Economics Insights articles are informal think-pieces by Ofcom economists. The analyses and opinions in these articles represent the views of the author and should not be interpreted as official Ofcom policy. This article was written by Arina Nikandrova from our economists team, before she recently left Ofcom for a new role.
In November 2019 the Competition Appeal Tribunal (CAT) upheld Ofcom’s decision to fine Royal Mail £50 million for breaching competition law. The original penalty was levied following an investigation into a complaint by Whistl, Royal Mail’s only competitor in the market for delivery of business letters, known as ‘bulk mail’. Ofcom found that Royal Mail had abused its dominant position by issuing in 2014 contractual notices to change its wholesale prices for other operators to access its delivery network which discriminated against its competitors that also undertook deliveries of bulk mail. The CAT upheld in full Ofcom’s Decision.
Before the CAT, a central issue was the extent to which competition law prevents a dominant firm from introducing pricing structures that might exclude competitors less efficient than itself. This Economics Insights piece briefly outlines the economics of this aspect of the case, discusses the issues that were at play and the CAT’s reasoning. There are a number of other important issues in this case that are likely to be of interest to competition economists, lawyers and policy makers – for instance, the implications of the non-implementation of the pricing once Ofcom opened an investigation, the status of the EU case law on pricing cases, and the application of the exemption for ‘Services of General Economic Interest’ in regulated markets. This short note does not attempt to cover all of those, but you can read the full judgment for more detail.
An appeal by Royal Mail to the Court of Appeal will be heard in April 2021. In the meantime, the case continues to generate great interest in the economics and law competition policy community.
Royal Mail’s price changes
In January 2014, Royal Mail issued contractual notices which introduced differential prices for final delivery of bulk mail collected and pre-sorted by other postal operators, known as ‘access operators’. Royal Mail’s notified access pricing changes introduced, for the first time, a differential between the different price plans available to its access customers. Although the exact details of the price plans, terms and conditions are extremely complicated, it is sufficient for present purposes to note that the effect of the notified price differential would have been for access operators that also engaged in delivery of bulk mail (i.e. competitors of Royal Mail in the delivery portion of the value chain) to pay a higher price per item than those access operators that relied entirely on Royal Mail for final delivery.1
Whistl was the largest access operator in the UK, collecting and sorting bulk mail from its business customers and handing it over to Royal Mail for final delivery. At the time Whistl was also expanding a new aspect of its business; rolling out its own delivery network in parts of the UK to compete directly with Royal Mail. In the areas where Whistl had not rolled out its final delivery service and for customers who had not yet agreed to move mail volumes to Whistl’s delivery service, it had no choice but to rely on Royal Mail for delivery of its mail. The notified price differential meant that Whistl would have to pay Royal Mail the higher access prices across all of its mail that continued to be delivered by Royal Mail than it would have paid had it halted its competing delivery network and used Royal Mail for delivery across the whole of the UK.
Whistl complained to Ofcom that the new access prices made its planned final delivery operations unviable. In February 2014, Ofcom announced the decision to open an investigation into Whistl’s complaint. Although Royal Mail suspended the introduction of its new prices pending resolution of Ofcom’s investigation, Whistl put on hold its plans to extend delivery services to new areas. In June 2015, Whistl announced that it was ceasing its bulk mail delivery operation in its entirety.
The As Efficient Competitor test in exclusionary abuse cases
In principle, market exit is not necessarily symptomatic of abuse by a dominant firm. Indeed, a well-functioning competitive process is expected to result in exit of inefficient firms that are unable to make profit when faced with vigorous competition against more efficient rivals. Therefore, the task for Ofcom and, on appeal, the CAT was to determine whether Royal Mail’s price changes were “competition on the merits” or whether they amounted to abuse of a dominant position.
The CAT adjudicated Royal Mail’s appeal in light of EU case law which emphasises the requirement to undertake an effects analysis, principally the two Post Danmark cases and Intel.2 These cases were decided following the European Commission issuing its 2009 enforcement priorities guidance on abusive exclusionary conduct by dominant firms (“the EC Guidance”).3 The EC Guidance was a watershed moment in competition law thinking. It signalled the Commission’s commitment to an analysis of the economic effects of exclusionary practices, thus marking a move away from the prevailing “form-based” conduct rules under which certain dominant firm pricing practices are deemed to be per se illegal. For its effects-based approach, the EC Guidance extensively references the application of the as-efficient-competitor test (“the AECT”). However, the EC Guidance also states in paragraph 24 that “in certain circumstances a less efficient competitor may also exert a constraint [on a dominant undertaking].”
An AECT aims to determine whether the dominant firm’s pricing structure involves below-cost pricing and thus could drive an equally efficient competitor from the market. In the context of Royal Mail’s pricing, Royal Mail put forward an AECT that involved checking whether an operator with Royal Mail’s delivery costs could profitably operate in some areas while having to rely on Royal Mail for delivery in other areas.
