1Introduction

Practitioners of EU law are used to focussing on the way EU law influences the judicial protection in the Member States. Generally, less attention is paid to the way national systems of judicial protection might influence the case law of the Court of Justice. The action for damages enshrined in article 340(2) of the TFEU is an example of how this second type of influence can take place. This article allows a party to bring a case before the European Union Courts, in order to obtain compensation for damage resulting from actions of one of the institutions. This type of liability does not directly influence the national regime of compensation for damages. The influence is even exerted the other way around, since the Court of Justice has to base the liability regime of the European Union ‘in accordance with the general principles common to the laws of the Member States’. In practice, this means that the applicant must show the existence of a wrongful act, actual damage and a causal link. There is no question that liability for wrongful acts is ‘common’ to the laws of the Member States.

The question of whether EU law should also accept a regime whereby liability exists in the absence of a fault, has been significantly more controversial. The Court of Justice has, for a long time, hinted at the existence of such a type of liability. This would mean that the institutions can be forced to pay compensation when an, in itself legal, act leads to ‘special’ and ‘abnormal’ damage. These conditions can be traced back to the principle of ‘equality before the public burdens’, which will be discussed in more detail later on. Although art. 340 TFEU does not explicitly rule out the possibility that the Union can be held liable in the absence of a wrongful act, in the FIAMM decision of 9^th of September 2008, the Court brought all speculation concerning the existence of a separate regime concerning no-fault liability of the European Union to an end. The Court of Justice ruled that ‘[a]s Community law currently stands, no liability regime exists under which the Community can incur liability for conduct falling within the sphere of its legislative competence in a situation where any failure of such conduct to comply with the WTO agreements cannot be relied upon before the Community courts.’C-120/06P and C-121/06P Fabbrica italiana motocarri Montecchio SpA (FIAMM) a.o. v. Council and Commission [2008] ECR I-6513.

This decision was met with fierce criticism. Authors have mainly been critical about the outcome of this case for the applicants in the FIAMM case. In this article, a more general analysis is made of the pros and cons of accepting no-fault liability claims against the European Union. This analysis is made against the background of the FIAMM decision, since this decision offers a good insight as to why the Court of Justice currently refuses to recognise the possibility of such a regime. For this purpose, I will first give a brief outline of the facts concerning the FIAMM case, the decision of the Court of First Instance and the Court of Justice (section 2). Section 3 will deal with the meaning of the principle known as égalité devant les charges publiques (‘equality before public burdens’, or, in other words, the equality of citizens in bearing public burdens) – the most important basis for compensation for damages – in a general sense, partly on the basis of Dutch and French case law. Next, I will deal with its potential added value to EU law. To this end, I will first examine whether, from a comparative law perspective, the principle of equality before public burdens (hereinafter called the ‘principle of equality’) may be meaningful at the level of the European Union (section 4). After some legal policy observations in section 5, section 6 focuses on the options for compensation for damages left at the level of the EU after the FIAMM judgment.

2Case Law of the Courts Regarding No-Fault Liability

Up until the FIAMM decision, neither the Court of First Instance nor the Court of Justice had ever explicitly ruled out the possibility of claimants putting forward a successful claim in damages despite the damaging act being lawful. However, both Courts have also never explicitly recognised that art. 340 TFEU (formerly art. 288 EU Treaty) implies such a possibility. Instead, for a long time both Courts opted for a conditional approach. This approach is illustrated by the Dorsch Consult case. In this case, a German firm of consultant engineers had been contracted to perform services for the Ministry of Housing in Iraq. When Iraq invaded Kuwait in 1990, the United Nations Security Council imposed an embargo on trade with Iraq and Kuwait, preventing trade by the Union with Iraq and Kuwait. The Iraqi government reacted by freezing all property and assets, which it owed to the states that adopted the embargo. This was harmful to Dorsch Consult, a company that still held outstanding debts with the Iraqi government, which it now could not retrieve. Dorsch Consult claimed compensation from the Union, reasoning that as a result of the trade embargo, it had foregone large sums of money that were still owed to it. The Court of Justice, following in the steps of the Court of First Instance, took the opportunity to stipulate which conditions such a claim in any case had to meet. The Court stated that:

‘If the Community is to incur non-contractual liability as the result of a lawful or unlawful act, it is necessary in any event to prove that the alleged damage is real and that a causal link exists between that act and the alleged damage. Secondly, with respect to the Community’s liability in respect of a lawful act, the Court noted that it was clear from the relevant case-law that, in the event of such a principle being recognised as forming part of Community law, a precondition for such liability would in any event be the existence of “unusual” and “special” damage.’Case C-237/98 Dorsch Consult v. Council and Commission [2000] ECR I-4549; [2002] 1 CMLR 41.

The Court went on to state that Dorsch Consult did not meet the ‘unusual’ requirement, since Iraq had to be regarded, even before the invasion of Kuwait, as a ‘high-risk country’. The damage alleged by the appellant could not therefore be regarded as exceeding the economic risks inherent in operating in the economic sector concerned.

This approach, in which the Court even went so far as to test whether the conditions for no-fault liability were adhered to, but refused to answer the preceding question whether this type of liability is actually accepted, was met with criticism. For example, Tridimas stated that this approach is not satisfactory, since it puts the cart before the horse and is liable to maintain uncertainty.T. Tridimas, The General Principles of EU Law (Oxford 2006), p. 495. From this perspective, the decision in the FIAMM case was very welcome, ending all speculation although maybe not offering all certainty that some commentators might have wished for. The FIAMM case was comparable to the Dorsch Consult case since the damage was also the direct result of a state taking retaliation measures. In this case however, these measures were a reaction by the United States to Council Regulations found to be incompatible with WTO agreements. The United States decided to increase the customs duty that FIAMM (and some other companies) had to pay when exporting their products to the United States, irrespective of the fact that FIAMM had nothing to do with these breaches, which took place in the banana industry. The Court of First Instance affirmed the standing jurisprudence that even though the Dispute Settlement Body had held that the Community Legislation was incompatible with the WTO agreements, this could not give rise to liability on the part of the Community on account of unlawful conduct by its institutions, since the WTO agreements are not in principle among the rules in the light of which the Community courts review the legality of action by the Community institutions. According to the Court of First Instance, this limitation did not stand in the way of the possibility of FIAMM to claim damages in absence of unlawful conduct. For the first time, the Court of First Instance recognised that there was such a possibility, which could require the Community to make good the damage suffered by operators who have to bear a disproportionate share of the burden resulting from such conduct. The Court of First Instance nevertheless concluded that in this case the conditions governing such liability were not satisfied because, in the light of the normal hazards of international trade, the damage suffered by the applicants was not unusual in nature.

The Court of Justice however, stated that the Court of First Instance erred in law by affirming the existence of a regime providing for the non-contractual no-fault liability of the Community. In this regard, it put forward two main arguments. First, it stated that although the principle of Community liability where one of the institutions has acted unlawfully is an expression of a general principle familiar to the legal systems of all the Member States, by contrast no such convergence of the Member States’ legal systems has been established regarding the existence of a principle of liability in the case of a lawful act by a public authority, in particular where that act is of a legislative nature. Secondly, the Court pointed out that the legislative context is characterised by the exercise of a wide discretion, which is essential for implementing a Community policy. Therefore, the EC cannot incur liability unless the institution concerned has manifestly and gravely disregarded the limits on the exercise of its powers.

The FIAMM decision has left many commentators wondering why the Court came to such a ‘harsh’ decision.See, f.e., Coutron, ‘Responsabilité pour faute et responsabilité sans faute en droit communautaire. Les approximations de l’arrêt FIAMM. Note sous CJCE 8 septembre 2008, FIAMM, aff. jointes no. C-120/06 et C-121/06’ [2009] Revue française de droit administratif  329, 340-341 and G.A. Zonnekeyn, ‘De “onschuldige” exporteurs in de kou: geen schadevergoeding voor schending van WTO-recht’ [2008] NTER 363-370. They have also pointed out that both arguments mentioned by the Court are not convincing. Before dealing in more detail witht this matter, I will first reflect in a more general way on the equality principle.

3General Observations on No-Fault State Liability

3.1Introduction

In order to analyse whether any added value can be expected from the recognition a no-fault liability regime by the European Courts, it is useful to offer a more general overview of this type of liability. Which principle(s) underpin(s) this type of liability? Which conditions does a successful claim have to fulfil and how are these conditions generally interpreted? Last but not least, to what situations does this type of liability apply?

