Important lessons on excessive pricing: the CMA publishes its full decision in Pfizer/Flynn

United Kingdom

On 15 June 2017, the Competition & Markets Authority (CMA) published the full non-confidential text of its infringement decision in a case concerning pharmaceutical manufacturer Pfizer Limited and its distributor Flynn Pharma Limited (Pfizer and Flynn).  In this ground-breaking and controversial case, the CMA found Pfizer and Flynn had imposed excessive and unfair prices for phenytoin sodium capsules, an anti-epilepsy drug used in the treatment of an estimated 48,000 patients in the UK. Such pricing was found to be an abuse of each party’s respective dominant position in breach of both UK and EU competition law.  The CMA had previously announced its decision on 7 December 2016 to impose a record fine of £84.2 million on Pfizer and a fine of £5.2 million on Flynn for these practices.  Both parties have appealed the decision.

The CMA’s full decision, which weighs in at over 500 pages, is a comprehensive analysis of the reasoning behind its infringement decision. The following article comments on general trends in the enforcement of excessive pricing cases and then highlights key insights which can be derived from the recently published full decision in Pfizer/Flynn and which help understand how prices may sometimes be considered excessive and abusive of a dominant position. Our recent law-now also highlighted important trends in abusive pricing cases.

Background to excessive pricing and the Pfizer/Flynn case

A new focus on excessive pricing

Abusive excessive pricing by a pharmaceutical company was the subject of one of the first decisions under the UK’s Competition Act 1998 (in the Napp Pharmaceuticals case in 2001).  Despite the Napp case, however, there has been relatively little enforcement against excessive pricing under UK or EU competition law.  That seems to have changed in the last two or three years, with a series of new investigations and decisions announced, very much focusing on the pharma industry.

The CMA’s Pfizer/Flynn case is not the only live excessive pricing case in the sector.  The CMA itself has announced three other ongoing cases (two of them related), involving Actavis, Amdipharm and Concordia.  On 15 May 2017, the Commission announced an investigation into concerns that Aspen Pharma was setting excessive prices for five of its anticancer drugs, following an infringement decision against Aspen in Italy. There have also been announcements in 2017 of new investigations into excessive pricing allegations against a number of companies in Spain and South Africa.

Why the sudden increase in excessive pricing cases?

It has long been thought that there are challenging practical difficulties in finding economic evidence to satisfy the legal test for establishing excessive pricing under EU and UK competition law and that these difficulties explain the low incidence of cases on this subject. The test derives from the 1978 EU case United Brands Co. v Commission and requires each of the following two points to be established:

  • first, the difference between the costs incurred and the price charged must be excessive; and
  • second, the price in question must also be unfair either in itself or when compared to competing products.

The first limb of this test is difficult to apply, since  there will usually be no ready-made, appropriate economic framework for analysing a fair cost/price spread on the market in question, particularly in a sector like the pharma industry, where monopoly profits are arguably needed to balance out expensive R&D. The second limb is no easier to apply, since it requires a benchmark comparator to be identified and justified, when e.g. like-for-like comparisons may be tricky in any sector between market-leading products and smaller new entrants. 

The challenges in applying the standard legal test for excessive pricing have previously led to a cautious approach from competition law regulators.  That seems to have changed, at least in the pharma sector, due above all to a number of widely publicised sharp increases in the prices of certain medicines where the size of the increase has created a more easily identifiable target for enforcement.  The prominence of these cases, which usually concern generic medicines with a captive patient population, can be explained in part by a broader societal context of public healthcare budget constraints.  In the UK specifically, there has also been a perception that price regulation has failed due to a “loophole” in the rules which allowed certain generic medicines to escape regulatory scrutiny.  The Health Service Medical Supplies (Costs) Act 2017, which received Royal Assent on 27 April 2017, closes this loophole by, in effect, allowing the government to regulate the price of any individual medicine not covered by an existing pricing scheme, yet there has been a certain amount of media and political pressure for competition law to step in and remedy this perceived regulatory failing.  The full decision in Pfizer/Flynn, for instance, explicitly acknowledges that the CMA’s investigation began as a result of approaches from the Department of Health.

What are the key points in the Pfizer/Flynn case?

A summary of the background facts

Before September 2012, the price of Epanutin (the previous brand name for  phenytoin capsules) was regulated under the Pharmaceutical Price Regulation Scheme (PPRS). In 2012, Pfizer and Flynn entered into agreements transferring Pfizer's Marketing Authorisations for Epanutin to Flynn for a nominal fee, while Pfizer continued to manufacture the capsules and exclusively supply them to Flynn. Following the transfer, Flynn genericised Epanutin and the product was withdrawn from the PPRS, meaning it was no longer subject to price regulation.