The intellectual genesis of the AECT is in predation cases where below-cost pricing by the dominant firm in the short-term seemingly has all the characteristics of competitive conduct, but in the longer-term is capable of excluding equally efficient competitors from the market to the detriment of consumers. The EC Guidance, however, envisages the AECT may be an appropriate means of analysis across all the main categories of dominant firm exclusionary pricing conduct, such as predation, margin squeeze, tying and bundling, and loyalty discounts.
Principles for designing a test of anti-competitive foreclosure
Any rule for assessing whether conduct is anti-competitive is potentially subject to two types of errors: false positives and false negatives. False positives result in disallowing pro-competitive behaviours and depriving customers of the immediate benefit of lower prices; false negatives result in allowing anti-competitive behaviours to persist and reduce the likelihood of successful entry. At the policy design stage, the choice of rule for assessing pricing conduct should take into account the likelihood and the cost of these two types of errors. Furthermore, once adopted, the rule for assessing conduct would induce dominant firms to change their behaviour to comply with the rule. Hence, optimal enforcement policy should take into account the dynamic response of dominant firms to this policy and weigh the risk of curtailing desirable conducts against the risk of underdeterrence of undesirable conducts in future.
Proponents of consistent enforcement across all pricing conducts argue that an AECT strikes the right balance between protecting entrants from abuse of dominance and giving the dominant firm freedom to respond aggressively to entry. It is also argued that an AECT provides legal certainty to a dominant firm because, by using the dominant firm’s prices and costs, it allows “self-assessment” of conduct using information in the possession of the dominant firm, providing a “safe harbour” for the dominant firm if the test is passed.
An alternative viewpoint is that while the AECT might, in theory, provide a single rule for enforcement, in practice making an AECT determinative, and ignoring the strength of other evidence, is an overly narrow approach to effects-based decisions. Further, even if such tests should have a determinative role for predation, that role may be less appropriate for other forms of pricing conduct because the expected balance of false positives and false negatives will differ. It is argued that the consequent balance of risks of overdeterrence and underdeterrence depends crucially on the nature of the conduct and a case-by-case assessment of the particular facts of the case. Arguably, conduct that does not bring direct benefits to consumers (e.g. lower prices) but is likely to result in negative consequences for competition requires a less permissive anti-competitive foreclosure test with a low rate of false negatives. An AECT reduces any conduct to a single equation and so cannot capture the nuances of the conduct under consideration.
Whether an AECT was suitable in the Royal Mail case
From an economic perspective, it is my view that there are three particular reasons why, as a matter of principle, it can be argued the AECT was not informative in this case:
- The price differential did not bring a direct benefit to consumers. It was not an instance where Royal Mail cut its prices to compete with new entrants where a predation standard would be the natural test. Accordingly, the potential cost of a false positive as well as the risk of limiting desirable conduct by other firms in future was much lower than in a predation case.
- The cost of a potential false negative was high as Whistl was Royal Mail’s only significant competitor in the foreseeable future.
- Other evidence as to the intention behind the conduct and its expected impact was clear and compelling. This was a case where exclusion was clearly feasible and the consequence of successful exclusion would be a large loss of competition. This evidence meant that there was a low risk that this particular case was a false positive.
During the CAT hearing, Royal Mail presented the price differential as a discount on pricing conditional on access operators providing forecasting information which reflected demand for UK-wide delivery. Royal Mail argued that prices could have been higher in the absence of the price differential. Viewed as a conditional discount, it might be argued that there are potential consumer benefits and hence stronger reasons for looking at the conduct as a form of predation and for placing more reliance on an AECT.
Ofcom argued that the price differential could not be viewed as a discount, but rather that it represented a surcharge for competitors. This was reflected in Royal Mail’s internal documents, which showed that Royal Mail had rejected price changes involving price cuts and diluting revenues. Ofcom also found that the intention behind the price differential, which was discriminatory in its application, was effectively to penalise access operators who sought to compete with Royal Mail in bulk mail delivery. The CAT was persuaded by Ofcom’s arguments (see para 614-616 of the CAT judgment).
In my view , the stakes for competition were high when assessing Royal Mail’s conduct. Whistl was Royal Mail’s first and only significant competitor in a market characterised by significant barriers to entry and economies of scale. When there is a monopoly, economic theory dictates that an entrant, whether efficient or not, will cause the market price to fall, provided the entrant’s marginal cost is lower than the monopoly price. Thus, a permissive anti-competitive foreclosure test that sets a high bar for finding an abuse, such as an AECT, would pose a significant risk of denying consumers the benefit of lower prices, and would also mean that there could be no entry or expansion in the bulk mail delivery market at all.
