Show Less
You do not have access to this content

The Passing-On Problem in Damages and Restitution under EU Law

Magnus Strand

‘Passing-on’ occurs when harm or loss incurred by a business is passed on to burden that business’s customers or the next level of the supply chain. In this authoritative book Magnus Strand provides the first comprehensive examination of passing-on in EU law damages and restitution. The analysis covers a broad range of contexts including competition damages and the repayment of charges.
Show Summary Details
This content is available to you


Magnus Strand

A.    What is passing-on?   1.001

I.     A short story   1.001

II.    The passing-on problem: A set of factual circumstances …   1.006

III.    … Triggering a set of legal problems   1.011

IV.    What situations are passing-on situations?   1.023

B.    The parties   1.030

I.     Introducing the acronyms   1.030

II.    Alpha   1.033

III.    Beta   1.036

IV.    Gamma   1.041

V.     The need for coordination   1.044

C.    Purpose and approach of the book   1.046

I.     Purpose   1.046

II.    Contexts and perspectives   1.047

III.    Structure of the presentation   1.054


I.    A short story

1.001 If you were a baker, your business would be very sensitive to wheat flour prices. Imagine a 5 per cent increase in wheat flour prices. How would you respond? Well, if you wanted to maintain your profit margin, you would consider increasing your own prices by 5 per cent. But what if your local competitor didn’t increase her prices that way? Maybe then it would be better to maintain your prices intact, in order to avoid losing market shares. But then again, if you know your bread is so much better than hers, and that your customers are not likely to buy her bread instead of yours just because of the increased prices, maybe it would be safe to make the 5 per cent increase? Actually, maybe you just might consider a 6 or 7 per cent increase, as it could still be blamed on the wheat flour prices?

1.002 Now if instead you ran a famous sandwich parlour, and the bakery from which you bought your bread increased its prices by 5 (or even 7) per cent, what would you do? Well, as above, your answer would depend on your need of a maintained profit margin, local competition, and your experience of customer behavior. You might keep current prices, absorbing the loss, or you might increase your prices, passing the loss on to your customers – perhaps even with an extra increase of profit margin if you felt confident in doing so.

1.003 Now, imagine that the increased wheat flour prices were caused by a new system for grain quality declarations introduced by your Member State, which included a charge to cover costs for examining the declarations submitted. Wholesalers on the grain markets had decided to respond to the new costs incurred by reason of this system by incorporating them, or rather an estimate of the incurred costs, in their selling prices. Some of them had even indicated this cost as a specific item on their invoices. However, after a few years, the grain quality assurance system was struck down by the European Court of Justice.

1.004 Following the judgment of the Court, the wholesalers bring an action against the national administration responsible for the grain quality declarations, claiming damages at the amount of costs illegally incurred. The national administration however will pay no damages, as the costs at issue were compensated for by increased selling prices. In some instances, the costs were even specified in invoices. Under such circumstances, the national administration argues, an award of damages would be pure profit – indeed, an unjust profit.

1.005 Contemplating these arguments as they are presented in the media, you – the baker or sandwich maker – will begin to wonder: if the economic burden of the grain quality declarations was passed on to me, should I perhaps try to bring an action for damages against the national administration? Or should I try to recover from the wholesalers? Meanwhile, the neighbouring café owner starts thinking along the same lines …

II.    The passing-on problem: A set of factual circumstances …

1.006 Situations such as the one in the above example are not easily solved in law, as they involve at least three – but usually many more – parties, general interests (such as compensation, justice and deterrence from illegal conduct) and, more often than not, severe problems with regard to establishing what harm has been suffered by whom and to what extent losses, that seem to have spread through the markets like ripples on water, can be traced for the purposes of reparation. The presupposition is that an economic burden can sometimes be passed on, in whole or in part, down the supply chain. The act of letting harm incurred pass through a business and move on to burden your own customers, at the next level of the supply chain, is known as passing-on. Consequently, the ‘passing-on problem’ is not a distinct legal problem but a set of factual circumstances that triggers various legal problems. Those legal problems will be referred to collectively as the passing-on problem.

1.007 In the above example the context of the EU internal market was used in order to give a brief illustration of the passing-on problem, but the problem as such can occur in a variety of legal contexts. In EU law, the passing-on problem belongs to the context of private enforcement, i.e., claims brought by individuals or undertakings enforcing their rights under EU law. More precisely it can arise in the context of claims for damages or restitution for a breach of EU law. More generally however, the passing-on problem can arise in the context of any action where the relevant injury has been incurred in the form of increased production costs in a business.1 Notwithstanding whether passing-on is recognized explicitly, implicitly2 or not at all, the problem may arise in restitution as well as in damages, and in contractual as well as in non-contractual liability. In the context of competition law damages, the Commission has stated that the passing-on problem is ‘at the heart of the way in which a system for damages for breach of antitrust law … functions.3

1.008 An analysis of the passing-on problem is not concerned with whether the party who allegedly gave rise to the initial incidence of economic disadvantage for its counterpart (be it the EU, a Member State or a private individual) has committed a tort or been unjustly enriched, as the case may be. Argumentation with regard to passing-on may in other words presuppose a certain position concerning the tort or unjust enrichment, either in the positive or in the negative. Assuming it was the Council, the Council’s legal representative in the proceedings might use a phrase like: ‘Even if the decision of the Council was illegally discriminatory, the Council cannot be held liable in damages because any injury suffered by the claimant will have been passed on to its customers …’. At this point in the argumentation, i.e., the point in which we proceed to discuss the passing-on problem, the issue of tort or unjust enrichment per se can in other words be put in parenthesis, because the passing-on problem arises only in certain problems having to do with how the tort or unjust enrichment can be legally remedied.

