The CMA’s pharma cases have generated a lot of work for the UK’s antitrust lawyers and economists, with a good proportion of practitioners having worked on at least one of Paroxetine, Phenytoin, Liothyronine or Hydrocortisone over the past few years. Dawn raids and large fines in these pharma cases highlighted the emergence of a more assertive CMA.
The CAT has now heard appeals of each of these CMA infringement decisions. The Court of Appeal has heard subsequent appeals in two out of four, with a third (Liothyronine) due later this year. Paroxetine went one better, having the honour of being the last preliminary reference to the EU Court of Justice in a UK competition case before Brexit, which led to the landmark Generics UK ruling. The results have been mixed with a win in Paroxetine but an early setback in Phenytoin (see also previous articles on Phenytoin and Liothyronine here).
Court of Appeal’s Hydrocortisone judgment is another vindication for the CMA
Last week’s Court of Appeal judgment in Hydrocortisone marked a significant positive result for the CMA following its record £260m 2021 infringement decision.
To the extent that we are approaching the “end of the beginning” in the life-cycle of this cluster of cases, it increasingly looks like the CMA’s aggressive interventions in the generics pharma sector will be judged a success.
The Hydrocortisone judgment shows the Court of Appeal grasping a case that looked procedurally to be getting out of hand and imposing order in the CMA’s favour.
The judgment addresses procedural and due process aspects of the case (rather than arguments of substance). The Court of Appeal highlights unusual and “even more unusual” procedural steps taken by the CAT. The CAT’s willingness to broaden the CMA’s original enquiry and to re-open factual questions seems to have been the issue that most vexed the Court of Appeal judges.
In its more generous passages the Court of Appeal gives the impression of the CAT as a well-meaning if confused appellate tribunal. See for example the references to the CAT doing “what it thought right” but having “somehow lost sight of the essential realities of this kind of appeal” (paragraph 16). Or references to the “obviously painstakingly considered” position taken by the CAT which resulted in a judgment that was “significantly overthought” (paragraph 109).
Elsewhere the Court of Appeal is more damning. It describes a CAT procedure that was “inappropriate in all the circumstances of this case” (paragraph 96) and is cutting in its descriptions of mistakes that the CAT made (paragraph 118).
The judgment will be a difficult read for the pharmaceutical and private equity companies involved as well as the executives who were cross-examined. But reading the judgment will likely be no less bruising an experience for the CAT.
Background on the CMA decision and CAT judgments
The original Hydrocortisone decision included findings of abuse of dominance (excessive pricing) combined with pay-for-delay type agreements in relation to the generic drug hydrocortisone. That drug is used as the primary replacement therapy for people whose adrenal glands do not produce enough natural steroid hormone, for example people with Addison’s disease (a life-threatening condition). The UK NHS had spent £500k per year on the drug in 2008 but this had increased to £70m per year by 2015. 10mg tablets saw price rises over 12,000% compared to the branded version of the drug. 20mg tablets saw price increases of nearly 9,500% compared to the branded price (branded drugs were subject to price control whereas unbranded drugs had free pricing).
Within two years of the CMA’s £260m infringement decision, the CAT issued four judgments. First, on 18 September 2023 it issued an “abuse of dominance” judgment that upheld the CMA decision on excessive pricing (ie on the Chapter II prohibition on abuse of a dominant position), although it reduced the total penalty imposed on Allergan. Second, on 29 September 2023, it issued a “cartel” judgment that agreed with the CMA on the merits on the Chapter I prohibition but sought additional evidence and an additional hearing in relation to due process. Third, on 8 March 2024, it issued a “due process” judgment (following the additional hearing it had requested) in which it overturned its own second judgment of 29 September 2023. Finally, it issued a fourth “penalties” judgment in relation to the cartel infringement relating to 20mg hydrocortisone tablets, which had only been appealed insofar as it concerned the penalty that the CMA had imposed. It is the CAT’s second and third judgments—and in particular its findings that the CMA did not properly put its case to the witnesses—that were the subject of the Court of Appeal’s judgment.
Essentially, the CAT was concerned about how the CMA had conducted cross-examination of two witnesses for Advanz, one of the appellants, during the appeal. Those witnesses had denied the existence of a cartel agreement. The CAT took issue with the fact that when the CMA cross-examined the witnesses, it did not put to them that it alleged that their company Advanz was indeed involved in an anti-competitive agreement, despite what they had initially said in the witness box. Since the CAT would, in upholding the CMA’s decision, be saying that the witnesses’ statements had been untrue, a detailed cross-examination was called for, according to the CAT. The CAT therefore found that its own “cartel” judgment was reached in a procedurally flawed manner, and had to be set aside.
Much of the interest of the Court of Appeal judgment is case specific. The cross-examination was the main cause of contention. The CAT’s procedure elaborated above was undoubtedly unconventional and we could spend a lot of time discussing that. But it is perhaps more instructive in the remainder of this blog to focus mainly on any points of wider application in competition law cases.
Like ships in the night – does the CMA need to prove dishonesty, and if so does it need to explicitly put that dishonesty to witnesses in cross examination?