The CMA found that, while phenytoin sodium capsule prices had been broadly stable before that date, from September 2012 the average selling price for 100mg capsules rose from £2.21 to £51 to £61 in March 2014. Between March 2014 and June 2016, the average price was £41 to £51.

A “very narrow” market definition

The CMA identified the following relevant markets as the basis for its conclusion that the parties were dominant:

  • the manufacture of Pfizer-manufactured phenytoin sodium capsules that are distributed in the UK; and
  • the distribution of Pfizer-manufactured phenytoin sodium capsules in the UK.

The CMA’s decision openly acknowledges that “this represents a very narrow product market”, its justification for which is twofold.  First, it argued that the market could not be broader for clinical reasons.  Phenytoin sodium has “a narrow therapeutic index", meaning that even small changes to dosage can give rise to a disproportionate change in the level of the drug in the body, which can result in therapeutic failure and even toxicity. Clinical guidance had recommended that patients who are stabilised on a particular manufacturer’s phenytoin sodium capsule should be maintained on that product and not be switched to another manufacturer’s capsule.  This clinical background underpinned the CMA’s narrow finding that there were no substitutes, but its second justification was that there had been limited relevant actual switching to a variety of other existing options which might otherwise have seemed alternatives (such as the other existing capsule product or tablet products). Interestingly, there was substantial switching to the competitor capsule product when it launched, but that switching stopped on issuance of the MHRA guidance recommending continuity of supply on existing medication.

The market was therefore defined at a level even narrower than the molecule  – in two ways: by excluding other formulations (tablets) and by excluding third party products and thereby making the market brand-specific.  This is highly unusual.  It was, however, clearly very much based on a specific set of facts, and in particular on (i) the fact that there were very few new patients to compete for, so that the key issue was about switching; and (ii) the high toxicity risks, and strong guidance against switching – guidance which the evidence suggested was followed. These factors are also highly relevant to the way the abuse analysis was approached, as will be seen below.

How were Pfizer’s and Flynn’s prices found to be “excessive” under the abuse rules?

As summarised above, the legal test under the abuse of dominance rules requires both (i) the price charged to be excessive when compared with the cost and (ii) the price to be unfair either in itself or when compared to competing products.  Since the case was run at both manufacturing and distribution levels (with mirrored markets), the assessment was carried out separately for each of Pfizer and Flynn, but there were many common features.

Prices excessive.  Under the first limb, the CMA found that both Pfizer and Flynn's prices materially exceeded their costs, plus a reasonable rate of return. The CMA examined the parties' costs and prices, and found the excess cost for the various strengths ranged from 30% to 705% for Pfizer and from 30% to 133% for Flynn. Key elements of the CMA’s assessment were:

  • The first step, based on past cases, was to try to get to some kind of “cost plus” rate of return.  As always in profitability analysis, a key question here was ‘rate of return on what?’ For Pfizer, the CMA indicated it would have preferred to use ‘return on capital employed’ (ROCE) on the basis that this was a measure used in pharma regulation (under the PPRS – see below) and the alternative of return on sales (ROS) was not appropriate given that sales revenue was based on the very prices which were the subject of investigation.  By contrast for Flynn, ROS was considered suitable, in part because Flynn did not employ significant assets.  Ultimately ROS had to be used for Pfizer as well because of the difficulties of measuring capital employed due to the lack of a dedicated production line.   These technical issues as to which measure, and the limitations on the available data, can be significant in these cases.
  • The second question was what is a reasonable rate of return. And here, the CMA started off with very different approaches for each of Pfizer and Flynn, but driven by common features and ultimately settling on a common benchmark.  For Pfizer,  the starting point was that a reasonable ROS should not be higher than the ROS earned over the rest of Pfizer’s business.  The rationale was that this was a very old drug with no recent innovation and low risks due to an established customer base.  So the CMA sought a benchmark which would not leave Pfizer materially worse or better off than it would be without this business.  For Flynn, those same factors together with Flynn’s limited activities in distribution and the function of its high supply price in inflating its ‘cost plus’ figures led  the CMA to reject an approach based on the ROS across the rest of its business.
  • Yet for both businesses the CMA settled on a benchmark (as a back-up argument for Pfizer and more central argument for Flynn) based on  the 6% target for a return on overall (cross product) sales found in the PPRS 2014.  The PPRS is a voluntary agreement between the branded pharma sector and the government whereby, among other elements, a profit cap on sales of branded medicines for scheme members is agreed.  The PPRS profit mechanisms did not apply to phenytoin capsules and the CMA acknowledged that, since the PPRS related to a portfolio of products, its 6% return-on-sales benchmark had limited value as an indicator of a reasonable rate of return for a single generic product.  Nonetheless, the CMA did finally choose to apply 6% as a benchmark for this very purpose because it “is the closest the UK comes to an agreed industry standard for returns on pharmaceutical products”.  This is one of the most controversial elements of the decision, given that the PPRS allows companies to “modulate” by increasing the price on some drugs and reducing the price on others – and typically companies will reduce prices on products which face greater competition. Many in the pharma sector and beyond may feel that the CMA has extracted tenuous probative value from a general regulatory cost target benchmark.
  • The third and final step was to look at the ‘excess’ actual over reasonable returns. And the relevant excesses (cited above) were far higher than those found in previous abuse of dominance cases (25% to 47%).