Rather than rely on an AECT, in this case Ofcom undertook an in-the-round assessment of all the circumstances of the case, including Royal Mail’s overwhelmingly dominant position and the evidence of a deliberate strategy to exclude Whistl. On the basis of its in-the-round effects analysis, Ofcom concluded that the price differential was reasonably likely to lead to a restriction of competition from the point at which Royal Mail issued the contractual notices designed to introduce the discriminatory prices. This was because, at the point at which the lower prices were no longer available in practice to a competing delivery operator (i.e. Whistl), there would be a significant increase in the competing operator’s access costs for the proportion of its mail that would continue to be delivered by Royal Mail. The resulting financial impact of the price differential on an end-to-end competitor’s profitability would have been material.
In my view , the evidence showed it was highly likely to be an effective foreclosure strategy for Royal Mail. To exclude Whistl, it was sufficient to introduce arrangements whereby Whistl’s prices went up if it expanded its bulk mail delivery operations. Thus, unlike below-cost pricing in predation cases, foreclosure through the price differential did not require a costly period of negative profits for Royal Mail. Furthermore, because of significant barriers to entry, exclusion of Whistl was likely to have a long-lasting anti-competitive effect on entry and expansion into bulk mail delivery.
The CAT Judgment
The CAT found that Royal Mail’s conduct did not involve any attempt to compete on price, quality or innovate in a way that would enhance consumer welfare. Rather Royal Mail intended to reduce competition by constraining Whistl’s ability to compete. As such, the CAT concluded Royal Mail’s conduct was not competition on the merits.
In my view , the CAT’s judgment indicates that limiting protection against exclusionary conduct only to competitors that are as efficient as the dominant incumbent would lead to underenforcement. Productive efficiency of an entrant, measured by the entrant’s costs, is not a component of consumer welfare. Instead, an entrant’s cost affects consumer welfare indirectly through its effect on the post-entry price. Less efficient competitors are capable of bringing prices down and thus increase consumer welfare, particularly when entering a monopoly market, as in the present case.
As to whether economic principles identify a clear class of cases in which the AECT should be used, the CAT noted in para 529 of its judgment:
“[I]n the current state of economic thinking, there is no well-defined class of cases in which the use of an AEC test is the appropriate way of identifying anti-competitive behaviour.”
As such, the CAT did not find it helpful to attempt to attach a label to the type of conduct at issue as a means of determining whether the AECT should be applied.
The CAT also noted that, in the context of this particular case, the concept of an as-efficient competitor is inappropriate because no realistic entrant would possess the structural advantages (e.g. significant benefits from scale and scope economies or VAT exemption) or disadvantages (e.g. obligation to deliver mail to any address throughout the UK six times per week) that characterise Royal Mail, due to its legacy as the former statutory monopolist and its position as the provider of the universal postal service.
Moreover, the CAT emphasised the practical difficulties of implementing an AECT in the circumstances of this case. For example, the CAT identified issues with conducting the AECT on the basis of Royal Mail’s current costs. The economies of scale in bulk mail delivery would mean that the long-run incremental cost facing an operator is lower at higher volumes. Entry of a competitor would drive the individual volumes of Royal Mail and the entrant below those of Royal Mail when it operates as a monopolist. Consequently, the appropriate cost to assign to the as-efficient-competitor should be higher than the costs of Royal Mail at current volumes. As such, an AECT that is based on Royal Mail’s current costs would be unduly lenient.
On the issue of legal certainty, the CAT noted in para 519 that “an AEC test as a self-assessment tool provides at best a very limited degree of legal certainty and, at worst, none at all,” because there is no canonical methodology for carrying out an AECT. Since there is significant discretion about, for example, the treatment of common costs, a dominant firm that carries out an AECT to self-assess conduct will face a degree of uncertainty about whether a competition authority will accept the proposed design of the test. Indeed, by cataloguing various problematic features of the Royal Mail’s AECT, the CAT’s judgment serves as a testament to the uncertainty inherent in undertaking any AECT.
Although the CAT did not make this point, it is clear that these uncertainties would be increased where price-cost tests are specified using a “reasonably efficient operator” benchmark (e.g. in an effort to address the point that the only viable competitor in a market may not be as productively efficient as an outright monopolist).
Where does this leave effects analysis in exclusion cases?
The economics in this judgment supports the view that “effects analysis” in foreclosure cases should not be limited solely to As Efficient Competitor tests. Not only is there doubt that as a matter of practical implementation such tests offer the clear legal certainty that is sometimes claimed for them, but the well-established economic reasons for using AECTs to define “safe harbours” in predation cases do not necessarily extend to all instances of exclusionary pricing conduct.
Footnotes
1As alluded to above, Royal Mail suspended the implementation of the pricing changes once Ofcom opened its investigation.
2 C-209/10 - Post Danmark I; C-23/13 - Post Danmark II; C413/14P - Intel
3 “Guidance on the Commission's enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings”