-state trade and a preliminary analysis based on prima facie evidence of a possible adverse competitive effect.

UK merger control expansionism

The jurisdictional thresholds under UK merger control rules are notoriously flexible. The CMA may take jurisdiction over a transaction if the activities of the parties overlap in the supply of goods or services “of some description” where they have a combined share of at least 25 per cent in the UK (the share of supply test). While the transaction must result in some increment in the parties’ combined share, the CMA has broad discretion over what constitutes an appropriate description of goods or services.

In a number of recent cases, including Roche/Spark and Sabre/Farelogix, the CMA has used the scope granted to it by the share of supply test to take jurisdiction over mergers with little or no obvious jurisdictional link to the UK. Several of these cases have been in innovation driven sectors like pharma and tech. Like the EC, the CMA seems particularly concerned about possible ‘killer acquisitions’ affecting innovation.

Roche/Spark

In February 2019, the biotech giant Roche Holdings (Roche) announced a $4.3bn acquisition of US-based Spark Therapeutics (Spark). In the UK, Roche’s activities included the supply of emicizumab, a novel treatment for haemophilia, but not one based on gene therapy. Spark had no commercial activities and no sales in the UK. It was however active in the development of a gene therapy treatment for haemophilia.

Since Spark was not active in the supply of any products in the UK, Roche took the view that there was no overlap or increment in the parties’ UK activities and did not notify the transaction to the CMA. 

However, the CMA disagreed. In June 2019 it issued an initial enforcement order (IEO) requiring the parties to hold their businesses separate pending CMA clearance. In the CMA’s view, Spark was active in the supply of haemophilia treatments in the UK given the “commercial realities of the pharmaceutical sector” where, evidence showed, companies with existing products alter their commercial behaviour to compete with pipeline products that remain in the R&D stage. Spark had employees in the UK active in its global R&D efforts and held UK/EU patents.

The CMA ultimately cleared the deal, in phase I and without conditions, on 10 February 2020.  However, the CMA has subsequently amended its official guidance to underline that pipeline products will be taken into account in assessing UK jurisdiction.

Sean-Paul Brankin is Of Counsel at Bristows. Isobel Thomas is a partner at Bristows bristows.com