Prices unfair. On the second limb, the CMA decided that the parties’ pricing was unfair in itself and that it did not need to show that it was unfair when compared to other products.  In summary, its reasoning was as follows:

  • There were no non-cost factors that increased the economic value of the products. The CMA rejected a series of arguments based on the value of continuity of supply to the NHS and the increased value from the MHRA guidance itself in effectively limiting the substitutes.  The CMA’s analysis here was essentially ends-driven: these were factors which enabled the excessive pricing and therefore could not, in its view, justify it.
  • The substantial disparity in the economic value and cost made the prices unfair in themselves.
  • The prices could only be maintained as the companies were shielded from competition.
  • The excessive prices had had adverse effects on the NHS and its resources, which had to be diverted from elsewhere to cover the cost of the price increase.
  • Finally, there was an obvious comparator in that the prices were increased from a lower base, i.e. following the transfer of the marketing authorisations to Flynn in 2012.  In the CMA’s view, even with Pfizer claiming the product was loss-making at the earlier price levels, the price increases went “far beyond” what was necessary to make it profitable.

Concluding comments

The point just made in a sense goes to the heart of this case.  The “sheer scale” of the price increases – made possible from a regulatory perspective through the de-branding – was clearly what prompted the CMA’s intervention, and is a central theme running through the entire decision.  Similarly, the CMA returns frequently to the age of the products, the lack of recent innovation, the stable customer base and guidance on continuity of treatment, and for Flynn the limited extent of its activities.  In many ways the decision is a short story masquerading as a novel: these core features drive the CMA’s conclusion that the returns are excessive; the conclusion is evident at the start, and the legal hoops are just that – legal hoops that have to be gone through, into which the same arguments are repeatedly inserted.

On balance, the significance of the Pfizer/Flynn case boils down to three key areas:

  • In competition law terms, the case is an unusually strict application of a previously underused category of infringement.  Applied literally, some of the economic benchmarks the CMA applies to determine what is “excessive” pricing could catch (and outlaw) high pricing of products in the pharma and also in other sectors which involve premium pricing.  Companies, however, should be particularly wary of high price increases from a low base for dominant products, since, based on this decision, such price increases of themselves may be interpreted as a self-evident indication of an excessive price.
  • The case is also an interesting example of competition law overlapping with sectoral price regulation.  The CMA’s decision quotes extensively from parliamentary debates of the new costs legislation and the CMA’s case reinforces the regulatory environment.  One of the interesting points here is the remedy.  The decision directs Pfizer to reduce its prices, and Flynn in turn to reduce its own prices based on the revised input prices.  But it does not prescribe specific levels.  This contrasts with the Napp, where a price relationship between two segments was specifically ordered.  The new legislative framework reinforces the government’s regulatory powers to mandate specific price levels for drugs and other products sold to the NHS, although  how that will be developed depends in part on detailed implementing regulations, which are expected imminently.
  • The open question is where the enforcement of excessive pricing could develop in the future.  Much will depend on the outcome of Pfizer and Flynn’s appeal of this case, with the hearings timetabled to last for the whole of November 2017.  But it is clear as a matter of law that the prospective existence of specific sectoral pricing regulation for unbranded medicines will not preclude the application of competition law.  There is therefore a realistic prospect of emboldened enforcement agencies investigating other cases of high pricing in the pharma and potentially other sectors.