On 18 May 2022,[1] the European General Court rejected Canon’s appeal against a €28 million fine imposed by the European Commission in 2019 for its breach of EU gun-jumping rules, just a few months after a similar judgment,[2] therefore vindicating the Commission’s aggressive stance on such breaches.

Background

The Commission’s fine was imposed on Canon due to its use of a warehousing structure in its 2016 acquisition of Toshiba Medical Systems (TMS). Such a structure involves splitting a sale into two steps—the target business is first transferred to an interim buyer who temporarily acquires the target from the seller on behalf of the ultimate buyer, without the need for regulatory clearance. The ultimate buyer then acquires the target after all required regulatory clearances have been obtained.

In this specific case, an interim buyer had acquired 95% of TMS for €800, whilst Canon paid €5.28 billion for the remaining 5% and share options over the interim buyer’s stake. Canon then notified its intention to acquire TMS to the Commission which cleared the transaction unconditionally. Only after such clearance and all other regulatory approvals did Canon exercise its share option over the 95% shareholding held by the interim buyer and take full ownership of TMS.

The Commission concluded that taken together, the two steps of the transaction formed a single notifiable concentration such that the first step (i.e. the acquisition of 95% by the interim buyer and 5% by Canon) constituted a partial implementation of the transaction prior to notification and clearance. In the Commission’s view, this infringed both the notification obligation (pursuant to Article 4(1) EU Merger Regulation or EUMR, requiring notification of a transaction meeting the EUMR jurisdictional thresholds prior to implementation) and the standstill obligation (pursuant to Article 7(1) EUMR, providing that transactions cannot be implemented before clearance).

The Judgment

In its judgment, the General Court rejected Canon’s arguments in their entirety.

First, Canon argued that the first step of the transaction did not amount to an acquisition of control and could not therefore constitute an early implementation of the transaction. In doing so, it relied on the 2018 preliminary ruling in Ernst & Young.[3] Although it was undisputed by the Commission that TMS was not controlled by Canon during the first step of the transaction, the General Court nonetheless held that the concept of “implementation of a concentration” is not only limited to the situation in which the ultimate purchaser acquires full or partial control of the target, “but also covers any transaction which ‘contributes’ to such a change in control” in whole or in part.[4]

Second, Canon submitted that there had been no partial implementation of the transaction at the first step. In response, the General Court held that:

  • The Commission was correct to consider that the first step of the transaction was “only undertaken in view of the ultimate transaction.”[5]
  • The Commission was also correct to find that the interim purchaser was created for the sole purpose of facilitating Canon’s acquisition of control of TMS, and not that such structure was devised solely in Toshiba’s interests as seller.[6]
  • Canon’s share options in this case were not “genuine” options that would give it the right to buy TMS at a later stage, but Canon effectively “paid the full price of the acquisition of [TMS] in exchange for a special, de facto automatic mechanism for acquiring it or for having the right to sell it to a third party of its choice.[7]
  • The fact that Canon bore the economic risk of the entire transaction at the first step was a relevant criterion for finding that the two steps amounted to one single concentration.[8]

Third, Canon argued that the Commission incorrectly conflated the question of whether multiple transactions amount to a “single concentration” with the notion of early implementation contrary to the notification and standstill obligations. The General Court rejected this, noting that transactions found to fall within the concept of a “single concentration” would de facto be captured within the scope of Article 7 EUMR and logically, Article 4 EUMR.[9]

Finally, Canon challenged the Commission’s calculation of its €28 million fine pursuant to Article 14 EUMR which, unlike the 2006 Fine Guidelines applicable to infringements of Articles 101 and 102 TFEU, does not provide a detailed scheme for calculating fines for EUMR breaches but merely notes that the Commission must consider the nature, gravity and duration of the infringement as well as general EU law principles such as proportionality. These arguments were also rejected by the General Court in their entirety and the fine was upheld in full.

Key Takeaways

This was a risky case for the Commission. Warehousing structures had long been discouraged, but it was not clear that the Court would follow along. In a judgment from September 2021,[10] the General Court judgment reasoned that the possibility of exercising decisive influence over a target was needed to breach the standstill obligation. Such a possibility either exists or it doesn’t—and it was certainly possible to find that Canon didn’t obtain that possibility at the first stage of the transaction, given that it held only 5% of the target.

However, the General Court instead emphasised the 2018 Court of Justice ruling in Ernst & Young which found that:

In the light of the foregoing, it must be concluded that Article 7(1) of Regulation No 139/2004 must be interpreted as meaning that a concentration is implemented only by a transaction which, in whole or in part, in fact or in law, contributes to the change in control of the target undertaking.[11]

In the Canon judgment this week, the General Court broadens this further, suggesting that “partial implementation” of a transaction includes any steps that could be regarded as contributing to a lasting change of control over the target, even partially.

Pending any appeal, we now know that warehousing structures don't work—but they were rarely attempted anyway. But the judgment has implications beyond that—any preparatory steps that could contribute to the ultimate change of control is potentially gun jumping. Consider how that works in a greenfield JV; the Parties notify the JV itself, but pending clearance need to enter into ancillary agreements—employment contracts for senior JV staff for example. Could these be said to be contributing at least partially to the implementation of the transaction? These sorts of questions were not the focus of the discussion in Canon, but will need ever more careful and fiddly counselling going forwards.