1.009 The passing-on problem has many aspects. The legal problems entailed include who among the parties involved should be able to sue whom, how far down the supply chain it is reasonable to trace the passing-on of an economic burden, whether or not to take passing-on into account in the quantification of awards, and of how to prove the existence and extent of passing-on. It is common to give structure to these problems by distinguishing the use of passing-on by defendants as a defence or ‘shield’ against an action, from the use of passing-on by claimants positioned down the supply chain as a ‘sword’. Passing-on as a ‘shield’ means arguing that a claimant cannot recover as the alleged harm has been passed on and any award would therefore constitute a windfall. This is also called ‘the defence of passing-on’ and will largely be referred to as such in this book. Passing-on as a ‘sword’ means arguing that the economic burden at issue has been passed on to the claimant, and that the competent court must therefore allow the action brought before it although there may not have been any direct nexus between the claimant and the defendant. This image of the sword and shield is quite helpful, but it fails to identify the specific legal issues triggered by the passing-on problem and, therefore, also fails to distinguish different legal issues pertaining to ‘sword’ use of passing-on: court access issues and substantive proximity issues (e.g., causation).

1.010 The major legal issues triggered by the passing-on problem are accordingly three, and these three will now be introduced with more detail.

III.    … Triggering a set of legal problems

Legal Problem 1

1.011 Who should have access to court for the purposes of bringing an action for damages or restitution, as the case may be, in a passing-on situation?

1.012 It follows from case law such as Danfoss4 and Kone5 that the Court of Justice is inclined to strike down national rules that automatically exclude certain classes of claimants from the possibility of bringing an action before the competent national courts. It furthermore follows from Directive 2014/104/EU on competition law damages that Gamma claimants must be granted access to bring an action for damages for a breach of EU competition law.6 As illustrated by Milutinovic however, it does not follow that EU law requires for anyone, no matter how remote from the initial tort or transaction, to have unfettered access to bring an action.7 Moreover rules hindering access need not be as blunt as an automatic bar. The Court itself has laid down certain limitations of access to court to bring an action under Article 340(2) TFEU that are of interest here, e.g., the requirement to show that the alleged harm pertains to the claimant’s own assets.8 The Court has also accepted national rules that bar a claimant from bringing an action for competition law damages where it is established that that party bears significant responsibility for the distortion of competition.9

1.013 This book includes an in-depth analysis of EU law on access to court and assesses the possibilities for Gamma claimants to bring proceedings within the various contexts listed above.

Legal Problem 2

1.014 To what extent is a causal or other nexus required for the purposes of establishing causation or, in restitution, that the enrichment of the defendant is attributable to the disadvantage of the claimant?

1.015 Although access to court may very well be a problem for many claimants who were not directly involved in the initial tort or transaction, the highest hurdle to jump may yet be the substantive assessment of whether there is a causal link between a breach of law and the injury sustained by such a party or, in the context of restitution, that the disadvantage corresponding to the alleged enrichment is attributable to such a party. As noticed by Toth the proving of such points can be difficult for claimants ‘in the field of economic and commercial relations where the cause of an event can usually be traced back to a number of factors … operating simultaneously or successively and producing direct as well as indirect effects’.10 This applies a fortiori to claimants downstream, who need to prove not only that harm or loss was suffered by its supplier (or supplier’s supplier …) but also that the same harm or loss was subsequently passed on to them, and that the connection is still strong enough for an award of damages or restitution to be payable. Downstream claimants are thus vulnerable to arguments of being too remote from the transaction at issue to bring a successful action. This set of problems has been bundled together here as problems of substantive proximity.

1.016 It has been endeavoured to distinguish these proximity problems from procedural bars, but it should immediately be admitted that this distinction is not always easily upheld, as a procedural bar may be legitimized by proximity reasons, as in Danfoss and Kone. There is therefore some overlapping. Nevertheless the grey zone is surrounded by more clear-cut cases, and the grey zone itself can be accepted and managed as such.

Legal Problem 3

1.017 How should awards be estimated – e.g., what heads of damages are available and to what extent should passing-on have an impact on the calculation of the award?

1.018 Access to court issues and proximity problems usually face downstream claimants rather than claimants who were directly involved in the initial tort or transaction. By contrast the issue of how awards are estimated constitutes a problem for both types of claimants, although most case law has concerned actions brought by the latter category. This is the specific context in which we are concerned with the classic defence of passing-on, which is perhaps the most often discussed aspect of the passing-on problem.

1.019 The underlying issue in this regard is whether the passing-on of the harm or disadvantage should be considered in the estimation of an award of damages or restitution, as the case may be. In damages the ideology of reparation arguably speaks in favour of considering passing-on, and to disregard passing-on it is necessary to let reparation yield to deterrence.11 In order to eliminate any lingering incentives to act in breach of EU law, the estimation of damages could thus be designed to ensure the disgorgement of Alpha’s profit from the breach. In restitution the ideology of reversal of an illegitimate transaction speaks against considering passing-on. These differing tendencies will be discussed and compared in this book.

1.020 Each of these themes 1–3 will be discussed separately in the chapters below. For every theme there will be issues concerning evidence. These will not be discussed separately but in connection to the discussion of each theme identified above. Anyone interested in evidence issues is also strongly recommended to see section 9.E.I where such issues are discussed in detail.

1.021 The above subdivision of the legal issues triggered by the passing-on problem may entail that a number of issues that should perhaps be distinguished in national jurisdictions will be discussed indiscriminately. These include the distinctions between causation issues and remoteness issues, between the right to take legal action and the occurrence of procedural hindrances, and between quantification of harm and estimation of the award. Too much attention to national distinctions would however have burdened the structure of this presentation, and it is hoped that readers who need to make such distinctions can identify which line of reasoning in this book is relevant to each specific class in the national context.