Where there is, generally speaking, communis opinio on the question of whether and on which conditions authorities can be held liable when acting unlawfully, for a long time there was no such clarity when dealing with no-fault liability. This was also pointed out by the Court of First Instance, which accepted a regime of no-fault liability, despite the fact that ‘national laws on non-contractual liability allow individuals, albeit to varying degrees, in specific fields and in accordance with differing rules, to obtain compensation in legal proceedings for certain kinds of damage, even in the absence of unlawful action by the perpetrator of the damage.’Case T-69/00 FIAMM v. Commission and Council [2005] ECR II-5393. This did not, however, stop the CFI from bringing this type of liability within the scope of art. 340 TFEU (then art. 288 EU-treaty). The Court of Justice took a somewhat less obliging view when stating that there is in ‘no way’ convergence of legal systems in the establishment of a principle of liability in the case of a lawful act or omission of the public authorities, in particular where it is of a legislative nature. There is something to be said for both views, although the question remains whether a lack of coherence on the national level should have repercussions on the level of the EU, as will be argued later on in this article. For this moment, it should be pointed out that, indeed, not a lot is clear on the ratio, meaning and scope of government liability in absence of fault.

This does not mean, however, that we should leave it at this. In fact, when looking at the way no-fault liability has developed in some Member States, it is quite possible to gain a better grip on this type of liability and its conditions. Especially the state liability regimes of France and Germany are apt for a further study on the meaning and scope of no-fault liability. In Germany, the administrative judges have developed the concept of the Sonderopfer, which allows for civilians and companies to claim damages from the government when their property rights are infringed (‘opfer’) in an exceptional way (‘sonder’) by an act in the general interest. A similar concept has been developed at the beginning of the 20^th century – to be more precise: the Couitéas-decision of 1923 was the starting point – by the French Council of State (Conseil d’Etat). Like its German counterpart, the principle of equality before the public burdens allows only for exceptional damages to be compensated. The damage has to be both ‘special’ and ‘abnormal’ in order to obtain compensation.

Both principles have been referred to by parties in cases before the European Courts; the same goes for different advocate generals in their opinions.See for a reference to the Sonderopfer-principle f.e. Joined Cases 54 to 60/76P Comté de Lohéac v. Council and Commission [1977] ECR 645, para. 19), and the Opinion of AG Reischl (Jur. 1977, p. 645 et seq., on p. 666). See for the equality principle Case 59/83 Biovilac v. Council andCommission [1984] Jur. 1984, p. 4057, para. 28, Case 26/81 Oleifici Mediterranei v. Council and Commission [1982] ECR 3057, paras 26-27) and the Opinion of AG Verloren van Themaat in the last case (Jur. 1982, p 3089-3090). In his opinion on the FIAMM case (Case C-120 and 121/06), FIAMM a.o. va. Commission and Council [2008], para. 62-63, AG Maduro pointed at both principles. Hereafter, I will concentrate on the French case law, since this case law has also influenced the case law in other countries (such as the Netherlands, Luxembourg and Belgium). Furthermore, most authors assume that in the context of EU law, too, the most obvious solution would be to base any non-contractual no-fault liability on the principle of equality.Van Casteren, Schadevergoeding voor rechtmatig EG-optreden (University of Nijmegen 1997), and the previous footnote. French law therefore seems to be an apt starting point for a further elaboration on the characteristics of no-fault government liability.

3.2Origins of the Principle of Equality as the Legal Basis for Liability

To put it briefly, the equality principle requires that excessive burdens resulting from actions that public authorities undertake in the general interest should not be borne by a limited group of citizens but by society as a whole.See also K.M. Scherr, ‘Public liability for administrative acts under French law’, European Public Law 2008. The development of this doctrine can be traced back to the French Revolution and the ideas of equality rooted in it.See in more detail S. Caporal, L’affirmation du principe d’égalité dans le droit public de la révolution Française (1789-1799) (Paris 1995). The principle of equality in French administrative law was developed exlusively by the French Council of State (Conseil d’Etat) in its case law. As early as the beginning of the 20th century, the Council of State realised that the French system of state liability would be unbalanced if it merely provided for the possibility of holding public authorities liable by reason of an unlawful act. Since there are some barriers to this type of liability – particularly where administrative authorities have discretionary power – a system of liability limited to unlawful acts could mean that many victims of detrimental public acts would not have an opportunity to obtain any relief. Hence, to a great extent, resorting to a form of objective state liability was a compromise. By awarding compensation for abnormal damage or loss without insisting that the public action had to be unlawful in some cases, the Council of State allowed the government to come off unscathed. Furthermore, victims of decisions taken by a public authority that exercised discretionary power or of hazardously negligent administrative actions could be compensated.P. Amselek, ‘La responsabilité sans faute des personnes publiques d’après la jurisprudence administrative’, in: Recueil d’études en hommage à Charles Eisenmann (Paris 1977), p. 233-262, at p. 257: ‘[E]n se référant, non pas au fondement de la faute, mais à celui du dommage anormal, le juge parvient tout à la fois à sécourir la victime et à ménager la personne publique en cause’. It is generally assumed that considerations of equity in particular prompted this development: if the court considered this fair, it resorted to one of the grounds within the no-fault liability concept (responsabilité sans faute) to justify its decision to award compensation.Amselek 1977, p. 256 qualificies this as ‘une jurisprudence d’inspiration charismatique’ en ‘une politique jurisprudentielle d’équité (…)’. In this context, Harlow characterises the equality principle as an ‘expression of social solidarity’.C. Harlow, ‘Rationalising administrative compensation’ [2010] Public Law, 334.

3.3Significance and Scope of the Principle of Equality

Even in French legal literature, there is a lack of clarity on the significance and scope of the principle of equality.See f.e. T. Debard, ‘L’égalité des citoyens devant les charges publiques: fondement incertain de la responsabilité administrative’, Rec. Dalloz 1987, p. 157-163 and C. Harlow, ‘The current state of state liability’ [2002/4] Rivista trimestrale di diritto pubblico 915–33 (at 932). For example, some authors take the view that the principle of equality explains not only why the government may be liable for lawful acts, but it even justifies the existence of state liability as a whole.See f.e. P. Duez, La responsabilité de la puissance publique (en dehors du contrat) (Paris 1938). Whenever such damage or loss arises, whether the conduct concerned involves any fault or not, this results in a ‘public burden’ (charge publique), which imposes a heavier burden on the victim than on others and for that reason, it requires the state to pay compensation. It should be noted that in practice French administrative law does not follow this view. If the Council of State holds the state liable for an unlawful act, it does not invoke the principle of equality in doing so. The scope of application of this principle has been confined to situations in which the state acts lawfully.

But even within that – extremely broad – field, the equality principle is not always applicable. As is shown by the relevant case law, the application of the principle of equality is reserved to specific kinds of acts: individual or general decisions of an administrative authority (décisions resp. règlements), factual action (in the context of the construction of public works, for example) and acts whose creation or approval requires parliamentary involvement, i.e. Acts of Parliament and treaties. If the damage or loss arises from conduct that is lawful but hazardous, the French Council of State applies the theory of risk as a legal basis. An example is the case where a bystander was wounded by a bullet from a firearm used by the police in the lawful execution of its tasks.CE 24 June 1949, Rec. 307 (Consorts Lecomte). If the question concerning the significance and scope of the principle of equality is to be answered properly, a more detailed look at French law is required.

The case law and the legal literature show that three key conditions are crucially important to liability based on equality. Before dealing with these, it should be noted that the principle of equality requires the damage or loss to be caused by a public authority; accordingly, the principle cannot constitute a basis for any damage or loss caused by external events, such as floods or natural disasters or if a private person inflicts damage on another person. In addition, the damage should be caused by an act; the case law shows that this concerns most of the time a written decision, although it might also concern a non-legal act, such as the lawful reconstruction of a road. It is difficult to conceive that the refraining from action might lead to a successful appeal on the equality principle.In the same sense: Van Casteren, Schadevergoeding voor rechtmatig EG-optreden (Deventer 1997), p. 240. This might be different when a public authority decides to refrain from action, but in such a case, the damage is strictly speaking inflicted by the act itself.See for examples in the French jurisprudence where public authorities decide not to let the police make an ending to the occupation of ports or pieces of land (see f.e. CE 30 november 1923, Rec. p. 789 (Couitéas) and CE 22 June 1984, D. 1986, p. 29 (Soc. Sealink).

A  The Damage Must Result from an Act Containing a Lawful Balancing of Interests

The presence of an act is not enough, however. In addition, the loss-causing conduct must be lawful. This condition is usually considered equivalent to the condition that the act must be performed ‘in the general interest.’As exemplified by the description of no-fault liability in Recommendation nr. R (84) 15 of the committee of ministers tot member states relating to public liability: ‘(…) reparation should be ensured if it is manifestly unjust to allow the injured person alone to bear the damage, having regard to the following circumstances: the act is in the general interest, only one or a limited number of persons have suffered the damage and the act was exceptional or the damage was an exceptional result of the act.’ When viewed from the perspective of the French administrative courts and their case law, this definition is not quite adequate. For example, in France the theory of risk is also applied as a basis for liability, in which case a public authority may be liable if activities undertaken by public authorities, even conducted without fault, create an increased risk to society.Scherr 2008, p. 220. Here too, it is safe to assume that such activities are undertaken in the general interest – i.e. are lawful – but that the principle of equality cannot be used as a basis for compensation. Accordingly, for hazardously negligent acts, the administrative court uses the risk theory as the basis for compensation.L.N. Brown & J.S. Bell, French administrative law (Oxford 1998), p. 195-196.

