This paper – which can be found here – deals with the the passing-on defence.


A bit of context may be in order here. “Passing-on” is the passing of damage suffered by a victim of a cartel to other parties, usually by increasing the price of re-sale of the cartelised goods or of the products for which the cartelised good is an input. Passing on may be invoked by an indirect purchaser in order to claim harm suffered by himself as a result of overcharges on the purchases of products or services from direct customers of the cartelist or from companies which have incorporated goods affected by the infringement into their own products or services: this is passing-on as a “sword”. Alternatively, and more commonly, pass-on may be raised as a defence to claims for damages on the ground that the claimant has incorporated overcharges, or part of them, in its downstream prices of products or services, thus reducing its actual harm: this is passing-on as a “shield”.

Passing-on is frequently raised in antitrust damages claims in the EU, and can be crucial to the quantification of damages (or even to a party’s standing to claim for damages). Nevertheless, passing-on has not to date been determinative in many cases in the EU, and has seldom been subject to any detailed expert quantification. In the majority of cases where the issue has been determinative of the outcome of the case – and a recent paper puts the number at 24 cases as of December 2016  – it has been raised as a defence. In more than half of those, the court rejected the passing-on defence entirely, whereas in about 40% of the cases the court determined that the claimant had passed on the overcharge entirely.

More importantly for the purposes of the paper attached, a number of senior court decisions seem to indicate a resistance to fully adopting an economic approach to passing-on.  In particular, a distinction between the ‘economic’ and the ‘legal’ conceptions of passing-on may be emerging: while the former is based on economic theory applied to the relevant markets in the case at hand, the latter is a less technical approach which is based on the requirements set out by the legal requirements of causality and reasonableness.

This paper is a critique of this trend, and in particular of the first Sainsbury v MasterCard decision by the Competition Appeals Tribunal (CAT), which was adopted last year. In that case, the CAT found that MasterCard had set excessive multilateral interchange fees (MIF).  The success of Sainsbury’s claim for damages then depended on whether Sainsbury had passed on the excess (Sainsbury is a supermarket chain). The Tribunal considered three passing-on issues – the passing-on of MasterCard’s MIFs by card acquirers to merchants; the passing-on of the MIF by Sainsbury’s to retail customers via higher retail prices; and passing-on when awarding pre-judgment  interest. The Tribunal found that the same overcharge had been fully, not, and half passed-on respectively. The key passing-on issue considered by the Tribunal was MasterCard’s claim that Sainsbury’s passed-on the entire MIF overcharge to its retail customers in the form of higher prices and therefore suffered no loss. This was hotly contested by Sainsbury’s, but its defence was rejected by the Tribunal because of lack of evidence.


The paper begins by describing the case, and by reviewing the situations where passing-on had been considered at the European level: namely, in tax cases (including an interesting discussion of the application of principles of unjust enrichment in this area) and in the context of the EU Damages Directive.  It then moves on to describe how the case provides a comprehensive review and clarification of the English law of competition damages and passing-on. In particular, the court set out its “legal test” for passing-on as a “defence” by  distinguishing it from “economic pass-on” at para. 484. The case can thus be read as setting out a two-part legal test for the passing-on defence. According to the legal test, for the defendant to succeed it must prove that there is: 1. an “identifiable” increase in the purchasers’ prices causally connected with the overcharge; and 2. a class of the downstream claimants’ who paid the higher prices.

The paper’s argument, put simply, is that there is no good reason to consider that there needs to be a difference between legal and economic tests, and that the reasons to so think are a mistake.

  • Regarding the requirement that “there must be an “identifiable” increase in the purchasers’ prices causally connected with the overcharge”: The court seems to think that, when faced with an unavoidable increase in cost, a firm can do one or more of four things, all of which can be considered passing on: (1) It can make less profit (or incur a loss or, if loss making, a greater loss); (2) It can cut back on what it spends money on – reducing, for example, its marketing budget; or cutting back on advertising; or deciding not to make a capital investment (like a new factory or machine); or shedding staff; (3) It can reduce its costs by negotiating with its own suppliers and/or employees to persuade them to accept less in payment for the same services; and (4) It can increase its own prices, and so pass the increased cost on to its purchasers.

The author notes that: (1) is not a reaction but a consequence of an overcharge, and can include partial passing-on (or even full passing- on that leads to lower quantities being sold). The reductions in costs described in (2) would lead to a contraction of the firm, reduce its output and sales, and lead to real losses by the claimant, which could provide grounds for claims for damages for lost profits. Reducing suppliers’ or other input prices (Response (3)) is not passing-on but “pass-back”, such as when a supermarket pays dairy farmers less for their milk. Passing back should be offset against the claimant’s damages when it does not arises as a specific reaction to the increase in the price of the cartelised goods, and thus breaks the chain of causation. Thus, only (4) would seem to be passing on from a pure economic (and legal) perspective, and considerations (1) to (3) are relevant only to the calculation of the original damages.

  • Regarding the second requirement that “there must be a class of the downstream claimants’ who paid the higher prices”: the court is wrong to consider that passing-on requires an increase in absolute retail prices. This is so for at least two reasons: (i) passing-on can happen when prices are unchanged but the purchasers respond by reducing product weight, quality or pack size. In these cases the customer is paying the same for less, and thus such a situation is consistent with increasing, constant and falling prices; and/or (ii) many cartels respond to and seek to arrest the decline in prices triggered by price wars, excess capacity and/or the entry of a new firm or imports which expand production and depress prices. These cartels slowdown the speed and extent of price declines so that the actual reduction in prices is less than the decline in counterfactual prices would be.

The court further distinguished legal and economic passing-on by asserting that the former balances the risks of under-compensation and over-compensation of claimants, while the second one does not. However, this attempted distinction has little to do with the economics of passing-on. The Tribunal was addressing the evidentiary concern of whether, when faced with uncertainty, the court should pay more attention to a Type I (undercompensating the claimant) than Type II (overcompensating the claimant) error. This must be the case since, if the Tribunal knew the passing-on rate, there would be no need for the second limb of its legal test. Hence, this second limb is actually an evidentiary requirement that must be fulfilled in order for the court to conclude that passing-on actually occurred.

The author’s last criticism of the decision is that the Tribunal’s test only applies when passing-on is used as a defence. The effect of the decision is that, where passing-on is raised as a “defence” the defendant cannot expect to deny the claimant compensation by default, mere assertion, and/or economic and empirical generalities. The defendant (i.e. the cartelist) must prove passing-on to the civil standard of more likely than not. However, where passing-on is used to mount a claim by indirect purchasers’, the normal rules of pleading harm apply, which leads to a more permissive approach whereby indirect economic and commercial evidence are allowed to be used to estimate the passing-on rate and quantum. This would reflect EU law’s  “principle of effectiveness”  by modifying the standard of proof, if not the type of evidence which can be used, to ensure that the “right to damages is not practically impossible and excessively difficult”.

I think the article provides a very useful overview of the differences between ‘legal’ and ‘economic’ conceptions of passing on. Further, I have two main comments about this paper, with which economists may disagree (but, after all, this paper is by an economist who is developing legal arguments): (i) it demonstrates how hard it can be to establish passing-on in practice; (ii) I think that, in practice, “passing on” is a legal concept, and that economics must be deployed to establish its content of this context. In other words, whether the legal and economic conceptions of passing on correspond to each other is ultimately a legal question.

Author Socials A weekly email with competition/antitrust updates. All opinions are mine

What do you think?

Note: Your email address will not be published