Publication
Publication
Competition authorities are in the business of predicting the future. As we've written previously, the task of comparing two possible futures – one with, and one without the merger (the counterfactual) – is hard enough in the best of circumstances, especially in fast-moving markets. The assessment is even more complex when the sands are potentially shifting in multiple directions, as is the case when there is the prospect of another merger in the same market (a parallel transaction). So how do authorities disentangle multiple possible futures in practice and why does it matter?
Parallel transactions are an often-overlooked area of risk for merger parties. Whether a parallel transaction is factored into the analysis will determine the “counterfactual” against which the transaction is assessed. This can be outcome critical: a counterfactual in which a parallel transaction takes place will invariably present a more challenging situation for merger parties than not, for the simple reason that the remaining number of players in the market will be lower.
As M&A picks up and we see increased deal volume, this challenge is a live one in an increasing number of cases. As we explain, while the position in the EU is clear, the UK regime can be more complex to navigate. This Platypus post attempts to provide some guidance on how merger parties can navigate the challenges raised by parallel transactions (or the multi-headed substantial lessening of competition (SLC) analysis). As we explain, for mergers where a parallel transaction is in play, the only way to be confident of slaying the Hydra in the UK is to plan to engage with all potential heads of an SLC analysis.
At the EU level, for deals which meet its mandatory and suspensory thresholds, the European Commission’s approach to the counterfactual is based on the ‘first past the post’ approach. The treatment of parallel transactions is therefore clear: whichever party makes it first to the starting line (in either Phase 1 or 2) secures the counterfactual in which the parties in the parallel transaction are deemed independent players.
Back in Platypus’ natural habitat, where there is no obligation to notify transactions (and where the legal threshold for finding competition problems at Phase 1 is low), the ‘first past the post rule’ may be relevant to the assessment of jurisdictional thresholds. A transaction which previously fell out of the CMA’s jurisdictional scope could be brought into scope as a result of a parallel transaction (for example where the supply test is based on a fascia count). When it comes to substantive assessment however, no such simple procedural rule applies and the assessment of the appropriate counterfactual in parallel transactions is necessarily more nuanced and difficult to predict.
Depending on the circumstances of each case, the CMA will adopt one of the following counterfactuals: (i) prevailing or pre-merger conditions of competition; (ii) stronger competition – e.g. where new entrants are expected; or, (iii) conditions of weaker competition – e.g. where a competitor is exiting or critically for current purposes, merging with another competitor (i.e. a parallel transaction).
The CMA’s assessment of the counterfactual varies across Phase 1 and Phase 2, reflecting the different applicable thresholds. At Phase 1, the CMA has a duty to refer a merger which gives rise to a realistic prospect of an SLC. This threshold is very low and, as a result, the CMA adopts a cautious approach to the counterfactual at Phase 1.
In particular, where the CMA is presented with multiple potential counterfactuals and each of those scenarios is a “realistic prospect”, the CMA will assess the merger against the counterfactual in which the merger firms exert the strongest competitive constraint on each other, and where third parties exert the weakest competitive constraints on the merger firms. That is, at Phase 1, the glass of counterfactual competition is always half empty (consistent with the low legal threshold for reference).
At Phase 2 on the other hand, the substantive threshold is whether the merger gives rise to an SLC on a balance of probabilities. This higher threshold translates into the CMA selecting “the most likely conditions of competition” as its counterfactual against which to assess the merger.
The principle that counterfactuals with and without the parallel transaction would be considered in Phase 1 (unless the parallel transaction could “clearly be ruled out as too speculative”) was explicitly written into the CMA’s previous (2010) Merger Assessment Guidelines. The current Merger Assessment Guidelines (MAGs), which were revised in 2021, removed specific guidance on parallel transactions, now only noting that, in Phase 1, the CMA will consider “multiple potential counterfactual scenarios” provided each is a “realistic prospect” (at paragraph 3.12). The MAGs also note (at paragraph 3.10) that “Significant changes affecting competition from third parties which would occur with or without the merger (and therefore form a part of the counterfactual) are unlikely to be assessed in any depth as part of the CMA’s counterfactual assessment” and (at paragraph 3.11): “The CMA seeks to avoid predicting the precise details or circumstances that would have arisen absent the merger.”
The CMA and its predecessor the Office of Fair Trading (OFT) have often avoided publishing detailed assessments of the likely counterfactual. When the decision is a Phase 1 clearance, on a cautious basis, the regulator typically concludes that there is no realistic prospect of an SLC under either counterfactual – see, for example, Xchanging Holdings/Total Objects; Skanska Construction/Atkins; NASDAQ/LSE; and UCL/Royal Free London.
