On September 3, 2024, in a landmark decision, the European Court of Justice – the EU’s highest court – ruled in favor of Illumina in its challenge to the EC’s unprecedented assertion of jurisdiction over a transaction that met no notification thresholds at either EU or Member State level.

Cleary Gottlieb acted for Illumina in overturning the General Court’s earlier judgment, which had found in favor of the EC.

The Court’s press release is available here and the judgment here.

Judgment

In 2020, Illumina agreed to acquire GRAIL for USD 8 billion.  The transaction was not reportable, but the EC nevertheless sought to review it by first requesting and then accepting a referral request by France under Article 22 of the EU Merger Regulation (EUMR).  This was the first time the EC had sought to apply Article 22 to review transactions that did not otherwise qualify for review.

The ECJ has now ruled this to be unlawful.  A Member State with domestic merger control rules cannot seek an Article 22 referral if the transaction does not fall within its national merger control rules.

In arriving at this conclusion, the Court reassessed the General Court’s literal, historical, contextual, and teleological interpretations of Article 22.

  • Literal interpretation: Even where a provision “appears to be clear,” the Courts may still resort to other methods of interpretation “to clarify [its] exact scope,”  justifying a historical, contextual, and teleological interpretation of the provision. [1]
  • Historical interpretation: The Court went beyond the General Court’s assessment to consider Article 22’s legislative history (travaux préparatoires), which the Court found to contradict the General Court’s conclusions.  Article 22, in particular, was enacted to address the absence of national merger control regimes in certain Member States and not the fact that certain concentrations that “could affect the internal market would, in any event, escape ex ante review” under the EUMR.[2]
  • Contextual interpretation: The contextual factors that the General Court considered in the judgment under appeal were inconclusive in supporting the broad interpretation of Article 22.  By contrast, other factors that the General Court disregarded supported Illumina’s and GRAIL’s position.  Notably, the General Court failed to consider the existence of a dedicated mechanism in the EUMR to revise the thresholds – Article 1(4) and (5) – allowing for the “rapid adjustment” of the thresholds should they no longer be “apt to capture concentrations with potentially harmful effects.”[3]
  • Teleological interpretation: the Court held, among other findings, that:
    • Article 22 was never intended as a “corrective mechanism” to remedy merger control deficiencies by allowing the EC to review below-the-thresholds mergers.  Its object was to allow (i) merger control on behalf of Member States that had no national rules and (ii) extend the “one-stop-shop” principle to avoid multiple national filings. [4]
    • The broad interpretation of Article 22 was also “inconsistent with” the EUMR’s objectives of legal certainty, effectiveness, and predictability, in particular as it compromised “a clear allocation”of competences among the EC and Member States, a “predictable system of control” for undertakings that would not require informal notifications to each NCA, and the timely review of concentrations.[5]
    • A broad reading of Article 22 was not needed to ensure the effective control of concentrations with significant effects in the EU, as below-the-thresholds mergers can be subject to control by NCAs and national courts on the basis of Article 102 TFEU as confirmed in Towercast.[6]
    • Article 22’s broad interpretation was at odds with the principle of institutional balance, as “it is for the EU legislature alone” to review the thresholds or to “provide for a safeguard mechanism” for the EC to scrutinize below-the-thresholds transactions, and “it is open to the Member States to revise downwards their own thresholds.”[7]

Implications

The judgment has a number of implications.

  • The judgment limits the main direct avenue the EC intended to use to scrutinize concentrations falling below both the EUMR and national merger control thresholds.  Such concentrations included so-called horizontal “killer acquisitions,” but also vertical acquisitions such as the Illumina/GRAIL transaction.
  • Unless a transaction is reviewable on the basis of national merger control rules (or the Member State has no national regime at all, which remains the case only for Luxembourg), an EU Member State will no longer be able to request that a case be referred to the EC. 
  • The information sharing on concentrations contemplated by gatekeepers per Article 14 of the Digital Markets Act (DMA) will be of less practical relevance now that national authorities can only use Article 22 EUMR to refer concentrations that fall under their national merger control regime, as Member States will generally already know about them through mandatory filings.  Nonetheless, the EC may still seek to encourage referrals by Member States with voluntary filing regimes, low jurisdictional thresholds (Cyprus, Sweden), or broad “call in” powers (Italy, Ireland) following an Article 14 DMA notification.
  • In the short term, the EC may consider pivoting to different methods of enforcement if it wishes to continue reviewing these types of transactions.  The recent Towercast judgment offers an alternative to enforcement through the EUMR, but one that is limited to acquirers that hold a dominant market position.  The Towercast case made clear that the EC may challenge concentrations ex post under general antitrust provisions, notably Article 102 TFEU, irrespective of the existence of a dedicated EU merger control regime.
  • In parallel, Member States may continue to take action to bring more concentrations within the purview of their national merger control rules.   In recent years, Member States, such as Italy and Ireland, have enacted rules allowing their competition authority to call in and review transactions that fall below national thresholds altogether.  Other Member States may follow suit or seek to lower their national thresholds.  The EC may rely on such mechanisms to allow it to review these transactions through Article 22 EUMR, in a way that was not foreseen by the legislator and may lead to further litigation.
  • Such national trends would be consistent with the use that national authorities have made of the EC’s broad (and unlawful) reading of Article 22.  To date, more than half of the Member States had either made or joined at least one request for a referral of a below-the-thresholds merger based on Article 22 EUMR.
  • In line with the ECJ judgement, the EC may, and should if it wishes to review more transactions at EU level, seek a revision of the EUMR’s notification thresholds.  The EUMR provides for a specific mechanism to do so (now Article 1(5)), and the EC has relied on it before.  The existence of a specific legal basis to change the thresholds should avert any perceived risk that initiating a legislative process to revise the EUMR may give Member States the opportunity to make other amendments to the EUMR that the EC may not support (which may be a reason why the EC chose to rely on an unlawful interpretation of Article 22 in the first place).
  • In sum, following this important judgement, there is no doubt that transactions falling below EU thresholds and that do not fall within national merger control rules will no longer be subject to merger control in the EU, in line with what had been the case over the 30 years of application of the EUMR and its precursors, and in line with the original purpose of Article 22 EUMR, which was to serve as a tool allowing Member States that did not have a national merger control regime in place to refer cases to the EC.

[1] Judgment of September 3, 2024, Illumina and GRAIL and Commission, Cases C-622/11 and C-625/22 (Judgment), paras. 127-128

[2] Judgment, paras. 146-148.

[3] Judgment, paras. 175-184.

[4] Judgment, paras. 191-199.

[5] Judgment, paras. 203-205.

[6] Judgment, paras. 211-214.

[7] Judgment, paras. 215-217; Judgment of March 16, 2023, Towercast, Case C-449/21.