1.022 What of other EU law problems triggered by the passing-on problem that are not covered at all? A few should be mentioned. One is that parties who were directly burdened by the initial tort or transaction need not necessarily respond to the costs thus incurred by trying to raise customer prices: they might also try to do so by cutting other costs. In this way the burden of the illegitimate cost originating from the tort or transaction may ‘pass back’ rather than pass on. In other words it would be pushed back up the supply chain instead of continuing down. Accordingly suppliers might suffer. Although their situation will not be analysed as such in this book, it should be pointed out that such claimants will find themselves in a situation reminiscent of that of downstream claimants, and much of the discussions below that concerns downstream claimants will be relevant to supplier claimants as well.12 Another legal problem that has already been mentioned is that potential claimants at any level of the supply chain might be held back from bringing an action by business reasons. This is not a legal problem but better analysed in economics. A final issue to be mentioned in the context of passing-on is the availability of class actions. Although the availability of a class action may promote private enforcement of EU law, such considerations fall outside the chosen scope of this book. Therefore it will not be discussed in depth but only be touched upon where immediately relevant.13

IV.    What situations are passing-on situations?

1.023 It is difficult to define precisely what situations are passing-on situations and what situations are not, but for the purposes of a legal approach to passing-on it is necessary to find a working definition of the concept. This section is intended to bring some stability to the discussion by indicating what economic events can plausibly trigger a passing-on situation, and what economic events cannot.14 It is not intended here however to conduct any review of the case law of the European Court of Justice and the existing legislation to delineate passing-on instances.15 Nor should the model suggested here be mistaken as being anything but a proposed working model. Nevertheless the suggestions submitted are based on the case law and legislation that will be presented and analysed in the following chapters. Interested readers are therefore encouraged to re-read this section in light of those chapters.

1.024 Passing-on argumentation is commonly used in EU law when a cost can be connected to the production, purchase or import of a specific commodity and that commodity is subsequently sold – as it was or following its incorporation into another commodity – to the next party in the supply chain. The most typical example of a cost that is passed on in this manner is VAT, which is explicitly designed to follow the commodity at issue down the supply chain. Other examples are less obvious, making it difficult to decide precisely what situations will be legally relevant passing-on situations. Is passing-on legally relevant only when products are resold unaltered (e.g., shoes)? Whenever the overcharged product is included in the final product (e.g., leather for shoes)? What about overcharges in other factors of production (e.g., shoe-making machinery)? The further away from the reselling of unaltered products the situation at hand is, the more remote it seems that an overcharge will have been passed on within the meaning of Directive 2014/104/EU.16 Volume effects (i.e., losses due to a drop in sales that follows an increase in prices) are perhaps the most remote, seemingly impossible to ‘pass on’ as such. Nevertheless, any business will arguably wish to cover all costs, and make a profit, through revenue. If so, they will not only take steps to manage costs, but also ensure that all losses are covered by pricing (at least in the long run, as price elasticity of demand may require adjustments of pricing from time to time). This would indicate a wide substantive scope for passing-on argumentation, with no lines drawn that could exclude economic burdens such as loss of profit from the burdens which could be ‘passed on’. Time is also a problem. Assume that a purchaser of products from a cartel company absorbs the loss incurred by reason of having paid an overcharge one year, as fierce competition on its selling market makes passing-on impossible. At the end of the year the overcharge is revealed and the cartel dissolved. The purchaser nevertheless raises its prices the next year, when there is less competition, to compensate for loss sustained the year before. Is this passing-on?

1.025 It is indeed difficult to pinpoint in general terms the character of the event giving rise to an economic disadvantage for the party or parties directly burdened by the initial tort or unlawful transaction. In American passing-on discourse Miller has submitted, as mentioned, that the passing-on problem can arise in the context of any action where the relevant injury has been incurred in the form of increased production costs in a business.17 Pursuing this line of thought, the harmful event may have taken the form of a direct product-related cost,18 the cost object of which was the specific commodity to which an infringement of EU law was distinctly related and which was subsequently sold to the next link in the supply chain (unaltered or as part or raw material for goods or services sold). This would typically be the case if the purchase price of the commodity had been raised and the raising of the price was somehow illegal. A related example which would not, I believe, qualify as a production cost in economic terms is the levying of a charge, duty, tax or other fee on the production, import, selling or the like of specific goods or services. In the gritz and quellmehl cases however the relevant event took the form of unpaid subsidies for the production of certain commodities, i.e., a loss of product-related income, which was subsequently sold to the next link in the supply chain.19 The economic disadvantage suffered and allegedly passed on cannot therefore be pinpointed in terms of whether it was a cost or a loss of income. We must accordingly proceed from Miller’s definition and find a better option for the EU law context.