What then are the ‘characteristic features’ of the concept of equality before public burdens? What do the acts, for which a claim based on equality can be made in French administrative law, have in common exactly? Several authors have pointed out that the principle of equality in this context assumes that the act causing damage or loss includes a volitional component.Zie f.e. G.C.A. Henriot, Le dommage anormal (Paris 1960), p. 54, who states that the equality principle applies to situations where ‘une intention directe’ is present. He describes the concept of a ‘charge publique’ as ‘un élément inhérent à l’activité administrative, au fonctionnement des collectivités publiques, élément constant, inévitable, prévu’ (p. 52). See also Chapus and Delvolvé, as stated below. The damage cannot be a mere incidental consequence of the act, but is the inevitable consequence of an act which leads to the result that some people’s interests are consciously sacrificed in the general interest. In this context, Delvolvé states that there must be an end-means relationship between the damage and serving the general interest: the principle of equality applies if the result the government intended to achieve with its lawful act can be attained only by inflicting damage on a citizen.P. Delvolvé, Le principe d’égalité devant les charges publiques (Paris 1969), p. 352: ‘Puisque la mesure dommageable est incontestable, les conséquences qu’elle entraîne directement doivent être considérées comme nécessaires a l’intérêt général; elles s’imposent dans ce but.’ This damage should arise as a necessary or – somewhat broader – inevitable consequence of the lawful act.See C. Debbasch & F. Colin, Droit administratif (Paris 2007), p. 515: ‘Le dommage n’a pas de caractère accidentel, il est la conséquence inévitable des mesures prises pas l’administration  dans l’intérêt général.’ Vgl. L. Romermann, Aufopferungshaftung in Europa (Tectum Verlag 2007), p. 92: ‘Statt desen wird die Kompensation all jener Schaden erfasst, die als unweigerliche Folge des Verwaltungshandelns auftreten und als solche sicher vorhersehbar gewesen sind.’

As a rule, the damage will therefore be the foreseeable consequence of the act, while by contrast, the damage for which compensation may be granted under the risk theory is nonrecurring and unpredictable in nature. This is not the case under the principle of equality. Under this principle, serving a particular interest (preventing animal diseases, improving access to the city centre, etc.) that arises from the process of balancing interests entails inflicting damage or loss.M.K.G. Tjepkema, Nadeelcompensatie op basis van het égalitébeginsel (Deventer 2010), p. 202. For the principle of equality to be applicable, the damage or loss must be the intended, or at least, the foreseenconsequence of the lawful balancing of interests. Chapus formulates this as follows:

‘The principle of equality relates to situations where the damage/loss is not of an occasional nature, as opposed to situations where liability based on risk is relevant. It does not concern any damage/loss that arises from an unfortunate combination of events that have occurred even though they could just as well not have occurred. It concerns damage/loss that is the natural, necessary and foreseeable result of specific situations or specific measures, as a consequence of which the interests of specific members of society are being ‘sacrificed’ to the demands of the general interest.’R. Chapus, Droit administratif general – Tomes I et II (Paris 2001), p. 1364.

It is still not clearly established whether the other Member States have adopted this interpretation of the equality concept as well. It is mainly French administrative law that is strongly characterised by differentiation in the principles underlying no-fault liability; in the other Member States, this is the case to a much lesser extent. Nevertheless, it can be concluded that the above description of liability based on the principle of equality is compatible with the way this principle works in other countries. An example taken from the Dutch case law may clarify this point.Supreme Court 18 January 1991, Nederlandse Jurisprudentie 1992, 638 (Leffers). In 1985, the Minister of Agriculture issued a ministerial regulation designed to eliminate the feeding of pigs with kitchen and slaughterhouse waste (‘swill’). These food scraps presumably caused an infectious disease among pigs, which the Minister wanted to put an end to as soon as possible. An inherent, inevitable consequence of this regulation was that farmers who fed their pigs on swill suffered huge financial losses: they had to change their business operations in such a way that pigs could from then on be fed in other ways as well. Moreover, the regulation meant that investments that had been made for this type of feeding (which had to be cooked in special boilers) were worthless. In particular, farmers who had based their business operations on swill feeding – and who were therefore not able to change over to another type of feeding – suffered such immense losses that they were entitled to compensation, according to the Dutch Supreme Court. This Court ruled that compared to his competitors, Leffers had been hit disproportionately heavily by the ministerial regulation. Further, the substantial adverse consequences resulting from this measure, which had been implemented overnight, could not be attributed to normal entrepreneurial risks. Although the principle of equality was not explicitly invoked, it is generally believed that the Supreme Court was guided by this principle.

The Leffers judgment shows that it is possible to apply the principle of equality to situations where the damage or loss – in this case, the loss suffered by pig farmers who fed their pigs on swill – is an inevitable and foreseeable consequence of a measure that serves the general interest. Based on this example, it can also be proved by means of Chapus’ statement that the principle of equality requires that the damage or loss is a necessary consequence of the lawful public action: it is evident that the purpose the Minister intended to pursue – to put an end to a potential source of spreading of African swine fever – could not be achieved without inflicting damage on the pig farmers.

B  The Special Burden

If the first condition is satisfied, this means only that the principle of equality is applicable. This by no means implies that the principle has been violated too. For that to be the case, two additional conditions must be satisfied. First, the damage or loss must be special. This element requires that the damage or loss should not affect too large a group of citizens. The damage or loss must be extreme, comparatively speaking, which means in this context: compared to other citizens in a comparable position.

It is not always easy to determine exactly with which other citizens the comparison must be made. Is it, for example, conceivable that all undertakings that suffer damage or loss as a result of a statutory regulation are hit disproportionately heavily, compared to all undertakings that do not come within the scope of the regulation? Or is a stricter approach required, which involves a comparison of the undertaking affected with others that are also affected by the action concerned, albeit less heavily? An analysis of the case law of the various national courts reveals that this latter approach is indeed the more customary. Usually, the reference group, i.e. those undertakings that have had to accept the damage or loss without any compensation, concurs with the group for which the measure had adverse effects. This means that as a rule, an undertaking will argue in vain that it has suffered a disproportionate disadvantage compared to undertakings outside its sector that have not been affected by the rule that has caused the damage or loss (and to which this regulation does not apply). This is also illustrated by case law of the Courts, who have applied the special criterion in several cases.See f.e. the decisions Case T-196/99 Area Cova v. Council and Commission [2001], Jur. 2001, p. II-3597, para. 167 and Case T-184/95 Dorsch Consult v. Council and Commission [1998], Jur. 1998, p. II-667, para. 28. Accordingly, if any damage is caused by a statutory measure that applies to a well-defined and determinable group of persons or undertakings, it is a rule of thumb that the group to which the measure is applicable has not been specially affected. They constitute the ‘reference group’, i.e. those who will have to accept the damage or loss without any compensation. Only citizens or undertakings that are able to distinguish themselves from the others within that large group may claim compensation. In this context, they will have to prove that the factors in respect of which they are distinct from the reference group do not provide any objective and reasonable justification for an unequal burden. This could concern the nature of the activity affected: the Leffers judgment shows that companies that specialise in some activity and are then confronted with measures that are introduced overnight and that impose far-reaching restrictions on the exercise of that very activity may in certain circumstances claim that they are suffering damage or loss to a disproportionate extent.

C  The Abnormal Burden

In practice, the most important condition, which proves to be a high barrier in many cases, is that the burden must also be abnormal. Usually, this condition is considered equivalent to the question of whether the damage or loss is part of the ‘normal societal risk’ or the ‘normal entrepreneurial risk’. This condition too, requires that the burden on the person affected should be compared to other burdens, but here the comparison is made at a more abstract level: in the end, it comes down to the question whether the damage or loss is to be considered among the risks that are inherent to life in a community together with others, or to the operation of a business in a specified sector. For example, this involves an examination of the question of whether the injured party had foreseen or couldhave foreseen the relevant damage or loss. Another important factor is the extentof the damage or loss: even though citizens and undertakings must accept damage/loss they suffer as a result of lawful public actions to a great extent, they are entitled in certain circumstances, to claim compensation if they are confronted with excessive burdens.

A few examples may lend colour to these somewhat abstract notions. For instance, the Dutch Supreme Court has ruled that in the case of the trade in animal products, ‘there will sooner be an objectively foreseeable risk of diseases and government measures taken as a result thereof for the purposes of public health or otherwise’.Supreme Court 20 June 2003, Administratiefrechtelijke beslissingen 2004, 84 (Harrida). In this context, another Dutch court, the Trade and Industry Appeals Tribunal, refers to the ‘basic risk’ of the outbreak of an infectious disease, which must, as a general rule, be borne by the entrepreneur, rather than the state.CBB 29 februari 2000, AB 2000, 206. Compare Conseil d’Etat 22 februari 2002, Rec. p. 52 (Michel): animal diseass formulering a risk which  should in principle be borne by those who have the right to hunt. With respect to damage resulting from traffic measures too, it is established case law that such loss should as a general rule be considered part of the normal societal risk. According to the Administrative Jurisdiction Division of the Dutch Council of State, anybody may be confronted with such damage/loss, which means that such loss can as a rule hardly be called abnormal. The French administrative court too, adopts the position that damage/loss due to changes in the infrastructure is part of the ‘risks the consequences of which cannot be regarded as abnormal’.CAA Lyon 8 February 1993, AJDA 1993, p. 706.

A risk attached to this criterion is that the requirements relating to abstract notions such as ‘normal’ and ‘foreseeable’ are stretched to such an extent that any possibility of obtaining compensation is rendered futile in advance, because even the most serious forms of damage or loss are not completely unforeseeable. A correct interpretation of this criterion should leave some scope for taking individual circumstances into account. The relevant case law shows that one of the reasons for concluding that specific damage or loss transcends normal entrepreneurial risks may lie in the way in which the measure has been implemented. For example, the swill measure from the Leffers judgment as such was not unforeseeable, but the sudden manner in which it had been implemented was unforeseeable. According to the Dutch Supreme Court, entrepreneurs trading in animal products cannot be expected to make sure ‘that they are able to change over to a different method of operation, which is not affected by the relevant prohibition, overnight, without the profitability of their company being lost.’Supreme Court 20 June 2003, Administratiefrechtelijke beslissingen 2004, 84 (Harrida).

Accordingly, the foreseeability test is comprised of two stages: the court has to examine not only whether the measure as such is foreseeable, but also whether the manner in which the relevant measure has been implemented is foreseeable. This could for example, concern the date on which the measure enters into force: when in August 1980, French offshore fishermen occupied all French harbours, all kinds of undertakings that were dependent on harbour activities suffered losses. The authorities adopted the lawful decision not to issue an order to vacate the harbours. The French Council of State took the view that transporters of persons and cars from Britain to France were affected disproportionately heavily, and it took into consideration that the losses concerned were sustained in the summer period, which was important for these undertakings. They suffered a loss that was special (compared to the burdens that fell on others that were affected by the government measure) and abnormal (since the relevant companies’ revenues largely depended on transport in the summer period).CE 22 juni 1984, Rec. p. 246, D. 1986, p. 29-30 (Soc. Sealink; ‘un préjudice dont la spécialité et la gravité ont été suffisantes pour qu’il soit regardé comme excédant les charges que les usagers du port doivent normalement supporter’).

4Comparative Law Observations

After these general reflections on the principle of equality, it is time to focus on the liability of the EU again. The manner in which the Court of Justice of the European Union (CJEU) derives arguments from comparative law in developing EU-related liability – pursuant to art. 340 TFEU or in the context of Member State liability – has invited critical comments on several occasions in the past. For example, on the subject of Member State liability, it was argued that the CJEU had suggested that this type of liability also found some support in the law of the Member States. For some countries though, the Francovich doctrine in any case was incompatible with the existing system of state liability, which did not allow for an action for damages from unlawful actions of the legislature.France is a good example. See D. Poyaud, ‘Le fondement de la responsabilité du fait des lois en cas de méconnaissance des engagements internationaux’, RFDA 2007, p. 525-534.  With respect to art. 340 TFEU too, it may be said that liability for unlawful legislative action by no means has a clear counterpart in every Member State. Even so, this did not prevent the ECJ from accepting liability for unlawful legislative measures of the institutions in the case of a sufficiently serious breach of a superior rule of law.

This example alone makes it clear that it is difficult to require a great extent of convergence in the development of art. 340 TFEU.Compare Beatson & Tridimas, New directions in European public law (Oxford 1998), p. 16: ‘[Liability] is not based in principles common to the laws of the Member States (…). [W]ith regard to state liability arising from acts of the legislature there are no general principles which are truly common to the Member States.’ For this reason, it is surprising that the ECJ attaches so much value to the small degree of convergence in its refusal to accept a right to damages compensation at the level of the EU. According to the ECJ, there is ‘convergence of those legal systems in the establishment of a principle of liability in the case of unlawful action or an unlawful omission of an authority, including of a legislative nature, [but] that is no way the position as regards the possible existence of a principle of liability in the case of a lawful act or omission of the public authorities, in particular where it is of a legislative nature.’ In the absence of a clear right to compensation for damages in the legal systems of the Member States, the ECJ would have been well-advised not to accept this right.

Certainly where comparative law is so important a factor, the Court might have been expected to address the question of to what extent this type of liability is shared by the legal systems of the Member States in more detail, thus offering a stronger factual basis to its decision. Besides, as is shown by their arguments, the other parties involved in the FIAMM case sometimes play fast and loose with the facts.See the statement of the Council, the Commission and the Kingdom of Spain stating that in only two countries the possibility of liability for legislative action is recognized (Opinion AG Maduro, para. 54). This does in any way not take note of the Leffers-case, which, as stated above, involved a lawful legislative act. The purpose of this section is to comment on the ECJ’s judgementfrom a comparative law perspective.

4.1Incorrect Use of the Comparative Law Method

The ECJ’s comparative law ground has been criticised in literature.See f.e. L. Coutron, ‘Responsabilité pour faute et responsabilité sans faute en droit communautaire. Les approximations de l’arrêt FIAMM’, RFDA 2009, p. 329 a.o, Tjepkema 2010, p. 814 et seq. It has been rightly pointed out that in developing Art. 340 TFEU, the ECJ should not be exclusively guided by a quantitative approach.See f.e. W. van Gerven, ‘Taking Art. 215 (2) EC Seriously’, in: J. Beatson & T. Tridimas, New directions in European Public Law (Oxford 1998), p. 44-45. It goes too far to only consider a legal principle ‘common’ if it is recognised as such in all Member States: with that approach, a few dissident Member States would have a power of veto.See for further reading F. Fines, Etude de la responsabilité extracontractuelle de la communauté économique Européene (Paris 1990), p. 35. For the same reason, it is not right to allow the opinion of the majority to be decisive, as this could result in the Court having to derogate from current case law if new Member States, whose laws differ in this respect, join the Union.

The only method that reflects the diversity of the various legal systems of the EU Member States and that is sufficiently flexible to be used as a basis for developing Art. 340 TFEU at the level of the EU is the theory of ‘value-based comparative law’. According to this theory, the ECJ should examine which of the key concepts identified in the Member States is best in line with Union law. In the end, the decisive question is whether a specified legal concept is consistent with the system and the special structure of the European legal order rather than simply following the Member States. According to this theory, a liability regime must only be rejected if it is not in line with the system and the special structure of the European legal order at all.Van Casteren 1997, p. 54. In his opinion in the FIAMM case, advocate-general Maduro defended this approach.

‘The decisive question is whether such a solution would best meet the specific needs of the Community legal system. That does not mean that the principle of public authorities’ liability for a lawful legislative act should be enshrined in the Community legal order inasmuch as it would appear to be the best legal solution, and as such must be adopted in the Community legal order.’Opinion in Case C-120/06, para. 56.

According to Kapteyn et al., it is up to the ECJ, at the end of the day, to examine the systems of the Member States and elements that contribute to ‘a fitting, fair and productive solution to the issue of Community liability’.P.J.G. Kapteyn, L.A. Geelhoed, K.J.M. Mortelmans (red.), Het recht van de Europese Unie en van de Europese Gemeenschappen (Deventer 2003), p. 388. Below, in section 5, I will argue that the question of what exactly amounts to ‘fitting and fair’ under EU law has a strongly political dimension. This is why I will leave this point for now.

4.2Solid Basis for a Right to Compensation for Damages in the Member States

Even if the quantity-based approach is replaced by a quality-based approach, it is clear that a solid basis for a right to damages compensation in the case of legislative and other acts in the Member States might well be a strong argument for accepting a right to damages compensation at EU level. To what extent is there a legal basis in the Member States for a right to compensation for damages? It can hardly be denied that the ECJ was right to observe that there is not a great deal of convergence between the various legal systems on this issue. For example, a general right to compensation for damages, whether statutory or extrastatutory, is not recognised, in any case, in countries with a common lawtradition, such as the United Kingdom, Ireland, Cyprus and Malta.See for more details Tjepkema 2010, p. 790 et seq. In these countries, the position is taken that apart from the traditional liability based on torts, it is not incumbent on the courts to create other forms of state liability. Since this concerns a pre-eminently political issue, this is a responsibility to be taken up only by the legislator.See Harlow, Tort law and beyond (Oxford 2004), p. 61. Harlow is of the opinion that the question whether a right to compensation exists in cases where no fault can be pointed out, is not a part of liability law, but is part of a ‘non-legal process’. The Scandinavian countries, Austria and Greece also show that state liability for matters other than unlawful acts has been primarily a legislative issue and that only occasionally were special laws enacted that provided for some kind of liability for lawful acts.See Tjepkema 2010, p. 792 et seq.

By contrast, in a second group of countries, the courts (usually administrative courts but sometimes civil courts) have played a leading role and were not afraid to accept liability in cases where specials laws do not provide for any compensation. Currently, there are nine EU Member States that belong to this second group: France, Spain, Germany, the Netherlands, Belgium, Luxembourg, Portugal, Estonia and Poland.Tjepkema 2010, p. 779-789. Some of these countries have adopted a general statutory regulation governing the liability of public authorities, which is comprised of not only unlawful but also lawful acts (Spain, Portugal, Luxembourg, Poland, Estonia, and soon the Netherlands). In these countries, the legislator is not the only one to determine whether and if so, to what extent, citizens who suffer any damages as a result of lawful actions are entitled to compensation. In addition, the courts play a role in this process as well: they can force the authority who has refused to pay compensation or who gave too little to repair this by offering some compensation. In these countries, it is therefore the judge, and not the legislator who has the last word in deciding whether compensation should be given. Finally, it is worth mentioning that in their review of decisions on compensation for damages, the courts consider the political dimension of this doctrine by carrying out a rather limited judicial review of whether the administrative authority gave a reasonable interpretation of the concept of ‘normal societal risk’.In Dutch literature, this limited review is a point of debate. Some authors take the view that the judge is perfectly capable of judging whether the damage has surpassed the normal societal risk has been surpassed or not. See the debates in Barkhuysen, Den Ouden & Tjepkema (red.), Coulant compenseren? Over overheidsaansprakelijkheid en rechtspolitiek (Deventer 2012).

4.3Dubious Limitation of Liability to Non-legislative Measures

The grounds given by the ECJ in the FIAMM judgment may be interpreted such that it leaves no room for a right to compensation for damages in the case of any damage or loss due to lawful legislative measures. The FIAMM judgment implies that if a legislative measure does not constitute a sufficiently serious breach, any form of liability under Art. 340 TFEU is excluded. To this extent, the unlawful Union act is of an ‘exclusive’ nature. By excluding legislative measures from any possibility to claim compensation for damages, the ECJ has in actual fact ‘clipped the wings’ of this doctrine.

With respect to legislative measures in particular, the exclusion of a right to compensation for damages is highly remarkable. First, it is at odds with the trend that was triggered by the Bergaderm judgment and that continued to shape the relevant case law afterwards, namely that, as far as the liability conditions are concerned, the ECJ no longer makes any distinction between legislative acts and other acts that cause damage or loss.Case C-352-98P Bergaderm v. Commission [2000] ECR I-5291. In the current case law, all these acts are assessed against the criterion of the sufficiently serious breach. If there is hardly any margin for discretion, the mere infringement of EU law may be sufficient to conclude that there has been a sufficiently serious breach.Vgl. Case C-5/94 Hedly Lomas v. Council and Commission [1996] ECR I-2553, para. 28. At the very least, it is remarkable that FIAMM since Bergaderm, the ECJ has subjected all acts of the institutions to the same liability regime (i.e. that of the sufficiently serious breach) and in doing so, it has no longer made any distinction between legislative and other acts, even if this distinction is made in the context of liability for a lawful act.

A second point of criticism is more fundamental. In none of the Member States that recognise the principle of equality is there any support for the exclusion of a right to compensation for damages in the case of legislative acts.In the same sense Van Casteren 1997, p. 252. Indeed the principle of equality has been very important to acts which involve discretionary power in particular. As stated above, the fact that any damages arise from the lawful balancing of interests – including interests served by economic policy choices – is even a prerequisite to the application of the principle of equality.The Dutch Council of State, for example, does not recognize a right to compensation for damages when the authority in question had no possibility whatsoever to take the damaged interest into account when taking the measure which caused the damage (ABRvS 16 June 2005, AB 2007, 11). Even though there are examples of successful claims for compensation for damages in the Member States that recognise the principle of equality, particularly in the field of infrastructure measures (the public works) and decisions, the option of bringing claims against the State on the ground of lawful legislative action is not excluded as a matter of principlein various countries, namely in FranceCE 14 January 1938, Rec. p. 25 (La Fleurette) and CE 21 January 1944, Rec. p. 22 (Caucheteux et Desmonts)., Spain,See Van Casteren 1997, p. 187, who discusses a decision of the Tribunal Supremo from 1993. See on Spanish state liability law also C. Plaza, ‘Member States Liability for Legislative Injustic’ [2010] REALaw, p. 27 e.v. Germany,See the decision Pflichtexemplar (1981) from the Bundesverfassungsgericht. LuxembourgSee G. Ravarani, La responsabilité civile de l’Etat (Luxembourg 2006), paras 278 and 330. and the aforementioned Leffers decision from the Netherlands. Admittedly, in some countries (such as Spain) this case law has been highly controversial and exceptional.Already in the 90’s the Spanish Supreme Court acknowledged that, given certain conditions, Parliamentary Acts which had not been declared unconstitutional (and therefore were in principle lawful) could give rise to the right to compensation for damages that breach legitimate expectations and/or that impose a special and onerous burdens on the plaintiff as a result of the implementation of its provisions (see, among others, STS of 17 February 1998 Nº Recurso 327/1993; STS of 29 February 2000, Nº Recurso 49/1998). In other countries however (such as Germany, France and the Netherlands), there is a substantial body of case law on no-fault liability. What is more, in these countries the leading caseson compensation for damages relate to losses due to legislative acts. Examples include the above-mentioned Leffers judgment, the German judgment in the Plfichtexemplar case and the French judgments in the La Fleurette and Caucheteux et Desmonts cases.

There is also an explanation for the fact that a right to compensation for damages is sometimes recognised in legislation. By its very nature, the assessment made by the legislator is abstract in nature. The legislator is primarily concerned with the question of whether a reasonable legal balance has been struck between the purposes he seeks to achieve and the impairment of the interests of those directly involved. Generally speaking, the legislator is unable, to take account of all adverse consequences that may arise from the application of legislation in a concrete case. For example, when he issued the ministerial regulation for the purpose of imposing a swill ban, the minister was presumably not aware that in the Leffers case the application of this regulation would in some instances result in an unexpected and excessive loss. In particular, the consequences arising from measures that affect large groups of people, and that must be taken within a short space of time, cannot usually be predicted with precision. In the case of damages or losses due to lawful legislation, the ‘well-considered’ sacrifice of interests remains, by its very nature limited to the general abstract level, i.e. to the question of whether a reasonable legal balance has been struck between the objectives the legislator seeks to achieve and the impairment of the interests of those directly involved. In the case of legislation, this element cannot possibly relate to the application of a law in a concrete case. Accordingly, the foregoing warrants the conclusion that in the case of legislation, the above-mentioned necessary relationship between damages/losses and the general interest does not exist between the disproportionate consequences and the lawful act. Usually, the disproportionality of the damages or losses concerns an unforeseen and unintended effect of the application of the power in concrete terms. It is precisely in those cases that the equality principle can entail liability of public authorities. In the field of economic law, the taking into account of individual circumstances entails specific problems however, as will be discussed in the next paragraph.

5The Problems of No-Fault Liability in the Field of Economic Law and the Influence of Efficiency
Arguments

When discussing the possible added value of a no-fault liability regime, one should not forget that the case law of the European Courts does not even show one example of a case in which this type of liability was successfully invoked. It should be pointed out however, that almost the same thing can be said when looking at the national regimes of compensation for damages in the field of economic law. Admittedly, the landmark La Fleurette case of the French Council of State took place in the field of economic law. This case concerned a law which forbade the production of ‘gradine’, which could replace dairy products; by doing so, the French government hoped to protect the dairy industry. La Fleurette specialised in the production of gradine and had to end its businesses. According to the Council of State, the legislator, who – neither in the law itself nor in the explanatory memorandum – paid any special attention to companies such as La Fleurette, could not have intended to ignore the damage to La Fleurette so the disproportionate damage suffered by La Fleurette had to be compensated. Even afterthis landmark ruling, there have been very few successful claims based on the principle of equality where the damage arises from the lawful Acts of Parliament in the field of economic planning. Terneyre calls the La Fleurette judgment a typical product of his time in this respect, being a ‘liberal reaction of the Council of State to the first manifestations of economic interventionism’.P. Terneyre, ‘Reponsabilité en matière d’interventions économiques’, in: P.-L. Frier & D. de Béchillon (red.), Dalloz Encyclopédie de Droit Public: Répertoire de la responsabilité de la puissance publique (Paris 1999), para. 122.

It should be noted that laws in the field of economics often imply that choices are made that are favourable for some and unfavourable for others, because some undertakings are allowed to do certain things which others are not, or because one group, often at the expense of another, is subject to a specific favourable regime. The detrimental act is often inherently discriminatory: intervention in economic life is guided by the idea that only those undertakings that are performing an activity that is the most needed or that favourably stands out compared to others for other reasons, should enjoy a benefit. This may explain why the introduction of a licensing regime does not usually involve the acceptance of any obligation to pay compensation: when the legislature introduces a licensing scheme, leading to the granting of licenses to some parties and to the refusal of a licence to others, there is an implicit understanding that the last mentioned parties cannot claim compensation, since they did not qualify for a licence. Granting compensation in one case will almost certainly set a precedent for other cases, making the implementation of economic policy extremely difficult.See also the Opinion of AG Koopmans in the Dutch case of the dairy farmer De Waal, who suffered damage after the introduction of a quota scheme (Supreme Court 18 April 1997, NJ 1997, 456). Koopmans pointed out that it is up to the legislator to take individual circumstances into account, by making exceptions and taking up hardship clauses. If the legislator did not take an individual circumstance into account (such as, in De Waal’s case, the unfavorable choice of a certain year as a reference year for the allowable quantum of cows to be held) the judge will have to bow to this, Koopmans stated. If the inequality complained of is the natural result of the objective the legislature intended to achieve, the added value of the principle of equality before public burdens – as well as the general principle of equality – will be limited.See also CE 29 mei 1985, Rec. p. 162 (Société ‘Trans-Union’), CE 13 oktober 1978, Rec. p. 370 (Perthuis).

This does not mean however, that Union money is never spent on damages inflicted by lawful acts of union institutions acting within the sphere of economic law. The cases in which this happenend are interesting enough to point out, since they show that it is not considerations of equity but efficiency that will often be an important reason for compensation. This may not come as a surprise. It is no secret that the Court of Justice lends great weight to legal policy when it comes to the recognition and development of new forms of liability.R. Meijer, ‘Rechtspolitieke afwegingen in de Unie- en lidstaataansprakelijkheid’, in: Barkhuysen, Den Ouden & Tjepkema (red.), Coulant compenseren? (Deventer: Kluwer 2012), p. 659-672. As is well known, this effect was of paramount importance to the ECJ in the development of Member State liability. Clearly, this liability was not just motivated by the fact that it may contribute to the loyal implementation of Community law, but just as much by the assertion that the effective application of EU law would be impeded if the Member States were not given the opportunity to hold the state liable for the faulty transposition of EU law. Consequently, liability has a primarily instrumental character in this context: it is a means to guarantee the aim of effective application of EU law.According to T.C. Hartley, The Foundations of European Community Law (Oxford: OUP 1999), p. 60 the development of Member State liability should even be regarded as ‘pure policy’.

In my opinion, the CJEU and the institutions in their capacity as legislator will not be guided by fairness arguments in practice. In all likelihood, a more important factor will be whether the grant of compensation, in cases not involving unlawful acts, may help to achieve the objectives of the Union. The manner in which the right to compensation for damages functions in the Member States also shows that this right quite often has a more instrumental function. In the Netherlands for example, compensation for the purpose of achieving environmental objectives is sometimes promised in the context of negotiations on a new environmental permit.See the report Schadevergoeding. Advies van de Evaluatiecommissie Wet milieubeheer over de toepassing van de schadevergoedingsbepalingen in de Wet milieubeheer, ECWM 2003/16. In addition, compensation is sometimes motivated by the wish to prevent injured parties from taking action against the detrimental action itself.See for further reading Tjepkema 2010, p. 739 et seq. If one considers the situations in which the institutions have awarded compensation payments without having performed any unlawful acts to date, it is clear that efficiency and/or social and economic arguments played an important part in these situations, more so than equity-related concerns.

The first example is derived from the Council decision of 26^th October 1995.PbEG 1995, L 264/28. The background of this decision was the problematic situation that had arisen for roughly 700 Spanish and Portuguese fishermen who carried out fishing activities in the waters subject to the sovereignty or jurisdiction of Morocco. Their losses were not caused by any legislative act, but by the termination of a fishery agreement between the Community and Morocco. This agreement had been terminated prematurely with effect from 30^th April 1995 with the intention to conclude a new agreement with effect from 1^st May 1995. Because it became impossible to round off the negotiations before the latter date, and the fishermen had had to discontinue their activities with effect from 30^th April, they were facing – in the words of the Council – ‘an exceptional situation of particular seriousness’. For this reason, the Council issued a decision granting the fishermen an indemnity for the losses that directly resulted from the failure to renew the agreement. According to the Council, one of the reasons for awarding this compensation was the risk that the sudden discontinuation could result in ‘a serious disturbance of the economy’. Clearly, the granting of compensation was not primarily motivated by the wish to alleviate the hardship of a limited group of fisherman, but more so by the wish to protect the local economy. This shows that the function of no-fault liability on the level of the European Union is above all instrumental: it is likely to be principly used where the granting of compensation can help prevent harm to interests protected by the EU institutions. From this perspective, one wonders whether it is up the EU courts to develop a system of no-fault liability, given the strong ties this type of liability has with EU policy.

This can also be illustrated by a second example concerning the losses undertakings may suffer if livestock has to be killed on a large scale after the outbreak of an infectious disease. Not only does EU policy determine what measures the Member States have to take – even though these measures are carried out at the national level – but it is also the Council of the European Union that has created the possibility of granting a financial contribution to the Member States to cover the costs incurred by them for the purpose of combating infectious animal diseases. Pursuant to a Council decision, a Member State that has incurred costs in combating diseases specified in the decision may claim a financial contribution, provided that the Member State has arranged for ‘swift and adequate compensation’.PbEG 18 juni 2009, L 155/30. In that case, the Member State qualifies for a financial contribution of 50% of the costs incurred for the purpose of indemnifying the livestock farmers and other costs, such as vaccination costs. Unlike what one might suppose, this decision is not primarily dictated by the need to satisfy the requirements imposed by the right to property. Rather, the reason for the decision is that it is highly important in the case of an outbreak of an animal disease to identify the sources of infection as soon as possible. If this does not happen, the process of combating diseases will become harder and lengthier and, as a result, more expensive. Awarding compensation is a suitable means to keep the threshold for notifications as low as possible and contain the risk that the area of infection will grow larger.See the conclusion of AG Jacobs, 29 November 2001, C-428/99 (Van den Bor), § 32.

In my opinion, it is difficult to underestimate the importance of this function of liability for a lawful act to the further development of this doctrine at EU level. Even though it is customary to regard compensation for damages as a ‘mere palliative’, meant for cases where an individual has to make a disproportionate contribution to the promotion of the general interest, it is to be expected that compensation for damages at the level of the EU will be relevant mainly if the economic interests of the Union itself are at stake. Viewed from this perspective, the outcome of the FIAMM judgments is hardly surprising, since it is hard to argue that compensating the FIAMM companies would in any way facilitate EU policy.

6The Future of No-Fault Liability within the EU

A final, obvious question that remains to be answered is whether there is a future for the doctrine of compensation for damages within the EU. Is it enough to maintain the current status quo, where no-fault liability is primarily policy based but where the CJEU has little time for equity-based arguments? Is something to be said for AG Maduro’s statement that the recognition of a regime of strict liability might serve as an outlet for the high barriers of the unlawful act regime?Opinion in Case C-120 and 121/06 FIAMM a.o. v. Commission and Council [2008], para. 57. And – as will probably be the case – if the CJEU sticks to its decision in the FIAMM case, what other ways of redress are available for undertakings who suffer damage as a result of legal actions of the EU? These questions will be answered in the following paragraph. I will first examine the possibilities of compensation for damages within art. 340 TFEU (para. 6.2). Then I will discuss the two ways of redress the ECJ pointed out in the FIAMM case: an action by the legislature (para 6.2) and the possibility of relying on the violation of property rights (para 6.3).

6.1Damage Compensation under Article 340 TFEU

Since the CJEU limited the refusal of the recognition of no-fault liability to legislative action, it would seem plausible that there is indeed some room left under art. 340 TFEU for a right to compensation for damages. It is doubtful however, whether this means anything in practice. In the first place, the CJEU interprets the concept of ‘legislative measure’ in a broad sense, it includes not only legislative acts but also certain decisions. ‘In the context of an action for damages, the [legislative – MT] character depends on the nature of the measure in question, not its form,’ according to the Court.ECJ 11 February 1999, Case C-390/95 (Antillian Rice Mills). Hence, even though decisions relate to individual cases, they can be of a legislative nature if they entail economic policy choices. A case in point is the Live Pigs case, where Italy and the Netherlands suffered a loss as a result of a decision addressed to them, because only in these countries had a swine vesicular disease been identified, and therefore only these countries were prohibited from exporting live pigs to other Member States for a specified period. When an action for damages was brought under art. 288 EU (art. 340 TFEU), these decisions were regarded as legislative measures, however. Other case law too, shows that the most relevant criterion for considering an act as ‘legislative’ is whether the institutions have any discretion in performing the relevant act.CFI 13 December 1995, zaak T-484/93 (Live Pigs). Compare CFI 15 April 1997, Case T-390/94 (Aloys Schröder). See also C. Stefanou & H. Xanthaki, A legal and political interpretation of Article 215 (2) [new article 288(2)] of the Treaty of Rome (Ashage: Dartmouth 2000), p. 89-90.

