Jean Monnet Center at NYU School of Law

Subsidiarity as a Rule and a Principle, or:
Taking Subsidiarity Seriously

Theodor Schilling[*]

Revised version

©Copyright: Theodor Schilling , 1995.

Table of Contents

I. Introduction
II. The subsidiarity principle as legal norm
III. The subsidiarity principle as a rule (the subsidiarity rule)
IV. The subsidiarity principle as a principle, in its technical sense (the subsidiarity principle as principle)

A. General remarks
B. Exclusive Community competences
C. The principle of subsidiarity and the supremacy of Community law

V. Conclusion


A. The 1991 Maastricht Conference on the Political Union was saved, it is sometimes claimed, by one word: subsidiarity [1]. Indeed, this word, or the concept expressed by it, introduced into the then EEC Treaty for the first time by the Single European Act (SEA), in the context of the environmental policy (Art. 130 r (4) EECT) [2], has been used widely throughout the Maastricht Treaty. It is part of the European Treaties now in at least four places: the second penultimate recital in the preamble of the Maastricht Treaty (expressly), Art. A (2) of the Maastricht Treaty (impliedly), Art. B (2) of the Maastricht Treaty (expressly) and, last but not least, Art. 3b (2) ECT [3]. It is now the second most often mentioned principle in the European Treaties; only the prohibition of discrimination is mentioned in more places. Its specific importance is underscored by the decisive rôle it played in the success of the Maastricht Conference and in the ultimately succesful efforts to dispel widespread popular concern about the Maastricht Treaty [4]. Plainly, therefore, it appears at the outset that it must be taken very seriously indeed [5].

However, a lingering doubt subsists. Could it possibly be that "the word that saved Maastricht" is just that, just a word, bare of any concept [6]? It may well be that this was the intention of some, or even many, of the delegations at the Maastricht Conference. However, it is not possible to ascertain how the individual members of the Maastricht Conference conceived of this word. Neither is it necessary. They introduced the word into what, after ratification, became the amended treaties, and it is there, in the treaties, where its meaning, the concept of subsidiarity, must be found [7]. In the context of Community law, the German adage according to which the law is wiser than the law-giver [8] must clearly apply. There was never any question of applying, to European law, anything similar to the American theory of original intent [9]. Possible mental reservations of some, or many, delegations at the Maastricht conference are therefore no argument against taking subsidiarity seriously.

B. The other arguments I want to make in favour of taking subsidiarity seriously presuppose a certain understanding of this principle. The subsidiarity principle has to be construed according to the general meaning of the words used in their respective context [10]. This task of construction is made relatively simple by the texts themselves which contain an authentic definition [11], albeit not a completely satisfactory one [12]. According to Art. 3b (2) ECT, the subsidiarity principle means that the Community shall take action only "if and insofar as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can therefore, by reason of the scale or effect of the proposed action, be better achieved by the Community" [13]. It is obvious that the subsidiarity principle, according to this definition, concerns only the relationship between Community and Member States, and not the relationship between society and state [14]. A somewhat different definition is given by the second penultimate recital of the preamble of the Maastricht Treaty, i.e. that "decisions are taken as closely as possible to the citizen" [15]. This definition appears to be much wider than the Art. 3b (2) ECT definition; whereas it encompasses the latter, it covers, in addition, intra-state relationships and possibly even the relationship between society and state. In the present context, however, only the relationship between Community and Member States is of interest.

It has been claimed repeatedly that the subsidiarity principle works two ways: that it founds a competence on the lower level of government, i.e. the Member State level, if the objectives of an action can be achieved at that level, and a competence on the higher level, i.e. the Community level, if the said objectives cannot be thus achieved [16]. Conversely, the subsidiarity principle has been seen as a "double-edged sword" which prevents both the higher and the lower level "from taking an action in areas properly falling within each other's respective sphere of action" [17]. This is certainly true of the subsidiarity principle if considered in abstracto, devoid of any Community context. Within this context, it is still true insofar as the creation of new Community competences is concerned; such a creation can, according to circumstances, validly be based on the subsidiarity principle. But otherwise it is not true in the Community context. This follows from the fact that under the Community system of competence allocation the residual powers are held by the Member States, whereas the Community is, in principle, free to act within the scope of its concurrent competences [18]. In doing so, it may at will restrict the Member States' possibilities of using concurrent competences. Conversely, under the concept of Community law supremacy [19], the Member States have no such power to restrict the Community's use of concurrent competences. Therefore, as the subsidiarity principle protects the use of powers by the level of government best able to achieve a certain object against the encroachment of another level, it can only be operative against the Community level. This applies particularly in relation to Art. 3b (2) ECT. That provision makes it clear beyond doubt that it restricts the exercise (and only the exercise) [20] of Community competences, and of Community competences only [21]. Nothing in the text allows the claim that the restriction of only Community competences is (only) "the primary political objective of subsidiarity" and not the legal meaning of the provision [22]. There simply is no way in which the Community could claim, under the subsidiarity principle within the meaning of Art. 3b (2) ECT, a competence not granted to it elsewhere [23]. Nor can it be said that Art. 3b (2) ECT prevents the Member States from acting even if a given measure might be best handled at the Community level as long as the Community does not in fact act, nor that it creates an obligation for the Community to act "where the prerequisites [of Art. 3b (2) ECT] exist" [24]. There is simply nothing in the wording of this provision that could be construed as obliging the Community to act under such circumstances [25].

C. It has been shown convincingly that, throughout the Community's history, its relationship to the Member States was balanced in this sense that the individual Member States had either a decisive say in Community matters (Voice) or the possibility to avoid the consequences of Community measures (Exit) [26]. This balance has first been shattered by the Single European Act. The individual Member State's Voice was reduced by the increased use of majority voting, and the (scarce) possibilities of Exit remained the same [27]. The Maastricht Treaty continued, in part, in the same vein. The enhanced participation of the European Parliament in the legislative process under Art. 189b ECT further reduced the Member States' Voice, as did the subsequent enlargement of the EC. In an enlarging Community, position and influence of all individual Member States decrease. At the same time, one Exit possibility, namely a strategy of non-compliance [28], was made more difficult first by the decision of the ECJ in Francovich [29] and then by the legislative introduction, in Art. 171 (2) (2) and (3) ECT, of sanctions for non-compliance with specific judgments of the ECJ. In addition, the Maastricht Treaty was set to lead to an important increase in legislative powers of the Community.

The Community provisions on subsidiarity introduced under these cicumstances are quite unique as regards the legal technique employed: Comparative research has revealed that the subsidiarity principle as understood here [30] is expressly only enshrined in the Community legal order [31]. In other legal systems, subsidiarity has been an aspect in the making of the respective constitution, and its directing force generally has been exhausted by the concrete regulations laid down in that constitution [32]. Rules in point are provisions on federalism, regionalism and local autonomy [33]. For example, in the German Basic Law (BL), prior to the constitutional amendments dealing with the Maastricht Treaty [34], subsidiarity was not mentioned expressly at all [35]; it was not dealt with, at least not in our context, in smaller commentaries on the constitution [36]. For the rest, according to the better, and near-unanimous, opinion, the subsidiarity principle, although the basis of some key concepts of the German constitution [37], cannot be used, as such, as a means to construe the Basic Law [38]. The same is true with respect to the U.S. constitution. Although ideas of having decisions taken as closely as possible to the people are clearly at the basis of some aspects of the specific American brand of federalism [39], in the construction and application of the constitution this principle is neither mentioned nor, not even under another name, applied [40].

It is not just a technicality that only the European Treaties contain an express provision sanctioning the subsidiarity principle. Rather, this circumstance signifies that, whereas in other multi-tier structures the constituent power assumes that it has done all it considers to be required by the subsidiarity principle in drafting the constitution as it did, the High Contracting Parties, in drafting the Maastricht Treaty, concluded that the general rules of the European Treaties did not respect the subsidiarity principle to the extent they considered necessary. For that conclusion at which, according to the textual evidence, the parties of the original European Treaties and of former amending treaties had not arrived, the Member States' growing discontent with the whole state of affairs just described was the main reason [41]. The development of the subsidiarity principle was a kind of antidote against the growing anxiety of some Member States that they might lose too many powers to the EC. The insertion of the provisions containing the subsidiarity principle must therefore be seen as a device employed by the present parties to remedy the perceived shortcomings in the European Treaties.

With one of these shortcomings, i.e. the unrestricted increase in Community legislative power, the subsidiarity principle as introduced in the EC Treaty deals directly. The EC must not use its powers in a way contrary to that principle [42]. The other shortcoming, i.e. the shattered balance between Voice and Exit, is not directly dealt with. Whereas Exit in the sense here used is the possibility to avoid the consequences of certain already enacted Community measures, the subsidiarity principle as discussed above is only a restriction on the future use of legislative and administrative powers [43]. However, as the subsidiarity principle is conceived as a device to remedy perceived shortcomings in the European Treaties, it must be looked at as offering, in terms of the Voice / Exit dichotomy, an additional Exit possibility. This assumption presupposes the development of an additional dimension of the subsidiarity principle. This development is the main enterprise undertaken in the present article [44]. Together with the Loyalty developed by the Member States and their institutions during two decades of enhanced Voice [45], such an Exit possibility, it is submitted, could serve to maintain, for the time being, an equilibrium otherwise threatened by the diminution of Voice brought about by the SEA, by the other amendments brought forward by the Maastricht Treaty, and by the subsequent enlargement of the EC. Such a development of the subsidiarity principle is in keeping with the view that it is one of the basic principles of Community law, and with taking it seriously.

D. One last point must be raised in this context: the well-known problem of Community legitimacy. It has been noted repeatedly that the legitimacy of the Community is, under democratic aspects, rather weak [46]. This state of affairs which, of course, could [47] and probably should be changed in the middle term, makes it necessary to construe transfers of power from the clearly and democratically legitimized Member States to the largely unlegitimized Community restrictively, and to stress, in reverse, limitations on such transfer including Exit possibilities. Prominently among these limitations ranges the subsidiarity principle, so that here is a further reason for taking it seriously. The question is then one of the consequences.

E. Taking subsidiarity seriously may have many different consequences in different contexts. I am here interested not in the consequences for the governance and administration of Europe. These questions have been dealt with widely and expertly. Instead, I am interested in the consequences for the application of existing law. Even in this restricted area, quite diverse questions do arise. The subsidiarity principle may have consequences for the relationship between Community and national courts. Subsidiarity, it might conceivably be claimed, does not allow for the Community courts' having the last word in subsidiarity questions. This question must be dealt with elsewhere [48]. The present article is mainly concerned with the influence the subsidiarity principle may have on the interpretation of primary Community law, and with the consequences of this influence on the solution of conflicts between Community and national laws.

The plan followed in this article is this. I shall show, sub II, that the subsidiarity principle as enacted in the Maastricht Treaty is a legal as opposed to a merely political norm. I shall try to demonstrate, sub III, that the subsidiarity principle as enacted in Art. 3b (2) ECT is a rule as opposed to a principle in the Dworkinian sense. Sub IV.A I shall argue that there must be stipulated, alongside this rule, a subsidiarity principle in its technical sense, in first line to protect the subsidiarity rule against possible circumventions by Community authorities. The contents of the subsidiarity principle in its technical sense can be developed from the relevant provisions of the Maastricht Treaty. Once thus developed, this subsidiarity principle can be applied to diverse sets of circumstances, thereby opening for the Member States an additional Exit possibility. It is proposed then to analyze, under two headings, the implications of this subsidiarity principle in its technical sense. With a view to checking the increase in Community legislative power, the implications for the interpretation of the concept of "exclusive competence" in Art. 3b (2) ECT will be discussed sub IV.B. With a view to opening an additional Exit possibility, the relationship between the principles of subsidiarity and of supremacy of Community law will be dealt with, sub IV.C, on the one hand with respect to Member State constitutional law, or other law protecting important national values, and on the other hand with respect to the interpretation of Art. 30 ECT.

II. The subsidiarity principle as legal norm

A. The Community is a community based on the rule of law [49]. This implies it is premised on legal norms. The question whether the subsidiarity principle is a legal as opposed to a merely administrative or political norm has been widely discussed, mainly de lege ferenda, i.e. before the adoption of the Maastricht Treaty [50]. A curious split of opinions [51] has occurred with lawyers coming out in favour of treating the subsidiarity principle as a political principle and politicians coming out in favour of treating it as a legal norm. There can be no serious doubt that it would have been possible, even in a community of law, to formulate the subsidiarity principle as a purely political principle, that is, legally to grant the Community full and unconditional exercise of its powers but to restrict that exercise as a matter of politics. In this sense, the subsidiarity principle has been called "a maxim of sound administration" [52], inherent in any system, at least in any multi-tier system, be it a federal or regionalized state or an International Organisation [53]. However, this is not what has finally been enacted [54]. Quite the contrary, from the wording, genesis and context of Art. 3b (2) ECT it is clear that this is a legal provision binding on the Community institutions [55].

This holds true for both questions to be answered in application of Art. 3b (2) ECT. As has been noted, the answer to the question whether the subsidiarity principle has been violated presupposes, under Art. 3b (2) ECT, an answer to the question whether the relevant matter falls into the exclusive competence of the Community [56]. The latter is indubitably a legal question [57]. However, the question whether the Community would violate the subsidiarity principle proper by exercising a given concurrent competence [58] is a legal question, too [59]. The wording ("the Community shall take action ... only ...") puts it beyond doubt that the provision is more than a political or administrative principle. This is, if need be, confirmed by the parallel wording of Art. 3b (1) ECT containing undisputedly a legally binding obligation. Second, its genesis shows Art. 3b (2) ECT as the result of a prolonged dispute between the Member States about the appropriate concept for sanctioning the subsidiarity principle [60]. In the end, the advocates of the view that the subsidiarity principle should be legally binding on all Community institutions and that this principle must, therefore, be incorporated into the Treaty itself as opposed to its preamble, won out over the advocates of a mere political and programmatic character of that principle. Finally, the context the subsidiarity principle found in being posited, in Art. 3b ECT, among two other legally binding norms governing the Community actions, i.e. the principles of attribution of (or conferred) powers and of proportionality, shows demonstrably that the subsidiarity principle, as laid down in Art. 3b (2) ECT, is a legally binding norm itself [61].

The question of the justiciability of the subsidiarity principle which has come to the foreground since the coming into force of the Maastricht Treaty [62] is for most purposes the same question as the question whether this principle is a legally binding norm, asked from the vantage point of common law centered not, as the civil law would have it, on abstract rules but on a prognosis of how the courts will decide [63]. It follows that the question whether the Community has the right to exercise a given concurrent competence, under the subsidiarity principle, this question being a legal question, is for the Court of Justice to decide [64]. We are here concerned neither with the exact modalities of this judicial control [65], nor with its inherent difficulties [66].

III. The subsidiarity principle as a rule (the subsidiarity rule)

Taking subsidiarity seriously does not simply require to treat the subsidiarity principle as a legal norm. There is the further requirement fully to develop that norm. In doing so, and to develop, in particular, an additional Exit possibility, it is useful to distinguish between two types of legal norms: rules and principles. This distinction, first developed, in English texts [67], to attack positivism [68], has since been used convincingly to describe and analyse positive law phenomena [69]. It is proposed to apply this distinction to the subsidiarity principle.

"The difference between legal principles and legal rules", writes Dworkin [70], "is a logical distinction. Both sets of standards point to particular decisions about legal obligation in particular circumstances, but they differ in the character of the direction they give. Rules are applicable in an all-or-nothing fashion. If the facts a rule stipulates are given, then either the rule is valid, in which case the answer it supplies must be accepted, or it is not, in which case it contributes nothing to the decision." "But this is not the way ... principles ... operate" [71]. "All that is meant, when we say that a particular principle is a principle of our law, is that the principle is one which officials must take into account, if it is relevant, as a consideration inclining in one direction or another" [72]. "This first difference between rules and principles entails another. Principles have a dimension that rules do not the dimension of weight or importance. When principles intersect ..., one who must resolve the conflict has to take into account the relative weight of each. This cannot be, of course, an exact measurement, and the judgment that a particular principle ... is more important than another will often be a controversial one. Nevertheless, it is an integral part of the concept of a principle that it has this dimension, that it makes sense to ask how important or how weighty it is" [73].

Under this distinction, the subsidiarity principle as laid down in Art. 3b (2) ECT must be considered, in spite of its name, as a rule (the subsidiarity rule). It is not a very well-defined or clear-cut rule, but a rule it is. The Community is only allowed to legislate, within its concurrent competences, when certain requirements are met [74]. There is no weighing of intersecting principles. Either these requirements are met then the Community is permitted to act or they are not met then it is prevented from acting. This point is well demonstrated by some authors who purport to weigh the subsidiarity principle against other principles. It has been claimed that the Community task, under Art. 2 ECT, "to promote ... economic and social cohesion and solidarity among Member States", on the one hand, and the subsidiarity principle, on the other, are contradictory principles [75], and that this contradiction can only be resolved according to unfettered discretion. This is not quite exact insofar as the subsidiarity rule in Art. 3b (2) ECT is concerned. Though the effects of Art. 2 ECT may well countervail, in some ways, the effects of the subsidiarity principle in general, this happens on a level before the application of that Article. If it is the objective of a proposed measure to promote cohesion and solidarity among Member States, it is very likely that that objective cannot be sufficiently achieved by the Member States. So, a simple application of the rule of Art. 3b (2) ECT will be likely to permit, in such cases, the adoption of the Community measure in question.

Another author confronts the subsidiarity principle with a host of other principles or rules [76]. A closer look at his arguments reveals, however, that he treats the respective conflicts as conflicts between rules, not principles in their technical sense. In particular, his discussion of the relationship between the subsidiarity principle and the principles of absorption [77] and pre-emption [78] of Member State competences shows that these principles are seen as severely restricting the field of application of the subsidiarity principle [79]. These other "principles" are seen as taking precedence over the subsidiarity rule in Art. 3b (2) ECT insofar as their respective fields of application overlap. This, however, is the typical result of a conflict between rules, not principles.

IV. The subsidiarity principle as a principle, in its technical sense (the subsidiarity principle as principle)

A. General remarks

1. Thus, Art. 3b (2) ECT contains a rule, not a principle. Nevertheless, the above discussion shows clearly that the field of application of that rule can be diminished, to a very important degree, either by manipulating the requirements for Community action there posited or by using other techniques, outside the orbit of Art. 3b (2) ECT, to achieve such a restriction. Therefore, the question arises whether the concept of subsidiarity can have any importance in preventing such a restriction. Clearly, the subsidiarity rule in Art. 3b (2) ECT does not apply here. The question is whether there is, in addition to that rule, a subsidiarity principle, with "principle" here understood in its technical sense as discussed above, and, if there is, what kind of influence that principle may have in Community law [80].

The subsidiarity principle principle not used in its technical sense variously has been called a constitutional principle [81], one of the general principles [82], and a structural principle of the European Union [83]; it has been said that it is at the centre of the definition of the new rôle of the Community itself [84]. It is, as stated above, the second most frequently mentioned principle in the EC Treaty, not only in the prescriptive provision of Art. 3b (2) ECT, referred to in Art. B (2) of the Maastricht Treaty, but also, if using a different phraseology, in the preamble [85] as well as in Art. A (2) of the Maastricht Treaty [86]. All these expressions of the subsidiarity principle supply the institutional support [87] necessary for arguing the existence, within the European Treaties, of a legally binding subsidiarity principle. It had been foretold, correctly, it is submitted, that this principle "will join the existing set of general principles [of Community law] and play a similar part in the Community legal system" [88]. Indeed, subsidiarity, taken seriously, cannot be restricted to the rôle of a rule within the technical limits of Art. 3b (2) ECT. Such a position would be quite defenceless against several different attacks, on the one hand, and would manifestly fail to develop the full potential of the subsidiarity principle, on the other. Subsidiarity, taken seriously, must be given an additional rôle as principle in its technical sense, with the aim, among other things, to defend the position of the subsidiarity rule [89].

2. The contents of the subsidiarity principle as principle must be found in the provisions of the EC Treaty, read together with Title I - Common Provisions of the Maastricht Treaty. The second penultimate recital of the preamble, providing that decisions must be taken as closely as possible to the citizen, and the identical text of Art. A (2) of the Maastricht Treaty certainly have a wider meaning than Art. 3b (2) ECT the requirements of which are much more specific. However, if applied to the relationship between the Community and Member States, their basic message is that the Community should only act when, and insofar as, a decision closer to the citizen, i.e. on a national level, is not possible. The same message can be gained from the subsidiarity rule in Art. 3b (2) ECT. This rule must be understood as the crystallization of one application of the subsidiarity principle to a narrower field, to a smaller circle of addressees, and with a more clearly defined content, i.e. the content of a rule.

This subsidiarity principle as principle is of much wider application than the rule in Art. 3b (2) ECT. It does not only apply to the "if" of the exercise of a concurrent competence, but also to the mode of this exercise [90], to the question, e.g., whether a regulation is necessary to provide for a certain legislative matter or whether a directive will do [91], and how detailed such a directive should be [92]. Further, it does not only apply to Community legislation and administration. As a pervading principle to be respected by all the institutions of the Community, including the Court of Justice, it also applies to the interpretation and application of Community law [93]. As stated, it is this aspect I am primarily concerned with in this article.

B. Exclusive Community competences

1. The interpretation of the concept of "exclusive competence" in Art. 3b (2) ECT.
a) Generalities

aa) According to Article 3b (2) ECT, "[i]n areas which do not fall within its exclusive competences, the Community shall take action [only] in accordance with the principle of subsidiarity ...". The concept of "exclusive competence" thus circumscribes the field in which the subsidiarity rule in Art. 3b (2) ECT does not apply, thereby potentially severely restricting the effect of the subsidiarity principle. Whilst learned authors agree that the concept of exclusive competence (as well as its opposite, i.e. concurrent competence [94]) is well known in federal constitutions but alien to the original EEC-Treaty [95], they disagree widely as to the meaning of that concept. The proposals discussed by academic writers and Community institutions can be classified, it is submitted, under four headings: those who see basically all Community competences as exclusive, those who consider Community competences once exercised as exclusive, those who assume an exclusive Community competence in all fields in which the amount of Community legislation surpasses a certain "critical mass", and those who understand the concept of "exclusive competence" in Art. 3b (2) ECT as a term of art.

bb) "Exclusive competence" has, as a comparative survey reveals, a very definite meaning which I shall refer to as its technical sense. Insofar as the competences of one level of government are exclusive, the other level is barred from legislating irrespective of any legislative activity by the first level. This is the undisputed meaning given to the term in the legal orders of the U.S. [96], of Canada [97], of Germany [98], of Switzerland [99], of Belgium [100] and of Austria [101]. Neither in court decisions nor in learned articles in those countries any dispute about the meaning of the concept of exclusive powers can be discerned. There is, one may suppose, a simple explanation for that fact namely that the concept of exclusive powers (or competences) of one level of government is only used, in all these countries, to designate areas of legislation from which the other level of government is excluded. No further consequences result from the use of that label. In legal orders like Austria, Belgium and Switzerland, in which all legislative competences are, in principle, exclusive competences of either the national or the sub-national level, this is self-evident. In such an order, to use the concept of exclusive powers in any sense wider than its technical one could serve no useful purpose. In legal orders like the American or the German, in which exclusive powers of both levels of government co-exist with concurrent powers, the interests of the national level are guaranteed, in the areas of concurrent powers, by the concepts of federal supremacy, or federal pre-emption, or both.

cc) Contrary to all the municipal legal orders considered above, in the EC context the concept of exclusive competence is used prescriptively in this sense that further consequences do result from the use of that label. As soon as a Community competence can be labelled "exclusive", the subsidiarity rule in Art. 3b (2) ECT does not apply [102]. It is that prescriptive use which defines the context in which the concept of exclusive competence within the meaning of Article 3b (2) ECT must be construed. Therefore, the technical sense of the concept is not necessarily the sense to be degaged from Art. 3b (2) ECT. In the EC context an autonomous interpretation of the concept is necessary.

b) Are all Community competences exclusive?

The most far-reaching proposal of such an interpretation [103] suggests "that in all matters transferred to the Community from the Member States, the Community's competence is, in principle, exclusive and leaves no room for any concurrent competence on the part of the Member States. Therefore, where the competence of the Community begins, that of the Member States ends. From then on, Member States no longer have the power unilaterally to introduce legislation." "[I]t follows that the principle of subsidiarity cannot apply to any matter covered by the original EEC Treaty" [104].

There is a very powerful argument against this view. It would reduce the impact of the subsidiarity principle to naught, by reducing the field of application of the subsidiarity rule in Art. 3b (2) ECT. But the mutually exclusive relationship between the concept of exclusive competence and the subsidiarity rule works in two ways. On the one hand, the subsidiarity rule in Art. 3b (2) ECT does not apply, according to that provision, in areas which fall in the Community's exclusive competence. On the other hand, however, the concept of exclusive competence used in that provision has to be understood taking due account of the subsidiarity principle as principle. This is an interpretation technique routinely employed in German constitutional construction: the so-called "Wechselwirkungstheorie" (theory of reciprocal effect). This theory, originally developed to restrict the impact of restrictions of freedom of speech admissable under the German Basic Law [105], is now employed to similar cases in which the restriction of a basic principle by other legal norms is permitted under the Basic Law. The Wechselwirkungstheorie is, according to common opinion [106], a special application of the principle of proportionality and as such, it is submitted, applicable in Community law. If one applies this theory to Art. 3b (2) ECT, this provision must be given a meaning leaving the subsidiarity rule as crystallization of one of Community law's general principles an important field of application. In the same vein, the principle of proportionality itself demands that the subsidiarity principle is not further restricted, by the exclusive competence exception, than is strictly necessary. Finally, the effet utile principle according to which the interpretation of a provision should grant that provision the effectivity aimed at by the legislature [107] militates against a wide interpretation of the concept of exclusive competence. It demands that the subsidiarity rule be given an interpretation under which it can fully play its rôle [108]. Accordingly, the view that all Community competences are exclusive is not tenable.

c) Are Community competences exclusive once they have been exercised?

aa) It has been claimed that a concurrent Community competence becomes exclusive once it has been used by the Community [109]. Similarly, in the Commission's communication on the principle of subsidiarity [110] it is stated that exclusive Community competence is "the possibility of depriving the Member States of the power to act" [111]. Those two statements are very close though not identical. Under the first statement, the competence is, before its exercise by the Community, a concurrent one, whereas it is, under the second statement, already an exclusive one although it leaves the Member States temporarily the power to act. Offering yet another nuance, the Court, in its Opinion 2/91 [112], after having stated "that the existence of [exclusive] competence arising from a Treaty provision excludes any competence on the part of Member States which is concurrent with that of the Community ..." continues thus: "The exclusive or non-exclusive nature of the Community's competence does not flow solely from the provisions of the [EEC] Treaty but may also depend on the scope of the measures which have been adopted by the Community institutions for the application of those provisions and which are of such a kind as to deprive the Member States of an area of competence which they were able to exercise previously on a transitional basis" [113].

bb) Assuming that the term "exclusive competence" is used in these statements in its technical sense, comparative law offers an example of a legal system that, according to some authors, corresponds to the Commission's statement. According to these authors, the Swiss order [114] is based, in principle, on mutually exclusive legislative competences of federal government and cantons [115]. When the Swiss federal government is granted, by constitutional amendment, a competence hitherto reserved to the cantons, this competence is, in conformity with Swiss constitutional principles, an exclusive one [116]. However, the cantons' competence is not considered ipso facto annulled; rather, it continues until the federal government makes use of the newly granted competence, and it is avoided only to the extent to which the federal government effectively uses its competence [117]. The competence discussed therefore closely resembles a concurrent competence [118]. What makes it an exclusive one is the fact that the ultimate transfer of competence to the dederal government is irreversible but for a new constitutional amendment. It is a deferred exclusive competence.

A claim that the concept of exclusive competence of the Community encompasses a deferred exclusive competence may be implied in the Court's statement according to which the Member States were only able to exercise competences "on a transitional basis" [119]. That language appears to suggest that the Community competence had been intrinsically exclusive in its technical sense from the outset and that a transfer of powers to the Communities was, therefore, irreversible [120]. However, such a claim has never been made expressly. The Community's model of competence allocation is fundamentally different from the Swiss. Whereas the Swiss model is based on mutually exclusive legislative competences granted to both levels of government for specific areas of legislation, the Community model is founded on competences granted to achieve certain ends [121] and characterized by the fact that these competences do not cover specific areas of legislation. It follows that, in the absence of any Treaty language to the contrary, the claim discussed is clearly not tenable.

cc) Comparative law does not know of any legal system under which a concurrent competence becomes an exclusive competence, in its technical sense, of one level of government by the mere fact of having been used by that level. However, there are, in logic, no reasons against such a proposition. Undeniably, there are no differences in cause or effect between such a model and the Swiss model of deferred exclusive competence; the only difference is as to the label of the competence before it has been used by the national level. But this difference is of importance once the interpretation of positive law is required. The label of deferred exclusive competence describes exactly, in the Swiss case, the constitutional competence allocation between federal government and cantons from which flow the legal effects of the use of that competence. The label of concurrent competence, applied in the case of Community law, describes a competence allocation from which such effects do not flow. Therefore, a different positive law basis would be needed to premise the claim that a concurrent competence once used by the Community becomes, by reason of that use, irreversibly a Community competence. In fact, this is an extraordinary claim that contradicts the general principle of parallélisme des compétences [122] according to which the legislature competent to enact a certain provision is also competent to abolish it, that is to enact an actus contrarius [123]. Such a contrary act would mean, in the present context and in accordance with the U.S. and German examples of concurrent competences used by the federal levels of government, that after the abolition of the Community provision it would be open to the Member State level to use the concurrent competence anew.

A positive law basis for the conversion of a concurrent into an exclusive Community competence, in its technical sense, by Community use of the concurrent competence once could possibly have been found in the first paragraph of the preamble to the original EEC Treaty according to which the Member States are determined "to lay the foundations for an ever closer union among the peoples of Europe" [124]. Read together with Article 5 ECT, according to which "Member States shall take all appropriate measures ... to ensure fulfillment of the obligations arising out of this Treaty ...", that passage could be taken to permit, even to order that the use of a concurrent competence, once used by the Community, could never revert to the Member States. The formation of an "ever closer" union could thus be seen as a one-way-road to an ever increasing centralization of powers. The idea of the acquis communautaire runs in the same direction.