Cases considering parallel transactions fall into two broad categories.
Where a parallel transaction has already received all regulatory approvals, the CMA considers the merged entity as a single player, regardless of whether closing has taken place. In Sika/MBCC, for its Phase 2 assessment, the CMA took into account a parallel transaction (Saint-Gobain/GCP) that received CMA clearance during the course of the CMA’s Phase 2 review.
This is where the Hydra starts to grow multiple heads and there are limited precedents that consider this scenario in detail. In 2006 in NYSE/Euronext, for example, the OFT in its phase 1 decision concluded that the parallel transaction (between Nasdaq and LSE) did not qualify as “sufficiently likely and imminent”. The OFT noted that there had been speculation as to many possible merger permutations among the major European and US exchanges but that no major transaction between key players had actually completed. That the parallel transaction would have required a substantive competition assessment appears to have influenced this decision – the OFT noted in particular that “to assume that Nasdaq would acquire full control of the LSE … would, among other things, prejudge competition clearance” and concluded that the appropriate counterfactual was the status quo ante. Conversely, for the assessment of NASDAQ/LSE, the OFT did take into account the parallel transaction. It noted that, irrespective of the likelihood of clearance and completion, it would consider the relevant counterfactual in this case on the basis of a completed merger between NYSE and Euronext.
In the more recent BT/EE, the CMA dismissed arguments from the parties that the parallel transaction (H3G/O2) was subject to merger control clearance in another jurisdiction (the EU), the outcome of which was “highly speculative”. The merger parties submitted that to predict such a counterfactual would require the CMA to prejudge the outcome of the European Commission’s review. They argued that taking a counterfactual with H3G/O2 merged was not realistic, and that the correct counterfactual should be the continuation of the prevailing conditions of competition absent the parallel transaction.
The CMA dismissed these arguments, noting that since the transaction had “been signed and announced … [i]t is not therefore too speculative”. The CMA did recognise that taking into account the H3G/O2 merger posed some challenges, as the Commission’s assessment of the H3G/O2 merger could lead to a range of possible outcomes (including an outright prohibition, an unconditional clearance, or a conditional clearance). However, the CMA concluded that the unconditional clearance of the H3G/O2 merger is the most appropriate counterfactual at Phase 1 as it is “the most cautious counterfactual for the purposes of the CMA’s competitive assessment … recognising that the CMA is not in a position to conduct a competitive assessment of the H3G/O2 merger”.
The CMA was not persuaded by submissions that it was possible to predict the likely outcome of the parallel transaction by reference to previous commitments accepted. The CMA instead countered that recent Commission decisions in this sector had uncertain precedent value since these related to competition in different Member States with different competitive environments.
The stage of the parallel review also appears to be relevant. The Saint Gobain/GCP merger investigation took place in the context of a parallel transaction involving the anticipated acquisition by Sika of MBCC. At that time, the parties in Sika/MBCC had fast-tracked the Phase 2 review to remedies. Unsurprisingly, the CMA concluded that there was no longer a realistic prospect that Sika’s and MBCC’s chemical admixtures businesses in the UK could operate under combined ownership in the counterfactual. The logic here is clear: the parties had waived their procedural rights and the only possible outcomes were an outright prohibition or a conditional Phase 2 clearance that, in turn, would be expected to restore the competitive situation existing before the merger (as noted in the BT/EE Phase 2 decision).
Merger review can be a process fraught with timing risks and uncertainties. Including a parallel transaction as a possible counterfactual significantly increases the factual permutations at play and makes it more difficult to assess the risk of a deal in advance (particularly when consolidation after the deal is signed could ultimately make a difference to the CMA’s assessment).
Approaching mergers with an understanding of parallel transaction risk is critical to ensure merger parties can adapt and respond effectively. The legal framework within which the CMA operates means that the CMA’s starting point, at least at Phase 1, will include assessment against a counterfactual in which a parallel transaction has completed. However, as the above cases demonstrate, the stage of the parallel transaction and the specifics of the case (e.g. whether there have been previous failed bids) are likely to be relevant considerations in the CMA’s assessment.
The bottom line is, that in the UK, merger parties wanting to avoid potential outcome-critical CMA intervention need to be confident that their transaction does not give rise to an SLC under any realistic counterfactual – i.e. to know they can slay all the Hydra’s heads.
