Download our article on the treatment of brand value in Le Patourel v BT
The CAT’s judgment in Le Patourel v BT, an excessive pricing claim regarding BT’s prices for standalone landline services, was handed-down on 19 December 2024. Despite finding that BT charged excessive prices under LIMB 1 of the United Brands test, the CAT concluded that the prices were not unfair in and of themselves under LIMB 2, in large part because BT’s additional services, referred to as ‘Gives’, (onshore call centres, blocking of nuisance calls etc) and brand value provided distinctive ‘economic value’ to its customers. It is not clear to what extent the CAT would still have dismissed the claim had it not placed weight on BT’s Gives and brand value under LIMB 2. Nevertheless, the CAT’s treatment of BT’s brand value across the two limbs of the test warrants reflection, as it is likely to have double counted the allowance for brand value across the two limbs of United Brands and assigned an implausibly high value to BT’s brand.
The Le Patourel claim was an opt-out collective proceedings case against BT on behalf of 2.3m BT customers, with Mr Le Patourel acting as the class representative. Mr Le Patourel claimed that BT charged unfair prices for Standalone Fixed Voice Services (SFV Services), i.e. landlines to residential customers, which amounted to an abuse of dominance, contrary to Section 18 of the Competition Act ’98. The judgment was handed-down on 19 December 2024, within which the CAT dismissed the claim.
Consistent with the well-established framework for abuse of dominance in the form of unfair pricing, the CAT assessed market definition, dominance and then abuse. The latter followed the two-limbed United Brands test.
The CAT defined the market as the provision of SFV Services to residential customers and disregarded BT’s claims that broadband bundles were in the same market.[1] With regards to dominance, there was a wholesale line rental product available during the claim period, allowing competing offerings from the Post Office and other broadband providers (TalkTalk, Sky and Virgin Media) but BT retained a market share of between 82% and 85% by revenue. The CAT ultimately found that BT’s high market share of a disengaged customer base (comprising a large number of elderly customers[2] who had stayed with the same provider for many years) in a declining market supported a finding of dominance.
Having found that BT was dominant, the CAT proceeded through the two LIMBs of the unfair pricing assessment, based on the United Brands Framework.
LIMB 1 – price vs competitive benchmark
All parties derived a relevant competitive benchmark, based on a measure of costs plus a reasonable return (with the latter based on margin benchmarking, as opposed to a return on capital approach) and compared this to the average revenue per user.
The claimants’ expert estimated relevant costs using BT’s Regulated Financial Statements (RFS), which adopt current cost accounting and a fully allocated cost (FAC) approach. Unfortunately, BT stopped reporting costs separately for SFV in its RFS in 2009, so the claimants’ expert indexed the 2009 figure using CPI for the claim period (from 2015/16 to date).[3] He then applied a margin of 8.9% (with an upper end of 10%), based on the margin earned by BT in 2006, which was the last year in which BT’s voice charges were regulated by Ofcom.
BT’s expert disregarded the RFS because they were out of date and used a FAC approach, which she did not consider gave sufficient flexibility for the recovery of common costs. Instead, she took total costs from the profit and loss account of BT Consumer (the retail division of BT Group, which includes landlines, mobile, broadband and digital TV) over the claim period, and allocated costs to SFV using a variation of standalone costs, with sensitivities for other methods such as FAC.[4] BT’s expert ultimately ended up allocating almost two thirds (62%) of the common costs in BT Consumer to SFV, when even by revenue (at the alleged inflated price) the share was 18% (using a 2015/16 figure; the simple average for the claim period was 12%). She then applied a 25% margin, based on the 90th percentile of her margin comparator dataset.
Despite accepting that the RFS was a relevant – albeit outdated – benchmark, the CAT did its own allocation of costs from BT Consumer to SFV, where it allowed 40% of common costs to be allocated to SFV and applied a 13.5% margin. The 40% common cost allocation figure was selected because it was lower than BT’s expert’s assumption, which the CAT considered carried a significant risk of being overestimated, whilst also reflecting “the principle that firms in competitive conditions should enjoy a considerable degree of flexibility in how those [common] costs are recovered.” The 13.5% fair margin was selected because it was close to the median from BT’s expert’s dataset of 14% and slightly above the 10% margin, at the top of the claimants’ expert’s range.
