In this book chapter, the authors criticise the CMA for relying on the same evidence of a gap between prices and costs in its assessment of each of market definition, dominance and abuse. When coupled with the absence of analysis of comparator prices – which, the authors argue, the CMA replaced with a failed search for justifications for a price-cost gap when finding that the price was ‘unfair in itself’ – this could serve as a precedent for a fragile and unreliable approach to assessing excessive pricing.
The paper is structured as follows:
Section 2 describes the framework for assessing excessive pricing under European law (and its British equivalent).
The paper builds on United Brands‘ two-step test, and particularly the requirement that am excessive price must exceed the “economic value” of the product to such an extent that the price bears “no reasonable relation” with that value. The legal test set out by the ECJ is as follows. First, the test for determining whether a price is excessive consists of comparing the price with all the “costs actually incurred” by the firm, including (i) the costs directly incurred in supplying the product or service, (ii) an “apportionment of indirect costs and general expenditure”, as well as (iii) a “reasonable return”. Together, these elements are commonly referred to as the “Cost Plus” test.
Where prices are in excess of Cost Plus, competition authorities should proceed to the second limb of the test, i.e. whether the price is unfair. There are two alternative approaches to this second limb. Firstly, competition authorities may show that a price is unfairly high by reference to comparable products. Secondly, competition authorities may show that a price is unfair “in itself”. This requires taking into account non-cost related factors, especially as regards the demand-side aspects of the product/service concerned.
While these approaches are alternatives, the authors consider that it would be perverse to rely on unfairness ‘in itself” when there are available comparators. Furthermore, if the purpose of an excessive pricing action is to prosecute prices that are so high as to be unrelated with the “economic value” of a product, then the threshold for determining whether a price is excessive needs to be set well above the price that prevails in a reasonably competitive market.
Section 3 sets out the facts of the Pfizer Flynn case.
These facts were already outlined in an earlier post, so I am not going to repeat them here. The authors do, however, provide an interesting overview of the regulatory framework in the UK, which is always a relevant consideration in pharma cases.
Section 4 assesses the CMA’s decision. This section comprises the bulk of the paper.
As regards market definition, the paper notes that the CMA built on the SSNIP test but observed that existing prices are already well in excess of 5– 10% above competitive levels. It therefore concluded that nothing constrained those prices from increasing above the competitive level.
The authors criticise this line of reasoning as circular. It will lead, in any case where prices are found to be at least 10% above competitive levels, as they will be in any case of excessive pricing, to the market always being defined to include only the products of the dominant company. Further, this approach fails to recognise that the competitive price level need not be cost-reflective. There is a problem in using ‘competitive’ prices to define markets when those ‘competitive’ prices emerge from price-cost evidence. While the CMA rightly did not base its market definition solely on price-cost evidence, its repeated reliance on cost data at different stages of the analysis risks setting a precedent in which all of those stages collapse into one observation regarding how price relates to cost.
The analysis of dominance is said to suffer from the same problem. It is true that the CMA dealt with barriers to entry and with insufficient countervailing buyer power. However the same price-cost evidence was cited as part of the dominance assessment as had been cited for defining the market.
As regards excessive pricing, the authors deal with each of the limbs of the United Brands test outlined above.
As regards whether the price was excessive in this case, the authors begin from the observation that firms must cover all of their costs, including indirect costs and returns to investors or lenders to the business. In pharma, indirect costs – namely costs with other products, and particularly with the research of products that never reach the market – are quite high. The CMA therefore sought to include an allowance for such costs, in effect creating a unit cost measure equal to long-run average cost. This is criticised because it attributes cost to higher margin and lower margin products equally, and thus is bound to find that the higher margin products under investigation have price-cost margins well above the firm’s average, and most likely above the market average as well.
As regards the reasonableness of the rate of return, the rate set by the CMA strikes the authors as being too low. In particular, by adopting the PPRS measure (i.e. 6%) the CMA applied portfolio profit rates to a single product. However, the CMA is not in the position of a regulator of a natural monopoly, who can typically correct for under- as well as over-recovery. If a company’s average return on sales is 6%, then some products will earn more than that while others will earn less. If 6% is a cap on any one product’s return, then the average return will necessarily be less than 6%.
