The UK Supreme Court recently handed down its judgment on appeals relating to competition damages claims brought by Sainsbury’s and other retailers against Mastercard and Visa.
Amongst other issues, the appeals considered the legal framework for pass-on, whereby it is alleged that claimants transferred any overcharge to their own prices to customers. In these cases, the pass-on argument was deployed as a defence, although it can also be used against defendants in claims involving indirect purchases. As recognised by the Competition Appeal Tribunal (CAT) in its original Sainsbury’s/Mastercard judgment, no cases under English law had previously dealt substantively with the pass-on defence. The Supreme Court’s judgment therefore provides potentially important insights into how the issue of pass-on should be considered in competition damages claims.
This article considers the potential implications of the judgment for the economic analysis of pass-on, with a focus on its use as a defence (although similar considerations may apply where it is used by indirect purchasers). In summary:
- the judgment may suggest some degree of convergence between the legal approaches to overcharge and pass-on;
- this could potentially lead to greater emphasis on empirical techniques for quantifying pass-on before courts, given that there are certain parallels between these techniques and those that are commonly used for overcharge estimation; and
- whilst such empirical evidence may not itself be sufficient to meet the legal requirement to prove causality, it may be used in conjunction with other factual and economic evidence to help inform a court’s overall assessment of a pass-on defence.
The Supreme Court’s conclusions in relation to pass-on
In its judgment, the Supreme Court clarified the law in relation to the standard of proof required to support a pass-on defence. It held that the ‘broad axe’ principle should apply, under which “unreasonable precision” is not required in order to prove the amount of pass-on. In support of this conclusion, the Supreme Court noted that the claimants would have been likely to account for any overcharge in their downstream pricing decisions together with many other costs, rather than on an individual cost-by-cost basis. The judgment also drew a comparison with the use of the pass-on argument by indirect purchasers, for which it is generally accepted that the broad axe principle applies, in finding that it is not clear why the fact that the burden of proof is on defendants to establish pass-on should mean that a greater degree of precision is required for quantification than would be expected of indirect purchasers.
The Supreme Court also confirmed that the burden of proof lies with defendants to establish the pass-on of any overcharge. However, it noted that the significance of this burden should not be overplayed because there is a “heavy” burden on claimants in relation to disclosure. In particular, the judgment explained that the claimants are likely to have exclusive access to much of the factual information which would be relevant to assessing whether and how any overcharge was passed-on to customers via their pricing.
Convergence in the legal approaches to overcharge and pass-on?
By reaching the conclusions set out above, the Supreme Court appears to have defined legal principles for the analysis of pass-on that are broadly similar to the principles already followed by courts for the analysis of overcharge, as summarised in the table below.
Another example of potential convergence in the approach to the analysis of overcharge and pass-on is reflected in the CAT’s January 2020 ruling in relation to disclosure in the Trucks proceedings. The ruling set out the CAT’s view that the quantification of pass-on would need to adopt some of the key characteristics of an overcharge-style approach, including the application of “established economic techniques” and large volumes of data, with general factual evidence serving a supporting role.
It remains to be seen whether the Supreme Court’s judgment and the CAT’s ruling signal a more general trend towards convergence between the legal approaches taken by courts to overcharge and pass-on. If it does, however, there are potential implications for the types of economic analysis of pass-on that may be considered relevant in future competition damages cases.
Potential implications of convergence for the economic analysis of pass-on
As explained by the CAT in its ruling on Trucks disclosure, an overcharge is typically quantified “through averages, extrapolations and aggregates”, rather than on the basis of a precise calculation for each individual transaction. Most accepted methodologies for estimating an overcharge involve employing a comparator approach, whereby the prices that would have prevailed ‘but for’ the infringement are estimated by reference to data in comparable, but infringement-free, situations. Such analyses may make use of statistical techniques to ensure that any comparisons are as like-for-like as possible. This can require large amounts of data, which in turn can mean that it is necessary to group transactions and customers together to generate average measures of the overcharge. It also allows the overcharge to be estimated rather than quantified with certainty, and statistical theories and techniques are required in order to understand the likely robustness of any estimate.
In principle, similar economic approaches can also be used to quantify the extent of pass-on (as noted in, for example, the European Commission’s pass-on guidelines). However, for various reasons, different approaches are often used in practice, such as analyses assessing the general relationship between prices and input costs in the downstream market and/or whether the claimant’s profit margins were lower during the infringement period than in the period absent the infringement. Nevertheless, where sufficient data is available to conduct such analyses, there can be similarities between these approaches and those used for overcharge estimation. For example, they may generate estimates of the extent of pass-on across transactions, rather than figures for each individual transaction, and statistical theories and techniques may need to be applied to understand the reliability of any estimates. It is therefore possible that any trend towards convergence, and particularly the application of the broad axe principle, could lead to greater (and/or more prominent) use of these approaches to quantifying pass-on.
The requirement to demonstrate causality: a fly in the ointment?
In its original Sainsbury’s/Mastercard judgment, the CAT concluded that a distinction can be drawn between the legal and economic definitions of pass-on. It explained that whilst “pass-on might be said to be a fact of economic life”, the legal definition of pass-on relies on demonstrating a causal connection between any increase in input costs as a result of an overcharge and an increase in downstream prices. This suggests that approaches to estimating pass-on of the type set out above may be considered insufficient if they are not complemented with factual and other economic evidence which establishes a clear link between the overcharge and the claimant’s downstream pricing decisions.
The CAT ultimately rejected Mastercard’s pass-on defence because it considered that a causal link had not been demonstrated, although it recognised that doing so poses significant challenges. Similarly, although the Supreme Court’s judgment recognised the inherent challenge associated with disentangling the effect of any overcharge on downstream pricing from other factors which may affect pricing, it did not seek to provide further guidance on how the issue of causality could feasibly be addressed within this context.
As a result, there would appear to be a question around the extent to which full convergence in the approaches to overcharge and pass-on could occur in the face of this causality issue. Nevertheless, it is worth noting that in its Sainsbury’s/Mastercard judgment, the CAT only granted interest to Sainsbury’s in relation to 50% of the overcharge on the basis that the rest of the overcharge had been passed on “in the non-legal sense”. Does this suggest that empirical analyses of pass-on might play a more prominent role in courts’ assessments of damages, even where they do not contribute to a formal legal pass-on defence?