Whilst the focus of this note is on the treatment of brand value, we note the following with regards to the CAT’s LIMB 1 assessment:
- If the (appropriately inflated) cost figures from the 2009 RFS were used to derive the common cost allocation percentage from BT Consumer to SFV, then the proportion would have been 9%[5] and under a revenue-based allocation, the proportion would have been 18% (using a 2015/16 figure; the simple average for the claim period was 12%). The CAT’s 40% common cost allocation appears high in this context.
- SFV service providers, like BT Consumer, pay for a wholesale line rental product from Openreach. The tangible asset base is therefore likely to be low. SFV was also a mature, legacy product. In our view, 13.5% is therefore likely to be a relatively generous allowance for a fair margin (e.g. in retail energy the CMA concluded that a 1.5% fair margin was consistent with competition for an established supplier). A return on capital cross check would have been helpful to test the validity of the 13.5% fair margin assumption. It may also have drawn the CAT’s attention to the risks of double counting and overvaluing brand value in LIMB 2, which we come onto below.
Under the CAT’s approach to cost-plus, it found excess profits, i.e. profits above the cost plus benchmark of 38%, compared to the claimants’ alleged 78%.
The CAT specified that any delta above 20% is excessive (arguably, with light justification, given the potential precedent implications of such a threshold), so the prices were deemed excessive under LIMB 1. However, it noted that the estimated 38% excess is ‘dramatically different’ to the 78% that was claimed.
LIMB 2 – Unfairness
Proceeding to LIMB 2, a range of arguments were put forward by both parties. The case appears to rely – at least in large part – on the assessment of whether the price was unfair in itself and within that: i) whether the delta under LIMB 1 was so excessive that it was self-evident that the price was unfair; and ii) whether the prices bore a reasonable relationship to economic value.
With regards to the size of the delta under LIMB 1, the CAT concluded that given the ‘dramatically different’ excess that it found compared to the claimant’s expert, the size of the excess was not sufficient for the price to be demonstrably unfair simply due to the size of the excess. In this regard, we note that the CAT’s cost-plus benchmark appears generous, given the 40% cost allocation and 13.5% fair margin (as discussed above).
It then proceeded to evaluate whether the price was unfair with reference to economic value. A key element of the CAT’s assessment here is the value of BT’s ‘Gives’, which included services such as onshore customer service and Call Protect (which allowed for nuisance calls to be blocked) and BT’s brand to SFV customers. Here, the CAT found that the Gives and BT’s brand value did constitute distinctive value to SFV customers and therefore “In the context where the excessive prices were at the level we found them to be, we conclude that they did bear a reasonable relation to the economic value of the SFV Services.” The CAT also considered the dispersion of prices and cost efficiencies under its interpretation of ‘normal and sufficiently effective competition’ in its unfairness analysis. It concluded that both elements of dispersion counted against a finding of unfairness under LIMB 2.
The CAT then went on to assess whether the prices were unfair by comparison. On 1 April 2018, BT entered into commitments with Ofcom to reduce the price for customers who only bought landline services from BT. The claimants submitted that the price under the commitments was a relevant benchmark but the CAT disagreed, as the purpose and context was different. BT suggested that the prices of its competitors showed that its prices weren’t unfair. In this regard, the CAT concluded that the position of BT’s competitors was “at least some evidence that its prices were not unfair and in any event is a counterbalance to the CR’s [Class Representative’s] comparators”.
The treatment of brand value
It is not clear to what extent the CAT would still have dismissed the claim had it not placed weight on BT’s Gives and brand value under LIMB 2. [6] Nevertheless, the evidence on brand value appears to have been a key part of its overall assessment of unfairness and in turn, abuse.
We consider that there are three main reasons why the CAT’s reliance on brand value under LIMB 2 warrants reflection.