The Court of Appeal described the issues in the Hydrocortisone appeal as simple. The CMA considered it could make a finding of a breach of the Chapter I prohibition without making a finding of dishonesty. The Court of Appeal agreed. The CAT had found that the CMA’s case necessarily implied dishonesty and that this dishonesty should have been put to the witnesses in cross-examination.
The Court of Appeal judgment is clear. There is no obligation on the CMA to allege or prove dishonesty in any particular case, even where that state of mind might plausibly be implied. The legislation does not require it and nor should the CAT. The Court of Appeal noted that the CAT’s suggested approach would complicate many, if not most, competition cases and “would make the CMA’s task impossible” (paragraph 88).
On whether the CMA had, in fact, put its case to the witnesses under cross-examination, the Court of Appeal had painstakingly gone back through the transcripts. It found the CAT was simply wrong to suggest that the main witness had been unaware of the case against him. It was clear from the transcripts that he had understood the CMA to be implying he had been involved in an unwritten pay-for-delay type arrangement (paragraph 61). The Court of Appeal complimented CMA’s counsel for the way they had cross-examined a “recalcitrant witness” without needing to go so far as putting that they were dishonest.
The Court of Appeal saw the approaches of the CMA and the CAT as passing by each other like “ships in the night”. The CMA had “stuck to its guns” in arguing that dishonesty need not be alleged or proved. The Court of Appeal agreed with the CMA and not the CAT.
An inappropriate procedure – what is it open for the CAT to do in an appeal of a CMA decision?
In a part of the judgment that may be a useful reference point in future cases, the Court of Appeal reaffirmed the conventional view of what the CAT is, and is not, able to decide on when reviewing a CMA decision.
The Court of Appeal confirmed that the regulator will be expected to defend an infringement decision on the basis of the material before it when the decision was taken and not by elaboration or extension of its evidence (paragraph 94). By contrast, an appellant is allowed to present a new case supported by new evidence – this is because a CMA infringement decision is quasi-criminal (for the purpose of Article 6 of the European Convention on Human Rights), and the CAT appeal constitutes the first judicial consideration of the allegations made against the appellant.
The CAT can put itself in the shoes of the CMA and make decisions or give directions that the CMA could have made (paragraph 93).
What the CAT cannot do is what it did in this case. That was to (i) deliver a provisional decision, (ii) re-open the argument on fundamental points, and (iii) then to reverse its own provisional decision.
The Court of Appeal noted, somewhat drily, that the CAT’s procedural error had left the Court of Appeal in a “difficult position” (paragraph 100).
Getting out of a difficult position
What is perhaps most remarkable about the judgment is that in spite of the procedural oddities, the Court of Appeal appears to have succeeded in tying up almost all the loose ends.
At the start of the hearing in the summer of 2024 the Court of Appeal had asked the CMA to amend its appeal to seek to overturn those parts of the CAT’s 29 September 2023 judgment that (i) dealt with dishonesty, (ii) declined to determine the companies’ appeals, and (iii) required a further hearing (see paragraph 10).
In so doing, the Court of Appeal ensured that it had all the pleadings and information it needed in order to resolve the case in a single judgment. The case relating to the pay for delay agreement will not be remitted to the CAT or the CMA. The only avenue left for the parties is an appeal to the Supreme Court. The final step in relation to this infringement is now for the CAT to rule on the parties’ appeals on penalties, which the CAT had so far declined to do since it had overturned its own “cartel” judgment.
This contrasts to Phenytoin where the Court of Appeal judgment led to the case being remitted to the CMA which then had to re-issue a new infringement decision. Of course they are different cases, and the requirement for the CMA to re-issue in Phenytoin was driven by a need for additional evidence which only the CMA could gather and assess (see our Phenytoin blogpost here). But the Court of Appeal’s desire to impose order this time will be a welcome development for the CMA.
In Phenytoin, the CMA’s original 2016 infringement decision was not decided upon by the Court of Appeal until 2020. It was only in 2022 that the CMA re-issued its new/remittal Phenytoin decision. The appeal of the re-issued decision was heard before the CAT last year and, to the extent it ends up at the Court of Appeal again that judgment is likely to be 2025 at the earliest (almost 10 years on from the original infringement decision).
The next steps in Hydrocortisone are as follows:
- In relation to the excessive pricing abuse, the CAT has ruled on substance and penalties. The Court of Appeal is still considering whether or not to grant permission to appeal.
- In relation to the pay for delay agreement that gave rise to the “cartel judgment”, the Court of Appeal has ruled out any further appeals on the question whether there was an infringement of the Chapter I prohibition. However, since the CAT has not yet ruled on penalties, it is now first for the CAT to decide on the parties’ appeals insofar as they relate to the imposition of the penalties. In doing so, it will be bound by the findings of the Court of Appeal.
- It is open to the parties to seek permission to appeal the Court of Appeal’s judgment to the Supreme Court.
So Hydrocortisone is not done yet, but it does seem like it will be resolved more quickly than Phenytoin and the courts seem to be getting to grips with how to impose order in these cases.
Leave a Reply