1.026 A common denominator for the different economic events that have been at issue in EU case law is that they have all constituted an economic disadvantage (instant or deferred) for the party or parties directly burdened by the initial tort or unlawful transaction. All economic events have also been very closely related to a specific commodity: unpaid subsidies for their production, charges on their import, turnover taxes and the like. A similar connection furthermore appears from Directive 2014/104/EU on competition law damages, where the infringement of competition law is understood as having occurred with regard to specific goods or services. In Article 14 of the Directive it is thus expressed that passing-on occurs where ‘goods or services [have been] the object of [an] infringement of competition law’ and the downstream party ‘has purchased goods or services derived from or containing them’.20 A preliminary demarcation of what situations are passing-on situations would therefore be to reserve the possibility of passing-on to economic burdens that are sufficiently closely linked to the commercial handling (production, purchase, transport or the like) of services or goods that are subsequently transferred to the next link in the supply chain (unaltered or as part or raw material for goods or services sold). Such a demarcation would make it legally irrelevant to the passing-on problem that indirect costs and fixed costs can also be said to be included, albeit less directly, in selling prices. It could, e.g., not be argued that an illegal overcharge on concrete had been passed on in the prices for products manufactured in a factory built with concrete that had been subject to the overcharge, or that an increase in insurance costs due to new and unlawful statutory requirements on insurance for the production in the factory concrete had been passed on in the prices for products. There are however more difficult examples. A charge on the import of machinery used for the production at issue would be more closely related to the production costs. Could such a charge be passed on in prices for the products? What if collective action in breach of EU law led to increased labour costs – could they be passed on?

1.027 It is submitted that a definition drawing on the wording used in Article 14 of Directive 2014/104/EU could be useful and would probably yield the most reasonable results. The Directive speaks of economic events connected to ‘goods or services [that have been] the object of [an] infringement’ and the subsequent distribution, to the next level in the supply chain, of ‘goods or services derived from or containing them’. It would follow from these requirements that a sufficiently close link must first be established between an infringement of EU law and an increase in costs for the commercial handling of specific goods or services. Of course, the existence of the infringement of EU law and the increase in costs must also be established. It would then need to be established that goods or services subsequently distributed to the next level of the supply chain were ‘derived from or [contained]’ the aforementioned goods or services. If such a solution were adopted then as a consequence volume effects, such as decreased sales following an increase in selling prices, could not reasonably be argued to have been passed on in the relevant manner. Such a requirement would also, arguably, make charges related to the machinery used in production legally irrelevant to the passing-on problem.21 The potential passing-on of labour costs would however remain open, as labour might be seen as being contained in products. It would ultimately be for the courts and the Court of Justice to determine issues such as these. It is nonetheless submitted that a demarcation of this kind would be sound and useful for courts, and that it would be for the parties to present evidence on the character of the economic event at issue and its relationship to the alleged infringement of EU law.

1.028 By contrast it is submitted that the lapse of time should be of little consequence to the possibility of passing-on an economic burden. Consider the production of matured whisky. If a producer pays a charge on malt which is levied contrary to EU law, and the relevant units of produced whisky are sold ten years later, it is submitted that passing-on should not be excluded by reason of the lapse of time. The relevant criterion should be whether the whisky ‘derived from or [contained]’ the malt.

1.029 This is not however the end of the story – only its point of departure, as the issue of whether or not the cost has in fact been passed on remains. Suffice it here to mention that the crucial points of this next step include the pricing policy of the party or parties who allegedly passed on an economic burden as well as the price elasticity of demand on the relevant market. Was the burden passed on? Could it have been? If the party passed on a burden by increasing its selling prices, did that party consequently suffer a decrease in sales? These are questions to address through evidence. It is for the EU legislature and the EU courts to set the burdens and standards of proof necessary for the purposes of identifying whether passing-on has occurred in the individual case. They have indeed been called upon to do so, and to scrutinize national rules on standards and the burden of proof. The resulting EU law on evidence of passing-on will be examined in the chapters below. This section has served to bring some stability to the discussion by indicating what economic events can plausibly trigger passing-on, and what costs cannot.


1.    Introducing the acronyms

1.030 The purpose of this section is to get acquainted with the parties involved in the passing-on problem and to introduce the acronyms I have chosen in order to refer to them without needing to use lengthy descriptions every time. This section also serves to illustrate the interests of the parties involved in the passing-on problem before making the plunge into legal detail. A number of arguments and possible positions will be mentioned, but for the purposes of this section, they will only be introduced, and accordingly they will not be elaborated upon in detail.

1.031 This book thus uses a simplified model of three persons: Alpha, Beta, and Gamma. Alpha can be the EU, a Member State, or a commercial enterprise. Beta is always a commercial enterprise – otherwise the passing-on problem would not apply. Gamma can be a commercial enterprise too, but can also be a consumer or a Member State – it is of little or no consequence to the analysis.

1.032 A legal dispute in which the passing-on problem arises has usually originated in a money transaction between Alpha and Beta. This transaction may have been the payment of a tax by Beta to Alpha or payment for goods or services sold by Alpha. In some cases the dispute originated instead in the omission of a money transaction from Alpha to Beta.22 Gamma, finally, is the heterogeneous class of natural and legal persons who can be found down the supply chain from Beta (i.e., they are Beta’s customers, customers of Beta’s customers, and so on down to final consumers). It follows that Gamma, as a common denominator for all these, may include both commercial enterprises and individual consumers. Beta and Gamma will both be claimants, while Alpha will be the defendant.23

II.    Alpha

1.033 Alpha obviously would not wish to pay damages or restitution, neither to Beta nor to Gamma.24 Alpha would use all available arguments to prove that no awards were payable at all, to any of the claimants. In relation to Beta, Alpha might argue that any and all injury suffered had been passed on to Gamma and that the passing-on of a loss meant that the loss no longer existed with Beta as an award can be payable only if the loss has been sustained more permanently. If the loss had not been sustained in this sense by Beta, Alpha would contend that it followed that no award was payable to Beta. Indeed, Alpha would submit, an award of damages or restitution to Beta (or Gamma) under circumstances where no loss has been sustained would be a windfall award constituting an unjust enrichment of Beta.25

1.034 In relation to Gamma however, Alpha might instead assert that the entire injury was absorbed by Beta and that no amount whatsoever was passed on to Gamma. Furthermore, if the ‘Gamma’ claimant at issue was not a final consumer, Alpha would also shamelessly argue that even if some part of the injury had been passed on to Gamma, Gamma would continue to pass it further on down the chain of distribution. And so on.