Second, although there are also acts that are of a purely implementing nature and that do not therefore entail any economic policy choice,See also J. Wakefield, Judicial Protection through the use of art. 288(2)EC (The Hague 2002), p. 64: ‘The administrative function stretches from the highest level of implementation of policy to the lowest level of simple application of rules within pre-defined constraints’. it is hardly conceivable that such acts could ever give rise to an obligation to pay compensation for damages. This is because disputes arising from such acts often involve implementation mistakes (miscalculations, faulty procedures etc.) that will give rise to an action based on an unlawful act.Examples of ‘administrative acts’ are mentioned by Biondi & Farley, The right to damages in European Law (2009), p. 103 e.v. In addition, the Union hardly ever concerns itself with acts without an intended legal effect, and here too, an action based on an unlawful act is more likely than an action based on a lawful act.

As stated in paragraph 4, the arguments that the Court presented to refuse a right to compensation for damages are not entirely convincing. At the same time, the FIAMM decision might come at less of a surprise upon realisation that the the CJEU’s line of reasoning is in fact quite consistent with its case law in the case of the (non-contractual) liability for unlawful acts. Generally speaking, the thresholds regarding this type of liability are very high, demanding that the infringement of a rule of law by the institution is ‘sufficiently serious’.Compare J. Jans a.o., Europeanisation of public law (Europa Law Publishing 2007), p.  258, who speak of a ‘strict criterion’. This is only seldom the case. With the Bergaderm judgment, the liability barriers have even been raised, since the ‘sufficiently serious breach’ criterion is also applicable to non-legislative acts. At the very least, it would be at odds with these high liability barriers if, mainly on the grounds of fairness, the CJEU created a regime of strict liability in which the lawfulness of the cause of the damage or loss is taken as the point of departure and required ‘only’ that the relevant act has led to a ‘special’ and ‘abnormal’ damage or loss.See my case note on the FIAMM-case in AB 2008, 311 and in the same sense R. Meijer, ‘Niet-contractuele aansprakelijkheid voor rechtmatig handelen. Toch geen beginsel van gemeenschapsrecht?’, Overheid & Aansprakelijkheid 2009, p. 156.In English law too, the high barriers relating to state liability for unlawful acts have been used not so much as an argument for the recognition of liability for a lawful act but as an argument against it.D. Fairgrieve, State liability in tort (Oxford 2003), p. 137: ‘[T]he English judiciairy has – until recently – shown a marked reluctance to allow negligence liability of public authorities, let alone any extension into liability based on risk or liability for lawful administrative action.’ This does not mean that compensation for damages as such is incompatible with the legal system of the EU, but this requires a more thorough reasoning than the argument that it is an outlet for the high barriers of the unlawful act regime.

In the end, it is doubtful whether a general right to compensation for damages under art. 340 TFEU will stand any chance. Tridimas is probably right when he points out that a right to compensation for damages will is only likely to exist where a measure has an ‘expropriatory character’.Tridimas 2006, p. 495. Similarly, Craig has stated that the conditions, which have to be met for a right to compensation to exist, should not be underestimated.Craig, EU Administrative Law (Oxford: OUP 2006), p. 781. Especially the criterion of the ‘risks inherent in trading’ will offer the Community Courts a good opportunity to refuse a right to compensation. The FIAMM decision has showed that even producers of batteries, who are not aware of banana market legislation at all, carry an inherent risk of retaliation measures, even where retaliation as such is not foreseeable. Furtermore, in the past the Community Courts have accepted that even damage stemming from unlawful acts belongs, up to a certain extent, to the ‘normal risks inherent in trading’.See e.g. Case C-27/60 Meroni [1961] ECR 1961-348 and Joined Cases C-83 and 94/76, BHNL, Jur. 1978, p. 1209. Of course, the courts’ stringent position on this can be criticised,See the critical notes of Anne Thies’ case note on FIAMM, Common Market Law Review 2006, p. 1165. but it is a position that is all too familiar. This is not only reflected by the difficulty in finding the institutions liable for unlawful action, but also in the case law concerning breaches of the right to property, where the Court has in the past refused to offer any kind of compensation to German importers of bananas, even though they lost an important market share after the introduction of new legislation, and even though they saw their profits disappear and had to fire a substantial number of their personnel.See the opnion of AG Gulmann in Case C-280/93, ECR I-4973, para. 83 et seq. This case is a notorious example of the reticence of the CJEU in liabity cases. See also I. Pernice, ‘Le recours en indemnité’, Cahiers du Droit Européen 1995, p. 641 et seq. on p. 653-654. In essence, Maduro is right in saying that these high hurdles are an apt way of preventing an unlimited development of no-fault liability. Still, the question remains of what then might be the exact added value of recognising this concept at all.

In the light of all this, one might wonder what would have been the added value of the recognition of the equality principle. In particular, it is the question of whether the advantage of the recognition of an almost theoretical type of liability weighs up to disadvantages which are not theoretical at all, when recognising the equality principle, it is highly probable that the Courts would have been overwhelmed with claims for compensation for damages. The prospect of dealing with these claims might well have withheld the ECJ from taking the last step towards the recognition of a right to compensation for damages.

6.2A Statutory Regulation

In the absence of an enforceable right to compensation for damages in the case of legislative action, affected undertakings are dependent mainly on the will of the institutions to award compensation in a decision or under a special scheme, for reasons of solidarity or for pragmatic reasons. This type of legislation and/or resolutions has not really taken off, but there are some examples. First, there are many regulations and decisions with a liability clause in which the non-contractual liability of the relevant institution is made subject to the general principles which the legal systems or the laws of the Member States have in common and which are therefore in line with the wording of art. 340 TFEU.See f.e. Art. 3 of the Decision of the Council of 17 december 1999, PbEG 1999, L 337/41, and Art. 21 of Council Regulation (EC) establishing a Community Fisheries Control Agency and amending Regulation (EEC) No 2847/93 establishing a control system applicable to the common fisheries policy. As long as one accepts that after FIAMM, compensation for damages cannot entirely be written off as a doctrine, there may also be a right to compensation for damages in these special areas.

For affected undertakings, the reference to the ability of the institutions to draft statutory provisions in cases of this kind is less favourable than founding the right to compensation for damages directly on Art. 340 TFEU because undertakings depend on the cooperation of the institutions to establish the existence of liability,. Still, the discretionary power of the institutions to adopt compensation legislation might be the most important road to compensation for the time being. In particular the much discussed issue on the way to offer redress to innocent victims of retaliatory measures, such as the FIAMM case, might be better dealt with by a special regulation than under art. 340 TFEU. First, this path is sensible because the CJEU can compensate victims without recognising in abstract terms that a right to compensation exists. By laying down the right to compensation for damages in a regulation, innocent victims are compensated, but the CJEU need not answer the question whether it is prepared to recognise a right to compensation for damages in a ‘normal’ situation where a lawful regulation causes damage.In the same sense M. Bronckers & S. Goelen, ‘De aansprakelijkheid van de EU voor schendingen van het WTO-recht’, in: Barkhuysen, Den Ouden & Tjepkema, Coulant compenseren (Deventer 2012), p. 689.

A second reason to prefer a compensation regulation above action on the basis of Art. 340 TFEU is that it is highly dubious whether a right to compensation for damages for innocent exporters could be based on this article. When a right to compensation for damages is based on art. 340 TFEU is read in the light of the equality principle – which seems most plausible – it is doubtful whether it could offer a solid basis for compensation in cases such as FIAMM. Indeed, in the FIAMM case the damage did not result directly from a lawful regulation or any other Community act, but from a response by a third country to an EU institution’s failure to comply with WTO rules. Since the damage did not directly result from a lawful act of a Union institution, it may be argued that any no-fault liability based on Art. 340 TFEU could offer no solution to the exporters if the principle of equality was the basis for such liability. The mere fact that the unlawful nature of an act has not been established is insufficient to ‘activate’ the principle of equality. For this to be the case, the damage should somehow be a necessary sacrifice to achieve any political objective (supra, para. 2), which was not the case in the FIAMM case. There is no support for this from the comparative law perspective either: no Member State recognises a right to compensation for damages based on the sole fact that the unlawfulness of an act cannot be established. The reference to the principle of equality has therefore been rightly called ‘artful’ by one commentator, this reference was said to be dictated by the single ‘need to grant some relief to those who were negatively affected by conduct which cannot be declared illegal’.M. Dani, ‘Remedying European Pluralism: the FIAMM and Fedon litigation and the judicial protection of international trade bystanders’ [2010] The European Journal of International Law, p. 331.

A third and last reason is that the method of legislation allows the scope of the obligation to pay compensation to be controlled properly. This relates not only to determining what items of damage or loss qualify for compensation but also to the interpretation of the criterion of ‘normal entrepreneurial risk’. If it is accepted that even innocent importers, who are not part of the sector in which the violation of WTO law has occurred, run some risk of facing retaliatory measures, it can also be argued that it should be laid down in a statutory provision that a specified percentage of the damage or loss suffered is to be borne by the injured parties themselves. This can take the form of a threshold, where the loss must exceed a certain percentage of turnover in order to qualify for compensation. It is also conceivable that a discount will be applied where the damage or loss is compensated in principle, not in full but rather, for example only 60%. In the discussion about the interpretation of the normal entrepreneurial risk, it is often argued that this also depends on political choices; it is the EU legislator that is the most appropriate body for making these choices.

6.3The Right to Property

In the FIAMM judgment, the ECJ pointed out not only the possibility of a statutory compensation scheme but also referred to the possibility that protection can be based on the right to property. It turns out that the case law of the ECJ on this point is for the most part oriented towards property protection under the ECHR. The fundamental rights of the ECHR are considered to be part of EU law as general principles. If and to the extent that they implement EU law, the EU institutions and the Member States are therefore bound by Art. 1 of the First Protocol to the ECHR as well. Article 6(2) of the EU Treaty codifies this case law and even extends the binding nature of fundamental rights as general principles of EU law to the entire European Union.C-36/75 Rutili Roland v. Ministre de l'intérieur [1975] ECR I-1219 and C-46/87 Hoechst [1989] ECR I-1549. Like the ECtHR, the ECJ puts great emphasis on the fact that the right to property and the freedom to pursue a trade or business are not absolute, but should be viewed in relation to their social function.Case C-280/93 Germany v. Council [1994] ECR I-04973.As the well-knownHauerjudgment reads: ‘The exercise of the right to property may (...) be restricted, provided that these restrictions in fact correspond to objectives of general interest pursued by the Community and that, with regard to the aim pursued, they should not constitute a disproportionate and intolerable interference with the rights of the owner, such as to impinge upon the very substance of the right to property.’Case C-44/79 Liselotte Hauer v. Land Rheinland-Pfalz, ECR 1979-3727.

The right to property test has many parallels with a test based on the principle of equality, although both principles do not completely overlap. A significant difference with the principle of equality test is that the legitimate objectives, which the institutions pursue in the general interest, are an independent factor in the context of the proportionality test. Restrictions can be imposed on the exercise of the right to property to the extent that they ‘in fact correspond to objectives of general interest pursued by the Community and that, with regard to the aim pursued, they should not constitute a disproportionate and intolerable interference with the rights of the owner, such as to impinge upon the very substance of the right to property.’Case T-13/99 Pfizer Animal Health SA v. Council [2002] ECR II-03305, para. 457). The nature of this test may be illustrated by the aforementioned Hauer judgment. Ms Hauer wanted to use her land as a vineyard, but she was refused a planting permit, because her site was said to be unsuitable for wine growing. She successfully objected to this refusal, but pending this action, a regulation prohibiting any planting of new vines came into force. The German court sought a preliminary ruling from the ECJ about whether the regulation was contrary to the right to property, primarily because the EEC Regulation did not make it clear whether the regulation prevented granting the authorisation (which would have undoubtedly been granted but for the regulation). The ECJ ruled that the right to property had not been infringed, partly because the measure was of a temporary nature and had been taken to deal immediately with a situation characterisced by surpluses. If, as the ECJ ruled, new planting of vines was permitted, this would from the economic point of view, only increase the surplus. For that reason, the temporary restriction did not affect the substance of the right to ownership.Case C-44/79 Liselotte Hauer v. Land Rheinland-Pfalz, ECR 1979-3727. In practice, this approach to the proportionality test will not easily result in the assumption that the right to property has been infringed. An ‘individual interest’ will quickly be defeated by these general interests, all the more so because in practice the Courts adopt a flexible attitude in its assessment against general interests relied on.Wakefield 2002, p. 130-133. To date, the ECJ has only in one case ruled that the right to property was violated and that the general interest could therefore not justify an infringement.Case C-588, Hubert Wachauf v. Bundesamt für Ernährung und Forstwirtschaft [1989] ECR 1989-2609.

Another important factor is that the restriction on the right to property that was at issue in Hauer directly related to her options for exploiting her land in the way she desired. It is hardly surprising that this meant an infringement of her right to property rights. In a normal situation where a new legislative measure limits a market share to a considerable degree, it is highly debatable whether there exists a right to property in the first place. A good example is to be found in the ECJ’s approach to the ‘banana cases’. A regulation on the common market for bananas had particularly adverse consequences for German importers of bananas. Previously, they had been able to benefit from favourable national schemes and after the entry into force of the regulation, they faced a substantial reduction of their earlier market share, which led to ‘tangible’ and ‘far-reaching’ changes in existing economic structures, the transfer of considerable profit potential to other Member States and mass redundancies.See the Opinion of AG Gulmann in Case C -280/93 Germany v. Council [1994], p. I-4973, paras 83 et seq.

‘The right to property of traders in third-country bananas is not called into question by the introduction of the Community quota and the rules for its subdivision. No economic operator can claim a right to property in a market share which he held at a time before the establishment of a common organization of a market, since such a market share constitutes only a momentary economic position exposed to the risks of changing circumstances.’Case C-280/93 Germany v. Council [1994] ECR I-04973, para. 79.

In this case, the rejection of the claim does not lie in the fact that there was no ‘fair balance’ (to use the ECHR terminology), but in the fact that there was no right to property to the previously existing market share, because this kind of market share constitutes only a ‘momentary economic position’. This approach is strongly reminiscent of that of the ECtHR, which in several cases involving loss of income due to new legislation did not even get round to the fair balance test because the loss concerned was one of the hazards of economic life and the right to property was therefore not adversely affected.

Consequently, it is by no means certain, first of all that a right to property exists in a specific case, and secondly the substantive fair balance test administered by the ECJ is extremely stringent. Relying on the ‘social function of the right to property’, the ECJ justifies even very drastic infringements of the right to property. Indeed, it is established case law that the objective pursued by the EU may justify restrictions that even have substantial adverse consequences for specific market participantsCase C-331/88 Fedesa [1990], ECR I-4023, para. 17) and Joined Cases C-11/04, C-12/04 en C-453/03 ABNA [2005] ECR I-10423. – a ground that cannot even be found in the ECtHR’s case law on art. 1 FP. For the ECJ, the fact that production restrictions could adversely affect the profitability of some undertakings and might even jeopardise their very existence is no reason to assume an infringement of the right to property.Case 172/83 Hoogovensgroep [1985] ECR 1985-2831, para. 23) and Case 258/81 Metallurgiki Halyps [1982] ECR 1982-4261, para. 13). In theory, the recognition of the principle of equality could have changed this very conservative compensation regime, partly because, under the regime of the equality principle, no weight is given to the general interest in the decision on whether damage or loss should be borne by the entrepreneur. In practice, the right to property test shows great similarities to the ‘abnormal burden’ test. In the context of the right to property, the risk of a restriction on a market share existing at any time may be relevant to a preliminary question on whether the right to property exists in a specific case, and in the context of the principle of equality, that very same question is relevant to the abnormal burden test.

7Last Remarks

The liability of the European Union on the basis of Art. 340 TFEU is a multi-faceted and complex matter. It not only involves a comparison with the laws of different Member States, but also issues concerning legal policy. After many years of flirting with the recognition of a no-fault liability on the EU level, the FIAMM decision is in the view of many commentators a disappointment. It is typical however, that the question concerning no-fault liability came about in the context of the WTO, with its special characteristics and impossibility of an action regarding unlawful action by the Community. It therefore remains to be seen if the FIAMM case was actually apt for answering the fundamental question on the existence of no-fault liability within art. 340 TFEU. In any case, the FIAMM decision implies that companies dealing within the boundaries of EU law have no possibility of enforcing a right to compensation when they cannot show that the EU institutions have acted unlawfully. This means that the right to compensation for damages will probably for a large part be part of legal policy of the Communities. Where compensation for companies can in some way help to fulfil the goals of the institutions, sometimes a right to compensation will be brought about, be it by a special regulation or an individual decision. Admittedly, this is a very limited approach of the matter of no-fault state liability, proving that the CJEU is not very open to equity based arguments and underlining the CJEU’s instrumental approach to liability.