On excessive pricing in the pharmaceuticals sector.

Recent EU and UK public enforcement decisions have put the pharmaceutical sector on notice for excessive pricing enforcement. From these decisions it is clear that, regarding long off-patent drugs, dominant undertakings should beware of setting prices at levels that significantly exceed reasonable profitability-based benchmarks, and that large [and opportunistic] price increases that are not associated with increases in costs, innovations or product improvements will not be tolerated.

According to EU and UK competition law, abuse of a dominant position can be divided into categories of exclusionary abuse and exploitative abuse. While the former category refers to the conduct that is aimed at excluding or weakening present competitors or potential entrants, the latter refers to a "purer" abuse directed at consumers or customers in the form of excessive prices. Although the enforcement has predominantly been on exclusionary abuses, recently there has been a surge of EU and the UK enforcement decisions based [primarily] on exploitative conduct in the pharmaceutical sector. Some of the notable decisions that we will discuss here are:

• Leadiant Biosciences was fined for excessive pricing in Spain, Italy and the Netherlands [ "Leadiant" ],¹
• Advanz Pharma was fined for excessive pricing in the UK [ "Liothyronine" ],²
• Auden / Actavis was fined for excessive pricing in the UK [ "Hydrocortisone" ],³
• The European Commission ["EC"] accepted commitments by Aspen in relation to the excessive pricing of multiple drugs [ "Aspen" ].⁴

These cases are all centred around drugs that have been off-patent for decades, and have experienced dramatic price increases, without any notable product improvements that could potentially legitimise such price increases. In addition, these cases involve:

• gaining an orphan designation for market exclusivity [high barriers to entry], in the case of Leadiant and Hydrocortisone, and
• strategic de-branding of branded drugs to escape from regulatory or health agency oversight, in the case of Liothyronine and Hydrocortisone.

In this paper, we will analyse the aforementioned decisions through an economist's lens and contrast the approaches employed by the EC and the national competition authorities.

The overall structure of this Insight will be as follows. We will begin by providing some implications of the recent decisions for dominant undertakings, and discuss the possibility of (follow-on) damage claims in light of these decisions. After that we will discuss the influential United Brands standard. Finally, we will provide a detailed analysis of the recent decisions with respect to market definition, and the employed methodologies in relation to the United Brands standard.

The recent decisions employ robust cost-plus and investor return methodologies. Although there are differences in the finer details of the analysis, there seems to be an overall consensus that the economic value (and hence the price) of long off-patent drugs should not be significantly different from their price benchmark, based on a reasonable return and/or profit.

Accordingly, these decisions suggest that dominant undertakings with off-patent drugs, should:

a) be able to justify [large] price increases in relation to costs,
b) consider that being outside of regulatory oversight or being in an exclusivity regime does not give free rein to pricing decisions,
c) consider if their products give a reasonable return or profit.

Below, we will elaborate on some of the challenges these responsibilities may bring upon dominant undertakings and some of the associated uncertainties.

First, the decisions could be read to suggest that whenever a dominant undertaking becomes significantly more efficient, it may need to reduce its price based on its decreased product costs.

Second, it seems that the undertakings need to keep an eye on the industry dynamics and the overall profitability of comparable drugs to get a sense of what reasonable profit/return would entail.

Third, it may be desirable for dominant undertakings to have a good grasp of the intricate risks involved within a sector, such as risks that may be associated with the emergence of a superior medicine or the possibility of a failure of a project. This assessment of the risk is important when calculating the company’s WACC, and when assessing the IRR of a specific project. Firms that misestimate inherent risks may calculate improper reasonable return benchmarks and may either end-up making less than reasonable return/profit or in contrast they may make excessive profits that could take the attention of competition authorities.

Fourth, the case-law does not yet make it clear what the lower boundary is at which prices become excessive. Even if one abstracts from the unfairness limb, non-cost factors and other comparators, the excessiveness limb does not provide a precise formula. The profitability benchmarks that are used in the excessiveness limb may diverge from each other, and it is not immediately clear what magnitude of deviation from the reasonable benchmark can be considered excessive.

Considering the decisions discussed in this Insight, the lowest measure of excessiveness is found in Leadiant in Italy. The Italian authority found that the price was excessive when it was 60%-100% above the cost-plus benchmark price, suggesting that a reasonable threshold lies somewhere below a price that is 1.6 times higher than the cost-plus benchmark price.