1.035 Under its assertive surface, however, Alpha would fear being held liable to pay multiple awards. Consider the possibility of one court holding that Beta did not pass on any loss, and that an award was therefore payable to Beta corresponding to the entire amount initially paid (but not due) by Beta to Alpha, and another court holding in parallel proceedings that Beta did pass on all its loss to Gamma, and that an award was payable to Gamma claimants corresponding to the entire amount initially paid (but not due) by Beta. Under such circumstances Alpha would be held liable to pay the same amount twice. If the Gamma claimants at issue here were moreover only the first tier of a long supply chain of potential claimants, there might be a risk of even further multiplication of the amount payable in parallel proceedings.

III.    Beta

1.036 Beta would have either: (1) paid money to Alpha that was not due; (2) paid an overcharge to Alpha, i.e., an amount of excess over and above the sum properly due to Alpha; or (3) have been denied a sum that was due from Alpha to Beta. Under any of these circumstances, it would not be likely for Beta to find it relevant whether this economic burden might have been compensated for through prices set by Beta vis-à-vis Gamma. In the first two instances Beta’s lawyer might argue that the situation should be settled by plain restitution of the nominal amount unduly paid to Alpha by Beta, as it was simply a solutio indebiti and should be treated as such. Whatever steps Beta may have taken to adjust to the loss constituted by that payment, the lawyer will argue, are business decisions of Beta of no concern to Alpha.26 In the third instance the argument would be similar although the point of departure would be that Alpha owes Beta money and should stop beating around the bush about it.

1.037 If pushed to scrutinize the passing-on defence in further detail, Beta would disagree with the notion that prices are set by summing up costs and adding a fixed profit margin. Beta would argue that prices are instead set with regard to market prices and Beta’s position on that market – admittedly with a view to making a profit, but profits are never guaranteed in business. There would be times when profits were high, thanks to low costs and high market prices, and times when profits were low (and even at times a loss) due to high costs and low market prices. Thus Beta would challenge the existence of a direct nexus between costs and prices. Beta would accordingly submit that if Beta had not incurred extra costs because of Alpha’s illegal conduct Beta would have enjoyed higher profits, which would serve to demonstrate that the pricing policy of Beta was not cost-based. Under such circumstances, Beta would conclude, Alpha must be made liable to pay a full award to Beta.

1.038 Beta might also challenge Alpha’s line of argument by following it through to its final consequence. Beta might then argue that if it were true that businesses used a fixed profit margin and that the injury had thus been passed on to Gamma, the payment of an award to Beta would consequentially also lead to compensation for Gamma as well as Beta. This would be so because, under the passing-on logic asserted by Alpha, a full award of compensation from Alpha to Beta would enable Beta to lower its prices while nevertheless maintaining its profits – passing-on the award, if you will, and thereby restoring balance to the supply chain.

1.039 Beta might also reject the notion that the payment of a compensatory award by Alpha to Beta in excess of loss actually absorbed by Beta would lead to an unjust enrichment. If anyone had been unjustly enriched in this situation, Beta would argue, it was Alpha. Alpha had profited from violating the law. That profit should rightly be disgorged and transferred to Beta, from whom it was exacted.27

1.040 Finally, however, and notwithstanding these possibilities of legal argumentation, Beta might still be reluctant to bring an action against Alpha for practical business reasons. Suing a business partner might be detrimental to business, especially if Alpha were a very important or even essential partner. Even if Alpha were an exchangeable partner, suing partners does not look good. For such reasons Beta might hesitate to pursue its claim in court.28

IV.    Gamma

1.041 From the point of view of Gamma, it would be clear that any award to be paid by Alpha must be payable to Gamma first and foremost. It was Gamma who ultimately suffered injury due to Alpha’s violation of law, while the injury simply passed through the assets of Beta. Alpha has been enriched, Beta is left at status quo – the actual loss would remain with Gamma, and that loss would be unfair and should be remedied by holding Alpha liable to pay an award to Gamma directly.

1.042 Gamma would, however, face great legal difficulties. Gamma would risk being legally defined as a third party with no direct nexus with Alpha. This could lead to a denial of access to court, leaving Gamma cut off from the possibility of bringing an action against Alpha. Even if Gamma were granted access to court the lack of a direct nexus would still be a problem in view of substantive proximity. In damages it would be difficult to prove causation. In restitution it would probably be even more difficult to leapfrog Beta and prove sufficient proximity between the breach of law by Alpha and the loss incurred by Gamma. To overcome these difficulties it would be crucial that Gamma was able to prove that the injury had indeed passed on from Beta to Gamma. Gamma might find such a burden of evidence difficult: should not such evidence instead be requested from Beta, whose assets this issue concerns and whose pricing policy gave rise to passing-on in the first place?29 By itself Gamma would at best be able to produce general evidence on price elasticity of demand on the relevant market. Such evidence would however be inconclusive with regard to the actual pricing policy applied by Beta in the individual case. And even if passing-on was proved it would remain uncertain whether the court would accept Gamma’s passing-on argument at law.

1.043 Gamma would have little sympathy concerning the issue of Alpha risking multiple liabilities. Gamma would almost certainly argue that whether or not Alpha has paid an award of compensation to Beta, and notwithstanding how that award was determined, Gamma’s prospects of bringing a successful direct action against Alpha should be undiminished. If Alpha violated the law, Alpha will suffer. After all, US antitrust violators pay treble damages. Why should European law violators be better off?