After the coming into force of the Maastricht Treaty, however, this basis is no longer available. Whilst it is true that the second penultimate recital of the preamble of that treaty follows the EEC Treaty's "ever closer union" language by declaring the Member States "resolved to continue the process of creating an ever closer union among the peoples of Europe", it qualifies that "ever closer union" by stating it to be a union "in which decisions are taken as closely as possible to the citizen in accordance with the principle of subsidiarity" [125]. Therefore, the "ever closer union" is now clearly perceived as compatible with the principle of subsidiarity; it has been made clear that the "process of creating an ever closer union" does not necessarily imply any centralizing tendencies [126]. It follows that that "ever closer union" language cannot any longer be used to found the thesis that the exercise of concurrent competences of which the Community has once made use can never be relinquished to the Member States.

In accordance with the principle of parallélisme des compétences, therefore, and in the absence of any positive law basis founding an argument to the contrary, the Community must be taken generally to have the power to abolish provisions which it had enacted, even without replacing them by other provisions. If it does so in some instances, there may be an obligation, under the principal of subsidiarity as principle, to do so [127] , the corresponding legislative competence may again be used by the Member States; the Community's competence had not become exclusive in the technical sense [128].

d) Does a Community competence become exclusive within the meaning of Art. 3b (2) ECT once Community legislation in a given field surpasses a certain "critical mass"?

aa) Comparative law does know of the possibility that national law, by surpassing a certain "critical mass" of legislation in a given area of concurrent competence, pre-empts that area and thereby excludes subnational legislation in that area. Under the German Basic Law, insofar as the federal government's competence is concurrent, Land legislation is excluded if, and to the extent that, the federal government makes use of its competence [129]. The latter rule closely resembles that of the U.S. concept of federal pre-emption [130]: in both cases, the states and the Länder, respectively, may lose their competence to legislate because of a legislative act of the federal government [131]. The (originally concurrent) federal competence thus becomes, in this respect, and within a restricted area of legislation, exclusive as from the moment the sub-national entity is barred from legislating. It does not, however, become an exclusive competence in its technical sense [132]. There remains, in fact, a major difference between those two types of "exclusive" competences. Whereas the concurrent competence turned "exclusive" by being used by the federal government can be reopened to the sub-national level simply by being "unused", an exclusive competence in its technical sense can be made concurrent and, therefore, open to the sub-national level only by an amendment to the rules on competence allocation which generally form part of the constitution. Thus, an exclusive legislative competence differs from a concurrent legislative competence that has been used pre-emptively insofar as the latter is reversible by means of general legislation [133], whereas the former is reversible only by way of constitutional amendment. Their exclusive effect, within the area of legislation respectively covered, may be exactly the same. The decisive point, therefore, that distinguishes, under that technical conception of exclusive powers, those powers from concurrent ones used pre-emptively is to know whether a (re)transfer of competences to the sub-national level of government is possible by general legislation on the national level, or only by constitutional amendment.

Under the German Basic Law, exclusive legislative competences are only those labelled just that. It is, therefore, undisputed that a concurrent competence used pre-emptively is not to be equated with an exclusive competence [134]. Similarly, under the U.S. constitution, there is no doubt that the federal government, after having legislated in a certain area and even having pre-empted that field, has the power to abolish that legislation, even without replacement, and by that means to make possible the States' use of their (re-established) concurrent competence. That power was absolutely essential, e.g., for President Reagan's policy of "new federalism" [135]. It is obviously not compatible with the idea of a kind of conversion of concurrent into exclusive competences by the federal government's use of the former.

bb) The conceptual distinction between an exclusive competence in its technical sense and a concurrent competence used pre-emptively and, therefore, turned in certain respects exclusive has repeatedly been disregarded in the discussion of Community law [136]. The Commission even appears to claim that the concept of exclusive powers was developed autonomously by the Community. "Historically", it states, "the concept of exclusive competence originally grew out of the obligation to establish the `common market', which was spelled out in very firmly binding terms ..." [137]. Therefore, the question is whether "exclusive competence" within the meaning of Art. 3b (2) ECT encompasses a concurrent competence used pre-emptively. The answer must duly take into regard the subsidiarity principle as provided for in that provision.

The restriction of the scope of application of the subsidiarity rule in Art. 3b (2) ECT to non-exclusive Community competences finds a self-evident [138] explanation if one considers the concept of exclusive competence in its technical sense. In an area in which the Member States have, by reason of such an exclusive Community competence, no power to legislate anyhow, an application of the subsidiarity rule would be obviously meaningless [139]. Seen in that perspective, the restriction put, in Article 3b (2) ECT, on the application of the subsidiarity rule makes perfect sense. The same, however, cannot be said if one considers the wider meaning of exclusive competence encompassing a concurrent competence used pre-emptively.

In the latter context, it is undisputed that the subsidiarity rule does apply as long as the competence has not been used pre-emptively by the Community. To borrow the words of the ECJ, this is so as long as no measures have been adopted "of such a kind as to deprive the Member States of an area of competence which they were able to exercise previously on a transitional basis" [140]. Disputes may arise however once the "critical mass" of Community legislation has been achieved, i.e. once Member State legislation is pre-empted and the Member States are therefore prevented from legislating anyhow. It could be claimed that from that time on an application of the subsidiarity rule again would be meaningless. Such an interpretation, however, would have an obvious disadvantage. It would burden the already difficult application of Art. 3b (2) ECT with the additional issue of whether a concurrent competence has been used pre-emptively by the Community and turned, by that fact, exclusive. The opposite interpretation under which the subsidiarity test is also to be applied in cases of Community pre-emption is better in bearing with the fundamental imperative behind the suiarity principle as principle, i.e., that "decisions are taken as closely as possible to the citizen" [141]. The test may reveal that no Community measures are needed [142], or even that the exercise of some of the concurrent competences involved can be turned back to the Member States. The case under consideration therefore has a closer resemblance to any other case in which the Community has exercised its concurrent competences to a certain degree than it has to a case of exclusive competence in its technical sense. In all these cases, there may be valid reasons, once a certain amount of measures in a given area has been taken on the Community level, to enact further measures on the same level. These reasons can be accomodated on the basis of a concurrent Community competence the exercise of which would be protected by the concepts of supremacy of Community law and/or of Community pre-emption. But it is of the very substance of Article 3b (2) ECT that those reasons must be checked against the subsidiarity rule, and there is no compelling reason to refrain from such a check just because a "critical mass" of Community legislation has been achieved [143].

The disadvantages of the contrary opinion are, perhaps, most clearly demonstrated by the Commission's stance. At the outset, the Commission has to admit that under the concept of exclusive powers it uses "it is impossible, legally speaking, to determine whether all these obligations to act entail exclusive powers for the Community ..." [144]. So, in the Commission's contention it is impossible, legally speaking, to determine to which legislative projects the principle of subsidiarity applies. The Commission then procedes to claim that "it is possible to speak of a genuine obligation to act leading, in the course of time and through the rulings of the Court of Justice, to the formation of a block of exclusive powers ..." [145]. That is to say the Commission admits (i) that it is impossible to determine the area covered by exclusive powers (wider concept) and (ii) that the powers claimed might, at most, develop into such exclusive powers over time. Nevertheless, it concludes that within that block of exclusive powers which it considers suddenly as existing and not any more as merely potentially evolving, "the Community does not have to demonstrate the need for action ..." [146], i.e. is exempt from the application of the subsidiarity rule. To rephrase the Commission's point of view, we do not know which powers are exclusive, and we hope (and trust) that, with the help of the ECJ, over time a "block of exclusive powers" will develop, but we will treat the Community powers as exclusive anyhow, at least for the application of Article 3b (2) ECT. Considering the subsidiarity principle's importance under the Maastricht Treaty, that is not an acceptable contention [147].

e) The concept of exclusive competence in Art. 3b (2) ECT as a term of art

Therefore, on the better construction, it is submitted, the concept of exclusive competence in Article 3b (2) ECT is to be understood according to its technical meaning as discussed above [148]. It remains to be seen which Community competences can be regarded, under this concept, as exclusive ones.

2. Exclusive Community competences

Which areas, then, can be considered to be covered by exclusive Community competences in their technical sense? As is well known, the EC Treaty does not expressly designate such areas. On the other hand, there can be no other source for such a competence but the treaty [149]. What is necessary, therefore, is an interpretation of that treaty [150]. Such an interpretation must be based on the concept of exclusive competence in its technical sense taking due account of the principle of subsidiarity. Two requirements must be met. The competence must be granted for a certain legislative area, which amounts to more than the power merely to regulate specific points or aims within such an area, and it must be exclusive in its technical sense.

a) On the basis of the second requirement, an interpretation according to which any potential competence of the Community, i.e. any competence transferred to the Community at all, is considered an exclusive one, is clearly not tenable [151]. Rather, only a competence under which Community legislation is made, by the EC Treaty, not only possible, but mandatory, can be considered an exclusive one [152]. As long as Community legislation is only possible, the Community has the power to abolish its legislation and, by that means, to allow the Member States again to legislate on that field. In all those cases, however, in which Community legislation in a certain area is mandated by the EC Treaty, the Community does not have the power to abolish legislation once enacted without replacement, at least in the absence of a Treaty amendment [153].

b) Nevertheless, on the basis of the first requirement, not all such competences can be considered exclusive competences. The main problem with the categorization of Community competences lies in the fact that there exist, in the EC Treaty, alongside each other two different types of competences. Whereas in normal federations competences are granted to the different levels of government by areas or fields of legislation [154], the original EEC Treaty knew, in principle, only competences granted to the then EEC to achieve certain ends [155]. That model is considered to have been advantageous to the construction of the Community legal order [156]. However, it has made the delimitation of Community and Member States competences difficult. Since the Single European Act, the Community has been granted, in addition to the above type of competences, further competences more in the mould of area-related competences as known to normal federations.

Competences to achieve a certain end are, as such, not apt to be exclusive competences in their technical sense. It is of the very nature of a competence granted to achieve a certain end that it does not completely cover a certain field of legislation. Whilst the concept of "field of legislation" is not well defined and may designate vast areas e.g. "the private law" as well as restricted ones e.g. "the law of scholarships and the promotion of academic research" [157] , the competence to achieve "the abolition, as between Member States, of obstacles to freedom of movement for persons, services and capital" [158] clearly does not designate an area of legislation but certain measures to be taken within given areas, e.g., in the provision quoted the areas of free movement of persons, services and capital. It cannot even reasonably be assumed that the Community competence for such abolition measures granted by Article 3 (c) EECT (original version) was exclusive, even when leaving the area requirement out of account. Measures by the Member States themselves doing away with certain obstacles to free movement could not conceivably be regarded as contrary to Community law, at least not in the absence of relevant Community measures [159].

Therefore, in principle, Community competences of the kind granted in the original EEC Treaty cannot be considered as exclusive. This view is amply borne out by the pre-Maastricht treatment, academic as well as judicial, of the issue. The concept of supremacy of Community law, the very heart of the ECJ's treatment of the relationship between Community and national legal orders, is meaningful only on the assumption of concurrent legislative powers of the two levels of government, and the most engrained, although misconceived [160], dispute at least in the German treatment of the issue concerned whether the Community law's supremacy excluded the validity [161], or merely the applicability [162] of incompatible national law.

c) An interpretation along these lines reveals as exclusive the Community competences for questions of common commercial policy (Article 113 ECT) [163] and of the common customs tariff (Article 28 ECT) [164]. In those areas, the rules of the (then) EEC Treaty (as amended) resembled closely the Swiss regulation of deferred exclusive competence [165]. They covered a certain area, and they provided for a phasing out of Member States' competences until the end of a transitional period [166] and therefore for an exclusive Community competence that became effective after the end of that period. Since then, Article 28 ECT grants the right to decide questions of duties in the common customs tariff exclusively to Community institutions and, according to Article 113 (2) ECT, the same is true for the implementation of the common commercial policy [167].

Other candidates for being considered as exclusive Community competences are the legislative competences for the establishment of a common organization of agricultural markets (Article 40 (2) ECT) [168] and, since the coming into force of the Single European Act, for the establishment of the internal market (Article 7a ECT) [169]. Both of them share, to some extent, with the two areas considered above the trait of commanding the Community to achieve certain goals by the end of a certain period. However, there are some difficulties linked with the notion that they are exclusive competences in their technical sense. As far as the establishment of a common organization of agricultural markets is concerned, the Community competence is exclusive with the minimal contents of para. (a) of Article 40 (2) EECT only [170]. The exercise of the Community competence is prescribed only in the limits of para. (a) [171], and, therefore, irreversible only in these limits [172]. A European market organization under para. (c), once established, could be abolished and replaced by common rules under para. (a) and, say, a consented co-ordination of re-established national market organizations.

As far as the establishment of an internal market is concerned, the Commission maintains that that establishment is an exclusive competence. The obligation imposed on the Community by Article 8a EECT [173] to "adopt measures with the aim of progressively establishing the internal market", it claims, is an "obligation to act ... clearly and precisely imposed by the Treaty itself", an obligation it considers to constitute the functional element of an exclusive power [174]. This view may have been arguable in the period between the coming into force of the Single European Act and the Maastricht Treaty, when no textual connection between Art. 3 (c) and Art. 8a EECT did exist. However, the "abolitionist" Community competence under Article 3 (c) EECT has been amended by the Maastricht Treaty. At the beginning of the paragraph, the words "an internal market characterized by ..." have been added so that it now reads "... the activities of the Community shall include ... an internal market ...". It has therefore to be asked whether the competence to establish an internal market characterized by the abolition of obstacles to free movement, since it is in this way that Art. 7a ECT now must be read, is a competence different in kind from the competence simply to abolish these obstacles, characterized above as non-exclusive. The answer clearly must be in the negative. The Maastricht Treaty has "characterized" the establishment of the internal market, in Article 3 (c) ECT, as a deregulatory exercise. The power granted to the Council in Art. 100a ECT to "adopt the measures for the approximation of [national] provisions ... which have as their object the establishment and functioning of the internal market" is, therefore, a power only to remove obstacles to the working of this market. Of course, this removal may include the enactment of unitary rules [175]. But this fact does not change the nature of the exercise. It follows that for the establishment of an internal market, for the reasons given above [176], an exclusive competence of the Community cannot be assumed.

This conclusion is reinforced by an effet utile argument. To consider the establishment of an internal market an exclusive competence in its technical sense would tend to reduce the subsidiarity rule to naught [177]. The establishment of an internal market can reach most or all areas of legislation. If the Community had exclusive competence in all these areas, the subsidiarity rule could there not apply. This result would be incompatible with the effet utile principle [178]. Further, the establishment of an internal market does not fit into a comparative scheme of exclusive competences generally granted to a federal government [179]. Whilst that is clearly not a dogmatic argument, it tends further to confirm suspicions already entertained as to the well-foundedness of a claim according to which this is an exclusive competence [180].

Therefore, of the Community competences granted to achieve a certain end, only the competences to establish a common customs tariff, a common commercial policy and common rules on competition in the framework of the common agricultural policy can be considered as exclusive [181]. What, then, of the area-related Community competences? Such competences were added to the original EEC Treaty mainly by the Single European Act, i.e. the environment (Article 130r et seq. EECT) and research and technological development (Article 130f et seq. EECT), and by the Maastricht Treaty, i.e. education (Article 126 ECT), vocational training (Article 127 ECT), culture (Article 128 ECT), public health (Article 129 ECT), consumer protection (Article 129a ECT) and trans-European networks (Article 129b et seq. ECT). The mere wording of these provisions makes it abundantly clear that the competences therein granted to the Community are concurrent competences only [182]. C. The principle of subsidiarity and the supremacy of Community law

"The application of the [subsidiarity] principle ... shall not effect the primacy of Community law" [183]. This sentence is no doubt correct insofar as the rule in Art. 3b (2) ECT is concerned [184]. The issue here to be discussed however, is the relationship between the supremacy of Community law and the subsidiarity principle as principle. There are some fields of legislation where it appears, at first sight, that the exhaustive use of sweeping Community competences, or the exhaustive interpretation of sweeping primary law provisions, may overreach certain limits of what is reasonable, or what is acceptable to the Member States. This is particularly true insofar as Member State constitutional law, or other law protecting important national values, is concerned. It is in these fields that the need for an additional Exit possibility is particularly strong and that it is therefore proposed to discuss the impact of the introduction of the subsidiarity principle on the concept of supremacy of Community law.

1. General remarks

According to the case law of the ECJ, Community law has supremacy over all Member State law, including all Member State constitutional law [185]. Whilst this proposition is rather doubtful as a matter of national constitutional law, on which, in the last analysis, Community law is based [186], it used to have a certain appeal as a matter of Community law [187]. After all, if Community law is to be applied uniformly throughout the Community, national rules of whatever level, being by definition only applicable within a single Member State, cannot be allowed to affect its application [188].

a) Supremacy of Community law is a legal technique necessary to achieve a uniform economic law throughout the Community [189]. This uniformity, in its turn, is a basic requirement of the internal market. The concept of supremacy of Community law over national law thus cannot be justified by the brute fact that Community provisions have been enacted by the Community. No consequences flow from this fact [190]. It can be justified only in view of the internal market and of the other purposes enumerated in Art. 3 ECT [191] which are still related basically to the activities of an economic community [192]. Indeed, when the original Member States concluded the EEC Treaty, they intended to create, in the interest of their industries, a bigger market. At that time, the idea of supremacy of Community law was not yet born. Thus, the EEC Treaty has not been the expression of considerations about the hierarchical relationship between economic provisions of this treaty, on the one hand, and aspects of national public policy [193], on the other. There was never any intention to grant economic law provisions, simply because of their enactment on Community level, supremacy over all other aspects of national public policy [194].

Insofar as aspects of national public policy remain within the exclusive legislative competence of the Member States, they are indeed not affected by the concept of supremacy of Community law developed only later on. However, it appears that today there is, in fact, no such thing as an exclusive Member State competence that were immune against Community legislation [195]. Even in the case of "normal" federated states in which the respective competences of the different levels of government are allocated on the basis of a catalogue, a clear distinction of fields of legislation within and without the catalogue is virtually impossible [196]. This difficulty is compounded in the case of the relationship between the Community and the Member States because many of the competences granted to the Community in the EC Treaty are granted not in view of a specific field of legislation but to achieve a certain end [197]. Whilst such competences cannot be considered as exclusive competences of the Community [198], it is also true that they do not permit the existence of truly exclusive competences of the Member States. Competences granted to achieve a certain end can invade virtually any field of legislation [199]. But if virtually no Member State competences are out of reach of Community legislative measures it must be asked where this leaves us with regard to the supremacy of economic Community law over basically non-economic measures of national public policy.

b) Supremacy of Community law is generally seen as a rule strictly to be applied in all cases of conflict between Community and national laws [200]. Indeed, that is the solution generally adopted by federal systems. Comparative law may show some reasons for that solution, and some adaptations undertaken by different legal systems. In Germany, e.g., the provision of Art. 31 BL according to which "Bundesrecht bricht Landesrecht" (federal law shall prevail over state law) is construed, within its sphere of application, as a strict rule. However, the German constitutional system provides for a rather strict separation of federal and Land competences for legislation on sub-constitutional levels [201] and therefore the sphere of application of Art. 31 BL is rather restricted [202]. In particular, the provision does not apply to conflicts between federal statutes enacted without federal competence (and therefore void per se) and Land law [203]. Further, there is no doctrine of absorption. In the U.S., the supremacy of federal law expressed in the supremacy clause of Art. VI (2) of the Constitution has a much wider sphere of application, as the area of concurrent competences in the U.S. is much wider than in Germany [204]. Here, too, the supremacy clause is in principle read as a rule. However, in its effects that rule is somehow reduced because it is a principle of interpretation that "if Congress is authorized to act in a field it should manifest its intention clearly ... The exercise of federal supremacy is not lightly to be presumed" [205]. It could even be said "that to the extent that subsidiarity has any parallel in U.S. federalism, that parallel lies in the law governing federal preemption of state law" [206]. In conclusion, whilst both legal systems considered know of the concept of federal supremacy and construe it as a rule, the concept's sphere of application is, in both countries, much reduced. In Germany this is so because the principle apportionement of powers relies on the concept of enumerated rights of the federal government and in the U.S. because the application of the supremacy rule is not a foregone conclusion but dependent on a clear manifestation of congressional intent to that effect.

Conversely, in the EC both these reductions of the effects of a supremacy concept understood as a rule do not apply. The exercise of powers by the Community is, at its present stage of development, scarcely hampered by a doctrine of enumerated powers [207], and Community law supremacy functions, according to the ECJ's case law, at least ante-Maastricht, per se [208], without any restrictive theory of interpretation being applied [209]. This state of affairs was, at least prior to the introduction in the EC Treaty of the subsidiarity principle, another impressive example for an eminent scholar's findings according to which the EC has, in some respects, developed way beyond anything to be found in federal states [210]. It is proposed to discuss here whether, and to what extent, if any, this position might have been changed, in relation to basically non-economic matters of national public policy, by the express sanctioning of the subsidiarity principle brought about by the Maastricht Treaty, and its insistence on decisions being taken as closely to the citizen as possible.

2. The concept of supremacy of Community law as rule and principle

a) The concept of Community law supremacy and the subsidiarity principle are expressions of contradictory evaluations of the relationship between Community and national laws. According to the subsidiarity principle [211] decisions should be taken if possible at a national level, not at the Community level, as is made clear by Art. 3b (2) ECT for the special constellation covered by that provision. On the other hand, once conflicting decisions exist on the national and Community levels, then, according to the concept of supremacy of Community law as developed in the ECJ's case law, the latter prevails.

On the level of secondary Community law some balance between those two contradictory evaluations is struck by the rule in Art. 3b (2) ECT permitting Community legislation only if, among other things, national legislation would not suffice to attain the objective of this particular piece of legislation [212]. If this requirement is indeed met in a given case supremacy of the Community legislation accordingly enacted basically protects the very objective that justifies its enactment in first place. In other words, if national legislation does not suffice to achieve the objective of a given piece of Community legislation, it must not be allowed to prevent or to hinder the latter's implementation. However, the subsidiarity rule applies only to secondary legislation, and even there it looks only at the objective of the Community legislation in question. It neither takes account of the respective legal importance pertaining to the overridden national legislation in view of its contents, nor is it applicable to a conflict between primary EC law and national law. Under both these aspects, the contradictory evaluation of the subsidiarity principle and Community law supremacy remains unresolved.

b) It is the very rationale of Community law supremacy to provide for a uniform economic law throughout the Community. Inversely, supremacy of Community law gets doubtful in cases in which conflicting national law does not endanger, or only marginally endangers, the achievement of the purposes enumerated in Art. 3 ECT. Indeed, against the comparative background described above, it is not plausible that in all those areas in which the Community may arrogate legislative competences no decision taken more closely to the citizen is possible throughout the whole area thus arrogated. There must be constellations in which it is perfectly conceivable that, although a Community measure enacted within Community competences contains provisions for a certain set of facts, the final decision on that set of facts is better, or as well, taken "more closely to the citizen", i.e. on a national level of government. If the subsidiarity principle can protect such decisions against Community law supremacy, this is the way in which it can play its rôle as offering an additional Exit possibility.