It is true that taking this approach to the average rate of return creates a risk that ‘the price of a product sold by a sufficiently large firm could never be found to be excessive, no matter how high it was, because it would always be a drop in the ocean’. Nonetheless, they suggest that competition authorities should always: (i) consider the profitability of prices of a ‘stand-alone’ firm only supplying the products in question, and assess the sustainability of such a firm; and (ii) avoid relying on calculations identifying excessive prices by reference to costs as regards a single product when defining the market or determining whether a company is dominant. Instead, a competition agency should treat each of these steps as independent filters, in order to ensure that cost allocation does not determine the whole case.
As regards whether the price is unfair, the CMA opted to focus almost exclusively on whether the price was ‘unfair in itself’. The authors strongly object to this. Firstly, the only meaningful benchmark for ‘economic value’ is the price of a similar product in a reasonably competitive market, so the ‘comparator’ version of this part of the test has a compelling logic in economic theory. Secondly, this is particularly the case if the alternative is for the CMA to fall back on the same price-cost analysis that led it to find the price to be excessive in the first place.
The result is that the CMA presumed that the price was unfair, absent some justification for the price. This reverses the burden of proof – if not in a legal, at least in a policy sense – because it presumes that companies need to explain to the regulator precisely why their prices are higher than cost. Most suppliers have no reason to consider whether there are ‘factors’ that justify their prices being above cost, and competition authorities should not require them to produce such explanations unless the authority can show that the prices exceed relevant comparators or are unfair in themselves.
The authors further consider that the CMA’s refusal to engage with comparators is puzzling. The CMA claimed that, in this case, there simply were no comparators available – but the economics of producing generic medicines can be similar for different products, because production costs are often a small part of the total cost of the supply chain. Consequently, the price of a similar capsule that has an entirely unrelated clinical use might be of interest, if such a product can be found priced under conditions of competition.
Section 5 looks at remedies.
Firstly, the CMA did not identify the level at which a price would not be excessive, which would seem to create legal uncertainty. Secondly, this was the largest fine ever imposed in UK’s competition law history. The authors, however, consider that fines are not a suitable penalty for excessive pricing cases. How likely is it that a penalty for excessive pricing will deter similar abusive behaviour in the future, and how great is the danger that effective deterrence of abusively high prices will also deter normal and pro-competitive business behaviour? Deterrence is likely to be most effective when the abusive behaviour is clearly distinct from normal business behaviour – which is not the case with excessive pricing.
Instead, the CMA could follow one of three approaches: (i) follow the lead created by the CMA’s own market studies. In almost all market investigations leading to remedies, the CMA and its predecessors determined that prices were excessive. As in those market studies, there seems to have been a possible regulatory reaction in this case, particularly as it could be said that there was a regulatory gap created by a failure to act on the part of the Department of Health. Thus, a recommendation for legislative reform to close that gap would have been a possible alternative to a fine in this case.
An addendum reflects the Competition Appeal’s Tribunal judgment on appeal from the CMA decision. ‘Although [the] focus in this piece was on policy rather than legal matters, the CAT judgment is highly relevant to several of the issues [discussed] here. The CAT rejected the grounds of appeal relating to market definition and dominance, citing the broad range of evidence in support of each and especially the clinical advice for continuity of supply. However, the CAT upheld the appeals against the CMA’s findings of abuse. [Regarding excessive prices], It found that the CMA had relied upon too narrow an evidence base in assessing prices against costs. However, its strongest criticisms appeared to be directed at the CMA’s approach to [unfairness], where the CAT criticised the CMA’s reliance on an ‘in itself’ assessment of unfairness (…) this test is an alternative to assessing unfairness against the price of competing products [which] does not allow the CMA to ignore arguments and evidence relating to comparators (…). The CAT also criticised the CMA for restating its price-cost finding as evidence for the assessment of unfairness.]
I am sympathetic to the authors’ argument that excessive pricing should be subject to a test with several checks and balances. As such, I agree that such a test should avoid ‘low evidential thresholds’ because this could lead to a great many markets in which prices could be found to be ‘excessive’, apparently enabling widespread regulatory intervention.
I am not sure that such risks are inherent to pure price-cost tests, though. After all, these tests could be subject to stricter evidentiary standards by coupling them with additional economic and non-economic criteria. In any event, I agree it is important not to reduce these factors to a single amorphous test, such as by collapsing ‘the two limbs of the excessive pricing test into one’.