V.    The need for coordination

1.044 It has been demonstrated above that passing-on is not a single legal problem but a specific set of factual circumstances that triggers a number of distinct legal problems: access to court issues, substantive proximity problems, and problems of how to estimate awards. Although solutions could, in theory, be found for each of these problems individually, it is evident from the above that those solutions must somehow be coordinated or the parties will risk severe adverse consequences – e.g., multiple liabilities for Alpha.

1.045 There are a number of suggestions on how to approach these coordination problems, e.g., on how to remove the risk for multiple liabilities and stop Alpha from using contradictory arguments in parallel proceedings. One suggestion is to join the actions at issue for the purposes of establishing that an award is payable per se.30 Another suggestion is to let Beta claim for all harm initially incurred and then let Gamma claimants sue Beta in turn.31 Suggestions such as these will be discussed in the chapters below and in particular in section 9.E.


I.    Purpose

1.046 The purpose of this book is to analyse and to discuss the approach to the passing-on problem that has been chosen in EU law. It must however be immediately stressed that there is not one single and coherent approach in EU law, but a number of approaches with a varying degree of legal detail. It will be discussed in this book if, how and to what extent these approaches are interrelated and compatible with each other. The merit of the approaches chosen will also be discussed. Reasoned suggestions on how to fill gaps will be submitted, and, on occasion, there will be suggestions for revision of the choices made.

II.    Contexts and perspectives

1.047 The heterogeneity of EU law on passing-on is due not only to the aforementioned range of legal issues triggered by passing-on itself, but also to the variety of legal courses of action in which it may arise. In EU law the passing-on problem has been explicitly addressed in three different courses of action: (1) in actions for damages brought against the EU under Article 340(2) TFEU; (2) in actions against Member States for the repayment of charges levied by Member States in breach of EU law; and (3) in damages actions brought against private individuals32 for a breach of EU competition law.33 In actions 1 and 2 an approach has been chosen by the Court of Justice and, in action 3, by the EU legislature. The discussion will also cover other EU law courses of action in which the problem may need to be addressed in the future.

1.048 As the perspective here employed is the EU law perspective, the legal issues addressed will also be embedded in the EU law context. This means inter alia that the law governing the respective actions for damages or restitution will not always be EU law only. Actions brought against the EU are governed by EU law exclusively, and the EU (in particular its courts) has full jurisdiction and responsibility to develop the conditions and criteria that apply to such actions. By contrast, actions for damages or restitution brought against Member States or against a private individual for a breach of EU law are governed by a complex combination of EU law and national law. Although the right or interest infringed will be protected by EU law, such actions must be brought before the national courts.34 With regard to damages actions for a breach of competition law, there is an instrument of EU law harmonizing aspects of such actions before national courts.35 With regard to actions for restitution of charges levied by a Member State in breach of EU law there is no EU law harmonization. Where there is no harmonizing instrument, EU law has traditionally relied on the national courts to provide appropriate remedies and conditions for private enforcement of EU law. This reliance on national law has been subject to an increasingly intricate web of EU law requirements, however, developed to secure the effective judicial protection of rights and interests protected by EU law. It is in this context that much of EU case law on the passing-on problem is found. It importantly follows that the passing-on problem is not only a concern for the EU courts and legislature, but also for their national counterparts. It also follows that the EU law requirements of effective judicial protection are crucial to this inquiry. Furthermore, there is an interesting extra dimension of tension, for in every discussion of the integration of EU law and national law it must be assessed how far the Court of Justice has been inclined to push the impact of EU law on national private law.

1.049 The passing-on problem will consequently be contextualized in three ways. First, in order to assess with full precision the policy options available to the courts seized with a passing-on situation, the three main legal aspects of passing-on (access to court, substantive proximity (e.g., causation in damages), and the estimation of awards) will be distinguished and examined in the context of general observations on those three aspects. Second, as EU law, just as any other legal system, has created specific rules and principles for specific courses of action, it is necessary to carry out the examination in the context of the different courses of action individually. Accordingly, there will be separate examination of the legal aspects triggered by passing-on in the context of damages actions against the EU under Article 340(2) TFEU, repayment actions against Member States, and competition damages actions. Third, all these discussions will remain within the paradigm of EU law. As a result of the contextual approach chosen for this book, many central topics in EU law damages and restitution will be covered.

1.050 The EU law discussed in this book must be distinguished from the comparative law field called European private law. Discussions in the latter field of research have in recent years revolved very much around the grand Draft Common Frame of Reference (DCFR) project,36 which was funded by the European Commission.37 As European private law is comparative law, its findings are academic, providing perspectives on national private law. By contrast, EU law is binding. It forms part of the law of all national jurisdictions of the EU Member States, providing a more or less detailed framework of rules and principles whenever the legal situation at hand comes within the scope of EU law. It has not however been intended to cover the particular problems that may arise in specific national jurisdictions in connection with the operation of integrating EU law and national law. Although problems particular to a certain jurisdiction may be discussed in connection to specific case law, such examinations are only intended to aid the abstract assessment of the interface between EU law and national law, to be carried out from the EU law perspective. This limitation has been chosen from consideration of the heterogeneous reading audience. On the one hand, this approach has the advantage of rendering this book interesting to lawyers of any EU jurisdiction who find themselves preoccupied by the passing-on problem. On the other hand, however, it also has the disadvantage of being unhelpful to all these lawyers in their efforts to understand how particular rules of their national jurisdiction, which may come into play in a passing-on situation, will fit with the EU law framework. This problem however underlies much of the presentation, and it is hoped that some of the case law and hypothetical situations discussed in this book will provide a good start for considerations of problems particular to the national jurisdiction at issue.