First, capital costs associated with brand value (and other intangible assets) should already be captured under LIMB 1 in the form of the 13.5% margin. The fair margin is a function of capital employed (strictly speaking, based on modern equivalent asset values for both tangible and intangible assets, which may differ from accounting asset values) and the cost of capital. The fair margin can be estimated by quantifying economic capital employed and the cost of capital, and multiplying the two together (sometimes referred to as the ‘return on capital’ approach (RoC)) or by identifying comparable firms in competitive industries and using the margins earned by those firms as a benchmark (often referred to as the ‘return on sales’ (RoS) approach). Despite the difficulties in identifying appropriate margin comparators, a RoS approach is sometimes favoured over a RoC approach because it is deemed by the experts involved to be more accurate or efficient than valuing the economic capital employed, including intangibles like brand (which can be hard to quantify), and estimating a cost of capital. Fundamentally though, RoS and RoC approaches are both estimating the same thing – the cost of capital. This cost of capital, approximated by the CAT via a 13.5% fair margin allowance, should account for the cost of capital associated with all assets needed to provide the service in question, including intangible assets, like brand. In addition, to the extent that ongoing expenditure in Gives and BT’s brand is included in the BT Consumer common costs, of which the CAT has allowed a 40% allocation to SFV, there will already be an (arguably generous) allowance for the costs of maintaining the brand value in the cost stack.[7]
Second, even if you allow additional brand value to be recognised under LIMB 2, beyond that included in the cost-plus assessment of LIMB 1 (i.e. you assume that the 13.5% is not sufficient to provide a return on BT’s intangible assets, including brand) , the implied royalty rates that would get you back to zero economic profits in this case[8] are implausibly high. This can be illustrated by taking the CAT’s estimate of the monthly economic profit per user of c.£6 as a percentage of the average revenue per user of c.£23, which implies a royalty rate of 26% of revenue, even assuming none of the allowed margin in LIMB 1 gives a return for brand value. A royalty rate of 26% of revenue is substantially above surveys of market norms, which suggest royalty rates of 5-7% of revenue[9], and is approximately double the fair margin. Neither of these implications seem likely.
Third, the barrier to entry and reason why BT had market power in this case, is the disengaged, vulnerable customer base. Where customers are disengaged, the incumbent’s brand value may be distorted above the level it would be under normal and sufficiently effective competition (where customers are reasonably engaged and informed) because the brand has a higher value by virtue of it being a tool to obtain and retain, disengaged and profitable customers. Allowing a value for brand that is significantly above market norms risks allowing a dominant firm to ‘price in’ the very feature of the market that is giving rise to market power.
Overall, whilst we agree that brand value is a relevant consideration in the assessment of unfair pricing cases and economic profitability more generally (where some level of brand value would be expected in a competitive market for the product in question), care must be taken to avoid ‘double-counting’ and allowing for implausibly high brand values. This is particularly the case where the feature of the market that is giving rise to market power is the disengaged customer base, as allowing a large brand value risks ‘pricing in’ the excess profits.
The CAT’s reliance on brand value under LIMB 2 in Le Patourel clearly has implications for future excess pricing cases. In addition, given the importance of the treatment of brand value to economic profitability assessments more generally, the implications may be felt more widely, reaching other areas of competition and regulation where profitability analysis is an important component of the economic evidence base.
Disclaimer
This document has been prepared by Kairos Economics based on publicly available information and is not intended as tailored advice for recipients. Kairos does not therefore accept liability for any reliance placed on the content by recipients.
The views expressed in this document are based upon our interpretation of the published CAT judgment, dated 19 December 2024, only. Evidently our views may differ if we had access to the underlying expert reports and evidence and/or if further evidence is made publicly available.
[1] There is a lengthy discussion on whether the amount of switching to broadband bundles during the claim period suggested a wider market definition, as submitted by BT and its experts. However, the CAT ultimately concluded that this switching was largely due to a secular trend, rather than broadband bundles and SFV being in the same market.
[2] Ofcom analysis finds that over 40% of voice-only customers (VoC) are at least 75 years old, and 40% live in DE socio-economic group households (Semi-skilled & unskilled manual occupations, unemployed and lowest grade occupations).
[3] A shorter claim period applied to VoC, due to commitments agreed with Ofcom in 2018, which related to VoC (a subset of BT’s SFV customers).
[4] Naturally, there is a great deal of detail on the different cost allocation methodologies (SAC combi, distributed SAC (or DSAC) and FAC) with numerous sensitivities – all of which are dealt with in the Judgment but which we don’t go into here, given the focus of this piece on the treatment of brand value.
[5] The claimants’ expert’s estimate of indirect costs from the RFS as a proportion of the BT Consumer overheads.
[6] Indeed, the levels of price and cost dispersion that would be expected under conditions of normal and sufficiently effective competition is also used to assist the CATs assessment of whether the prices were unfair in and of themselves under LIMB 2.
[7] The inclusion of brand value as part of costs under LIMB 1 is recognised in OECD (2011), Policy Roundtables: Excessive Prices, page 57 and footnote 156.
[8] Note, we are not suggesting that all economic profits are unfair i.e. that the test should be zero economic profits/perfect competition. Instead, we are illustrating the scale of the likely brand value overstatement by demonstrating how large the brand value would need to be for economic profits to fall to zero.
[9] See, for example: KPMG, ‘Profitability and royalty rates across industries: Some preliminary evidence’ (2012)