Furthermore, it is useful to note that relatively recent judgment of AKKA / LAA indicated that only prices that are “significantly and persistently” excessive can amount to be an abuse and that in some of the aforementioned decisions, the authorities based their excessiveness assessments on the prices or margins/returns being “significantly” higher than respective reasonable benchmarks.⁵

In Aspen, the European commission used this terminology and attempted to clarify what significant deviation from a given benchmark could mean. The EC, while accepting that what significant entails may be context-specific, seemed to suggest a distributional approach to detect significant deviations. In assessing the magnitude of excessiveness of the prices, it considered the distribution of comparator products’ margins and noted that the EBITDA margins of the undertaking lay significantly above the 90-percentile of comparators, whereas gross margins were slightly higher than the respective 90-percentile.⁶ This implies that the price is only considered excessive if the resulting margin is significantly above that of the highest comparator margin. It remains to be seen whether the case-law provides more clarity on a threshold point for excessiveness and whether the distributional approach would be widely embraced.

Fifth, in Liothyronine and Hydrocortisone, the CMA rejected the use of a portfolio pricing approach citing the CAT judgment of Napp. However, economic theory suggests that a portfolio pricing approach might be relevant in some cases where certain cost related decisions concern multiple products and firms face real uncertainties about the success rate of its individual products. It remains to be seen how the authorities and the courts of the EU would approach to the portfolio pricing approach - would they outright the reject the approach or would they embrace it in some situations?

The aforementioned public enforcement decisions punish or correct abusive conduct of dominant undertakings and try to deter such behaviour from happening in the future. These decisions also open up the possibility for follow-on and stand-alone claims on behalf of damaged parties. Parties who might have been damaged as a result of the aforementioned conduct, except for the conduct in Aspen as this was a commitment decision, may claim compensation, without having to prove the existence of an infringement.

Depending on the national healthcare system, the excessive pricing conduct might have resulted in damages with national healthcare agencies, private or public healthcare insurers, hospitals, pharmacies, taxpayers, or the patients / consumers themselves.

Potential claims would be based on a counterfactual price to estimate the damages incurred. To arrive at a counterfactual price, the particulars of the case need to considered and various benchmarks can be taken into account. Potential counterfactual prices, among other things, could be:

a) A historical non-abusive price point, potentially corrected for cost increases.
b) The price of a comparable product.
c) The [average] price resulting from the cost-plus analysis in the infringement decision.
d) Prices of the same product in another geographical market.

The courts may decide on a single benchmark as the counterfactual price or take the average (or maximum) of various viable benchmarks. Alternatively, the court could add a small but significant differential on the used benchmark.

Another important factor to consider is the extent of pass-on. Depending on the health care system in which the infringement occurred, the excessive price may have been paid by different institutions But, if those institutions have then recovered their costs by increasing premia, increasing patient contributions, applying for extra government funding, etc., the damages could have been fully or partially passed-on to other actors in the health care system.

The influential United Brands standard on excessive pricing has given shape to the recent enforcement decisions in various ways. In United Brands, the ECJ ruled that if a dominant undertaking reaps benefits that result from its dominant position and charges excessive prices that bear no reasonable relation to the economic value of the product, then it would constitute an abuse.⁷ This implies that dominant undertakings need to act against their short-run (rational) instincts of reaping benefits and, to some extent, self-regulate, i.e. set prices as-if there was "normal and sufficiently effective" competition.⁸

United Brands also sets forward an influential two-stage criterion to determine whether prices constitute an exploitative abuse.⁹ The two-stage criterion is cumulative and consists of an excessiveness limb and an unfairness limb. While the first stage [excessiveness limb] evaluates whether prices are excessive in relation to costs, the second stage [unfairness limb] considers whether prices are unfair in themselves or in comparison to competing products.

The recent enforcement decisions on excessive pricing use methods based on the United Brands two-stage criterion, albeit with different cost-excess methodologies [excessiveness limb] and varying interpretations of the relevance of comparator benchmarks [unfairness limb].

In our analysis, we will describe the methodologies the authorities employed in relation to each of the two limbs. However, this does not mean that certain facts or analysis cannot be relevant for both limbs at the same time. For example, excessiveness can be investigated using the reasonable return for comparable products, while [un]fairness of prices can be assessed considering the additional innovativeness which could be also relevant for the cost-plus calculation.

Before analysing the decisions in detail, it is useful to briefly explain the approaches that the authorities used in determining the existence of market dominance.

In all aforementioned cases, the national authorities and the EC employed narrow market definitions. Markets were defined on the national level, and constituted only the molecular substance, or even narrower, only the specific production method of the molecular substance. Dominance of the respective undertakings was determined by mainly relying upon the very high market shares (>60%) on the respective markets.¹⁰

It is important to note that the authorities based their narrow market definitions on the observation that other competitor drugs or competitor products manufactured by alternative methods did not constitute significant competitive constraints even though some of these competitor drugs might have been, to some extent, therapeutic substitutes.¹¹

Furthermore, some authorities cited additional factors such as market exclusivity, exclusivity agreements with suppliers, and the lack of countervailing buyer power to have contributed to the dominant position of the undertakings.¹²

Through the excessiveness limb of the United Brands standard, competition authorities attempt to assess whether the profits or returns of a firm are excessive compared to the costs and expenses related to the product.

In effective competition, firms may be expected to earn a reasonable level of profit or return on their costs or investments. Without such profits, firms may not enter the market, or may not make the necessary investments to continue or improve production. In a market where one firm is dominant, the firm may leverage its market power to earn above-reasonable or excessive profits, by charging excessive prices.

The threshold of what level of profits (and prices) can still be considered reasonable is context dependent and there are multiple ways to measure profitability. In any event, an economically sound profitability measure would consider direct and indirect costs that can be associated with the production of the product and should take into account the inherent risks and opportunity costs of such activity and industry-specific intricacies.

In the context of excessive pricing, competition authorities can use various methods to assess the profitability of firms, and the excessiveness of their prices. These methods can be divided into two categories:

• Cost-Plus: Authorities calculate a cost-plus benchmark price for the product, which can be derived from adding a reasonable gross margin, return on sales [ROS] or a reasonable return on capital employed [ROCE] multiplied with an appropriate capital base measure to the [per-unit] cost. This benchmark is then compared to the realised price. The authority will then determine if the realised price is unreasonably or excessively above the benchmark,
• Reasonable Return: Authorities will investigate what level of internal rate of return (IRR) is
  to serve as a reasonable benchmark. This benchmark is then compared to the IRR realised on the cash-flow of the product-project. The authority will then determine whether the realised IRR is unreasonable or excessively above the benchmark, from the viewpoint of an [ex-ante] investor.

In the following subsections, we will explain what approaches the authorities used in relation to this limb in the cases at hand. In Table 4–1 below we summarise the findings of the different decisions. It is apparent from the table that most decisions employ more than one method to determine the excessiveness of the prices or profits.

The estimation results are typically a range of values. This stems from the fact that either the relevant price information is confidential, or from the fact that some authorities conducted multiple calculations with slight variations. Such variations are done to ensure the robustness of the estimate. The authority in question does not wish to estimate a precise figure but give a reliable indication that the prices and/or profits were indeed excessive.

The lowest degree of excessiveness found across the different cases is 60%-70% over the cost-plus benchmark price for Leadiant in Italy, while the highest degree is for Liothyronine in the United Kingdom with 2433% over the cost-plus benchmark price.

The percentages in the table are calculated as one would an overcharge in cartel damages, the percentage that the actual price or return is larger than the relevant benchmark. The 60%-70% excess of Leadiant in Italy should be interpreted as the realised prices being 60%-70% above the benchmark price. Put in other words, the Italian Leadiant prices are 1.6 to 1.7 times higher than the benchmark price.

Table 4-1. Price excessiveness methodologies and results.

Source: Respective decisions. Excess over cost-plus is calculated by dividing the excess by cost-plus. Excess over internal return is calculated by dividing the excess by the reasonable return.¹³

Leadiant

Leadiant is a case characterized by sharp price increases subsequent to or just before the firm attained an orphan designation and market exclusivity from the European Medicines Agency (“EMA”). The competition authorities of Spain [“the CNMC”] and Italy [“the AGCM”] employed both the cost-plus and IRR analysis method, while it appears that only latter method was used by the Dutch competition authority [“the ACM”]¹⁴. We start by explaining the common IRR analysis.

The CNMC, the AGCM and the ACM, each calculated the internal rate of return [IRR] that resulted from the orphan designation application. Specifically, the CNMC, the AGCM used the realised and expected cash flows in the market exclusivity period which ends at 2027.¹⁵ The authorities then compared the estimated IRR to the weighted average cost of capital [WACC]. In this context the WACC serves as the rate of reasonable return that investors would require to invest in the company.¹⁶ All authorities used a “conservative” WACC of 15%, and found that the project yielded excessive returns compared to this benchmark, indicating excessive prices.¹⁷ Specifically, the Italian authority noted that the IRR was at-least 150%-200% higher than the WACC, whereas the Spanish authority found that the IRR can be infinite and is 100%-110% according to the most conservative calculation, or 567%-633% above the benchmark.¹⁸

In general, the IRR analyses were conducted after most of the cash-flows were realized and the success of the project was already largely determined. Therefore, one could adjust the cost estimates or the WACC to realistically account for the project-specific ex-ante risks that investors might have faced.¹⁹ To this end, the ACM, in its IRR analysis, took into account the possibility that the orphan designation application would fail, when calculating the IRR of the project. However, the Spanish and Italian authorities considered that the of cost of capital already accounted for the low risks associated with the project.²⁰