1.051 There are three courses of action in which the passing-on problem has arisen, namely:

1.    Damages from the EU under Article 340(2) TFEU,

2.    Restitution from a Member State, and

3.    Damages for a breach of EU competition law.

1.052 There are also some alternative courses of action in which the problem might occur:38 namely

4.    Restitution from the EU,

5.    Damages from a Member State,

6.    Restitution pursuant to a breach of EU competition law, and

7.    Damages and restitutionary actions brought against private individuals for a breach of EU law other than competition law.

1.053 Among these alternative courses of action, 4 and 5 exist in EU case law, 6 also exists in case law but in very particular circumstances, and those in 7 are only being discussed hypothetically. Nonetheless potential passing-on approaches will be discussed, albeit more briefly.

III.    Structure of the presentation

1.054 Part I of this book includes two chapters. In this first chapter the passing-on problem and its characteristics in EU law have been introduced, in a nutshell, alongside the approach chosen for the purposes of this inquiry into the legal issues triggered by passing-on. Chapter 2 will be devoted to introducing the EU law requirements developed to secure the effective judicial protection of rights and interests protected by EU law. These requirements are crucial to the passing-on problem, irrespective of which specific legal aspect of the problem is discussed. Nonetheless there are differences in their scope of application, depending on whether the action is under the exclusive jurisdiction of the EU Courts or should be brought before the competent national court.

1.055 Parts II, III and IV include the main three Chapters 3, 5 and 7 of the book, in which existing EU law on passing-on is discussed in relation to each of the relevant courses of action. Parts II, III and IV also include Chapters 4, 6 and 8 in which alternative courses of action are discussed. These are courses of action where the passing-on problem has not, as yet, arisen. Each chapter begins with an introduction to the action, tracing its most salient features. With regard to the alternative courses of action their interrelationship with the main action will be addressed. Focus then turns to the three major legal issues connected with the passing-on problem, as already mentioned above: access to court; substantive proximity; and estimation of awards.39 Each assessment begins with general observations on the area (e.g., on causation under Article 340(2) TFEU) before proceeding to discuss the existing case law or legislation on passing-on in that particular context.40 The character and length of the respective presentations and discussions will vary considerably, as some have not been touched upon at all by the EU Courts or legislature, while others have been the subject of extensive case law developments. Notwithstanding these differences, it will be consistently endeavoured to find the decisive problems, to address them with attention to case law and legal debate in the context at issue, and to submit reasoned opinions and suggestions where appropriate.

1.056 Comparisons, analysis across contexts, a few conclusions and overall suggestions will follow in the final Part V which comprises only Chapter 9.

1   This was first brought to my attention by Richard A Miller, ‘Federal Antitrust Law: Price Discrimination: Proof and Measurement of Damages in Treble Damages Actions‘ (1962) 60 Michigan L Review 1104, 1124n 66. See Section 1.A.IV for an endeavour to find a more precise understanding of what circumstances will relevantly trigger the passing-on problem.

2   Recognition of the defence of passing-on can follow implicitly from the method used by courts to estimate damages, especially in damages law, without it being considered as an issue at all.

3   Commission, ‘Staff Working Paper Annex to the Green Paper Damages action for breach of the EC antitrust rules’ SEC(2005) 1732 para. 158.

4   Case C-94/10 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet [2011] ECR I-9963. A summary of relevant facts and first holdings is available in para. 5.030, with further holdings summarized in para. 6.042.

5   Case C-557/12 Kone AG and Others v ÖBB-Infrastruktur AG EU:C:2014:1317. A summary of relevant facts and holdings is available in para. 7.035.

6   Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014 on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union [2014] OJ L349/1 Art 12(1).

7   Veljko Milutinovic, The ‘Right to Damages’ under EU Competition Law: From Courage v. Crehan to the White Paper and Beyond (Wolters Kluwer 2010) 242.

8   Case 72/74 Union syndicale-Service public européen and others v Council of the European Communities [1975] ECR 401 para. 21; case 114/83 Société d'Initiatives et de Coopération Agricole et al v Commission of the European Communities [1984] ECR 2598 paras 3–5; and parallel case 289/83 GAARM v Commission of the European Communities [1984] ECR 4295 paras 3–5.

9   Case C-453/99 Courage Ltd v Bernard Crehan and Bernard Crehan v Courage Ltd and Others [2001] ECR I-6297 Operative part para. 3.

10 Akos George Toth, ‘The Concepts of Damage and Causality as Elements of Non-Contractual Liability’ in Ton Heukels and Alison McDonnell (eds), The Action for Damages in Community Law (Kluwer Law 1997) 192.

11 The respective roles of these two interests in the passing-on problem will be the subject of Section 9.E.

12 This ‘passing-back’ problem is acknowledged in Directive 2014/104/EU (n 6) Art. 12(4). See also rec. 38 and Art. 11(4)–(6).

13 E.g., in paras 7.205–7.206.

14 For a similar overview of ‘likely pass-on scenarios’ see Milutinovic (n 7) 183–6. Cf. also the examples provided in Elisabeth Eklund, ‘Indirect Purchasers – Is there Anything New in the Directive? An Introductory Overview of the Current and Future Status of Indirect Purchasers in the EU’ in Maria Bergström, Marios C Iacovides and Magnus Strand (eds), Harmonising EU Competition Litigation: The New Directive and Beyond (Swedish Studies in European Law vol. 8, Hart Publishing 2016) 271–4.

15 It should be mentioned however that the EU Courts have not in their case law delineated the use of passing-on reasoning but have rather been reluctant to make clear demarcations with regard to the circumstances within which passing-on might be considered.

16 Directive 2014/104/EU (n 6).

17 Miller (n 1).

18 I admit I am no economist and I am sure this could be stated with more economic elegance. If the cost at issue has been challenged for being contrary to EU law, it would probably be economically more accurate to speak of deferred costs (becoming direct costs only if the challenge is unsuccessful). I wish to thank my mother-in-law Charlotte Byström, Head of Finance at Linneaus University, for input on economic terminology. Any misunderstandings are my own – and evidence that I should have listened better to my mother-in-law.

19 Case 238/78 Ireks-Arkady GmbH v Council and Commission of the European Communities [1979] ECR 2955; joined cases 261 and 262/78 Interquell Stärke-Chemie GmbH and Diamalt AG v Council and Commission of the European Communities [1979] ECR 3045; joined cases 64 and 113/76, 167 and 239/78, 27, 28 and 45/79 P Dumortier frères SA and others v Council of the European Communities [1979] ECR 3091; and joined cases 241, 242, 245 to 250/78 DGV, Deutsche Getreideverwertung und Rheinische Kraftfutterwerke GmbH and others v Council and Commission of the European Communities [1979] ECR 3017. These cases are extensively discussed in Section 3.E.

20 Possibly an elaborated working definition for the purposes of competition law will be supplied by the Commission in its upcoming Guidelines for national courts on the passing-on of overcharges, expected in 2017. See also Directive 2014/104/EU (n 6) Art. 14 which includes rules on the burden of proof.

21 A situation of this kind was brought before the United States Supreme Court in Hanover Shoe Inc v United Shoe Machinery Corp 392 US 481 (1968).

22 See e.g. the so-called gritz and quellmehl cases (n 19).

23 Actions brought by Gamma against Beta are possible e.g. in Germany (see Anne Howard and Anneliese Blackwood, ‘Litigating Infringements in the National Courts’ in Vivien Rose and David Bailey (eds) Bellamy & Child: European Union Law of Competition (7th edn, Oxford University Press 2013) 1257), but have not been addressed by the European Court of Justice as such. The Court has nevertheless made reference to this possibility in joined cases C-192/95 to 218/95 Société Comateb and others v Directeur Général des Douanes et Droits Indirects [1997] ECR I-165 para. 24; and in case C-94/10 Danfoss A/S and Sauer-Danfoss ApS v Skatteministeriet [2011] ECR I-9963 para. 26.

24 Knowing that Alpha may be facing substantial penal sanctions due to its violation of law, especially in the competition law context, this is understandable.

25 These are all arguments on how to estimate the award (Alpha is arguing, along pragmatic lines, that it should be estimated to zero).

26 For an argument of this kind see e.g. Peter Birks, Unjust Enrichment (2nd edn, Clarendon Law Series, Oxford University Press 2005) 221: ‘Everyone takes steps to adjust to a minus.’

27 See e.g. Francis Hubeau, ‘La répétition de l’indu en droit communautaire’ (1981) 17 Revue Trimestrielle de Droit Européen 442, 451.

28 This is a business concern to be weighed against the possible benefits from pursuing the claim. As it is a business concern it will not be analysed as such in this book. Nonetheless, this aspect is occasionally mentioned in legal debate and will be mentioned again below.

29 In this kind of situation, Beta and Gamma would have little incentive to give mutual aid in litigation against Alpha. It will be discussed below how to design an approach to the passing-on problem where such conflicts of interest among the potentially injured parties may be avoided.

30 See e.g. Antitrust Modernization Commission, ‘Report and Recommendations’ (2007) <> (last accessed 31 May 2016) 277, for a litigation model that would have such an effect.

31 See Howard & Blackwood (n 23) 1256–7.

32 The term ‘private individual’ will be used throughout this text (with the exception of quotes using different terminology). It is intended to serve as shorthand for any kind of natural or legal person capable of carrying rights and/or obligations pursuant to EU law, who is not an emanation of the Member State within the meaning of EU law. The criteria used for identifying legal bodies as emanations of the Member State within the meaning of EU law were laid down by the European Court of Justice in case C-188/89 A Foster and others v British Gas plc [1990] ECR I-3313 para. 20.

33 It deserves to be pointed out that an action for damages or restitution for a breach of competition law may also be brought against an institution, agency or other emanation of a Member State; see e.g. case C-242/95 GT-Link A/S v De Danske Statsbaner (DSB) [1997] ECR I-4449 which was for restitution by reason of a breach of competition law.

34 It is another matter that the national courts have the opportunity, and at last instance the duty, to refer questions on EU law that may arise in such proceedings to the European Court of Justice for a preliminary ruling; see Art. 267 TFEU.

35 Directive 2014/104/EU (n 6).

36 The very physical result of which is Christian von Bar and Eric Clive (eds), Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference (DCFR) Full Edition (6 vols, Sellier European Law Publishers 2009).

37 The Commission called for such endeavours in ‘Communication from the Commission to the European Parliament and the Council: A More Coherent European Contract Law: An Action Plan’ COM(2003) 68 final section 4.1.1.

38 Apart from the actions listed here one might also discuss the importance of passing-on in the context of the duty to repay illegal state aid. However, in that context the circumstances are quite different from the circumstances of the actions discussed in this book, as the defendant is not arguing disimpoverishment ofthe claimant, but arguing disenrichment of itself. The term ‘disimpoverishment’ as an option to ‘passing-on’ has been suggested for the purposes of restitution law by Birks (n 26) 219–21; and Michael Rush, The Defence of Passing On (Hart Publishing 2006) 11–17.

39 The choice of approaching each EU law theme in the context of the specific courses of action inevitably causes some repetition. However, every other sequence considered for this presentation either resulted in even more repetition or risked rendering the discussion unintelligible.

40 In Chs 4, 6 and 8 the discussion will be speculative as there is no case law and no legislation.