The AGCM and CNMC also conducted cost-plus analyses. Both authorities added a reasonable ROS to the costs associated with the product. The cost-base that the authorities considered, accounted for the costs that were directly related to the product, and indirect costs that were estimated by taking an appropriate percentage of shared costs that the undertaking incurred for its whole portfolio. The AGCM and CNM estimated the reasonable ROS benchmark by using the average operating margins of a comparable set of pharmaceutical companies in the EU in the relevant time period, which were given at 21% and 22%, respectively.²¹ Using these cost-bases and profitability benchmarks, the Spanish authority found that the cost-plus excess amounted to 146%- 223% [depending on the variation of the relevant period], whereas the Italian authority obtained an excess figures of 60%-70% and 90%-100%. As a result, both authorities find that Leadiant’s prices were excessive.²²

Liothyronine and Hydrocortisone

The Competition & Markets Authority decision [“the CMA”] issued two almost simultaneous decisions in the cases Liothyronine and Hydrocortisone. Both cases involved the de-branding of drugs and sharp price increases. The CMA used the cost-plus method to assess the excessiveness of the prices. Since the approaches that the CMA employed in both cases are slightly differed, we will explain them separately.

In Liothyronine, the authority calculated the cost-plus benchmark price by using the relevant direct and indirect costs and a reasonable ROCE multiplied by the employed capital base of the company. The capital-base in this calculation consists of the product-attributable tangible and intangible assets, and working capital. The authority considered that a reasonable ROCE rate is a reasonable WACC of 10%. To obtain this estimate for the WACC, the authority used a reasonable cost of debt and a cost of equity that was calculated using the capital asset pricing model [CAPM] formula and an associated risk estimate which was derived from the average volatility of a set of comparable drugs.²³ The resulting analysis of the CMA showed that the differential between the realized price and cost-plus price ranged between 900% and 2433% in the relevant period with a weighted average of 1894%, indicating excessive prices.²⁴

Upon receiving objections from the parties, the CMA also conducted some sensitivity checks based on:

a) A higher cost of capital of 15% [excluding the working capital],
b) A larger capital-base to incorporate the possibility that the product may fail,
c) A different method of allocating shared costs.

These sensitivity checks increased the cost-plus price but still indicated an excess in the range of 321%-1985%.²⁵

The CMA rejected the other adjustments put forward by the parties, such as the inclusion of a positive small company premium for the cost of capital estimation, incorporation of portfolio-pricing, and multi-firm approaches that would have increased the cost-plus price.²⁶

In Hydrocortisone, the CMA calculated the cost-plus price by calculating the direct and indirect costs and adding a reasonable ROCE rate of 5-10%, multiplied by the relevant capital base of the undertaking. The CMA concluded that this ROCE rate was appropriate as it was equal to the WACC estimate that the undertaking used in its valuation of the company before acquiring it.²⁷ The CMA calculated that the average excess for 10 mg tablets amounted to be 879% whereas the average excess for 20 mg tablets was 702%.²⁸

The CMA also received some objections to its calculations. These objections were centred around the non-appropriateness of using a single-product cost-plus analysis, the non-appropriateness of using ROCE for an allegedly “asset-light” business, and the low [reasonable] rate of return on capital used in the cost-plus. The CMA rejected these adjustments.²⁹

Aspen

The case of Aspen originates from the Italian competition authority decision in which it had fined Aspen for excessive pricing of several off-patent cancer drugs. The European Commission followed up to investigate the same infringement for the whole EEA, excluding Italy, and issued a commitment decision. The EC conducted a cost-plus analysis by adding a reasonable gross, and earnings before interest, taxes, depreciation, and amortization [EBITDA] margins to the relevant cost-base.³⁰ The EC relied on the gross and EBITDA average margins of comparable products, which were 54% and 23% respectively.³¹ The EC found that the prices of Aspen’s drugs are 280-300% higher than the cost-plus price calculated with these margins, indicating that the prices of Aspen were excessive.³²

In calculating the cost-base of Aspen, the EC did not take into account the acquiring costs that Aspen incurred when taking over the drugs from another firm. The EC justified this omission by noting the circularity problem that arises from this line of argument, as the acquiring price is [possibly] a function of expectations of future prices – which might have been anticompetitive.³³

Nevertheless, the EC conducted a complementary IRR analysis while taking into account the acquisition costs, and derived an IRR around 40%-50% which lies ‘well- above’ the ex-ante WACC estimate of Aspen which was given at 10%.³⁴ According to the EC, these additional analyses confirmed its cost-plus assessment that the prices were excessive.