It is not easy to define these constellations more exactly. Their origin lies in the unique way the Community's competences are granted, i.e. as competences to achieve certain ends [213]. There is therefore no real limit to the competences of the Community. Basically, it must be said that secondary Community law, once enacted within Community competences, primes, in case of conflict, national law enacted even outside these competences, for the simple reason that there is no area "outside these competences". Similarly, sweeping provisions of primary Community law prime in principle all conflicting national law regardless of the legislative purposes pursued by that law. It is, however, evident that these results clash with the subsidiarity principle as principle. Learned authors, and the Member States, have tried to avoid it. So it has been claimed that the principle of supremacy of Community law over national law is a principle "whereby each law is supreme within its sphere of competence" [214]. Supremacy of Community law outside this "sphere of competence" has then been dubbed "absorption" [215]. As there is no circumscribed sphere of Community competences, however, this concept of absorption appears to need some further development.

The Member States have claimed priority for non-economic interests, protected by national law, over economic interests protected by Community law [216]. This is, it is submitted, a more promissing approach to the problem. If one looks not at the sphere of Community competences which is unrestricted anyhow because of the very nature of these competences but at the conflicting national measures one may come up with a viable definition of a sphere of absorption. In all cases in which the national measure is not intended to regulate a subject matter dealt with in Art. 3 ECT (or is intended to deal with such a matter in a way which is, because of other provisions of the EC Treaty, outside a direct authorization of the Community) a possible conflict of this national measure with EC law is situated in the sphere of absorption. As the starting point of this analysis is the conflicting national law, it can be applied to conflicts both with primary and secondary Community law. The question is then one of the consequences.

c) Supremacy of Community law is best seen as a rule insofar as a conflicting national norm directly concerns one of the purposes announced in Art. 3 ECT. Here, the subsidiarity principle is not applicable; clearly a decision taken more closely to the citizen is not possible in this area without threatening the rationale of Community law supremacy, i.e. a uniform economic law. Even so, some doubts remain. Take, for example, the conflict between legislations referring to two different aspects of economic policy, e.g., the law of co-operative corporations and competition law. The relationship between those two areas of legislation is problematic in some legal systems to which both concepts are known [217]. Within the German legal order, e.g., competition law provides for some exceptions from general competition rules for certain types of co-operatives. For the rest, the courts apply competition law to co-operatives insofar as clauses in their statutes that restrict competition are not indispensable for their functioning and purposes [218]. In doing so, the courts rely on a discussion of substantive law, as both areas of legislation are governed by statute law [219]. A similar conflict between Community competition law and Member State (Danish) co-operative law had to be decided by the Court of First Instance [220]. The Court there discussed the substantive aspects of the conflict to some degree [221], but finally relied on Community law supremacy [222]. As a consequence, competition law prevails mainly because the law of co-operatives has not (yet) been made the subject of Community legislation [223]. Even if such an outcome is unsatisfactory, it is probably, in an area clearly governed by Art. 3 ECT, inevitable.

On the other hand, in cases situated within the sphere of absorption the rationale of Community law supremacy does not justify to consider that supremacy as absolute. This is most clearly seen in cases in which the relevant national set of provisions is, because of its contents, of primordial importance under national constitutional law. The idea of national constitutional values being primed by, e.g., technical Community provisions on the right of establishment is repugnant even to partisans of Community law supremacy [224]. However, that cannot mean that Community law supremacy is simply to be denied in these areas. The contrary, even insofar as a conflicting national norm does not directly concern any of the purposes announced in Art. 3 ECT, the rationale of Community law supremacy cannot simply be disregarded; a uniform application of EC law throughout the Community remains, even in those cases, a worthwile effort. On the other hand, the express sanctioning of the subsidiarity principle in the Maastricht Treaty has pushed a different point of view to the fore, i.e. that matters should be dealt with as closely to the citizen as possible. This point of view requires some restrictions to be put on Community law supremacy. These contradictory aspects, it is submitted, can best be reconciled by considering Community law supremacy over national law enacted in areas not directly mentioned in Art. 3 ECT, and therefore situated in the sphere of absorption, as a principle in its technical sense.

There is a parallel to this view of Community law supremacy in one theory of constitutional construction. In the case of sweeping constitutional provisions, it is possible, and has proven useful, to see one and the same provision as a rule and as a principle. The core of the provision can be considered as a rule, and around this core there may be a penumbra that can be seen as principle [225]. In such a case, conflicts between the core of the provision and other rules are to be decided according to the all-or-nothing fashion characteristic of conflicts of rules, whereas conflicts between the penumbra and other principles are to be decided according to the respective weight of the principles involved in the actual conflict. In other words, the penumbra of the norm does not automatically prevail in case of conflict with other principles, but has to be balanced against them. The same result can be deduced from the principle of effet utile [226] which, applied to the present context, demands that both principles must be granted effectivity.

This model can be applied to the concept of Community law supremacy so that here again the distinction between rules and principles proves to be helpful [227]. Community law supremacy is a sweeping concept indeed. Its core has been identified, on the basis of its rationale, as securing the unity of economic law throughout the Community. Insofar it is, as stated above, a rule, and strictly to be enforced. The concept's effects beyond this rationale, however, i.e. in the sphere of absorption, are best seen as its penumbra, as a principle having the aim to protect the unity of Community law only insofar as compatible with rival principles. In particular, the subsidiarity principle introduced by the Maastricht Treaty, the importance of which for the assumption of a principle of Community law supremacy has been demonstrated above, has posited a rival principle against the concept of Community law supremacy [228], and must therefore be balanced, in individual cases, against the latter.

d) Against a similar balancing approach used by the German Federal Constitutional Court in a different context, it has been claimed that to weigh the competing interests in an ad hoc manner diminishes the predictability of decisions [229]. This is a serious question. Predictability, it is submitted, is best served if the deciding authority here: the ECJ follows, in deciding individual cases, a well settled set of standards and values and a uniform method of reasoning from these standards, and if this reasoning is embodied, and therefore made public, in the decisions themselves. Gut decisions, and decisions that do not reveal the true reasons for which they have been reached [230], make predictions very difficult or even impossible. It is under these aspects that the predictability of decisions should be discussed.

It is thought that the balancing test allows to discuss, in a structured manner, those points of fact and of law that are, or may be, decisive for the solutions found. It appears even possible that, on the basis of this test, it would be easier for judges with different legal backgrounds to agree not only on results but also on the reasons for those results, given that the balancing test is largely free of legal technicalities. The reproach to the balancing approach quoted above was aimed specifically to the German model of balancing concretely, in each individual case, in the context of free speech conflicting with provisions on defamation, the respective interests at stake; these interests were, on the one hand, the individual plaintiff's reputation and honour and, on the other hand, the value or importance of the particular speech [231]. Here, therefore, brute facts must be balanced against one another. However, in the context of balancing Community law supremacy against the subsidiarity principle, protecting respectively abstract Community and national values, the situation is quite different. In view of the abstractness of the values involved, the evaluation of brute facts will generally not be necessary in coming to a conclusion. Therefore, it is submitted, predictability of decisions is not restricted, but rather enhanced, by the approach here advocated.

e) The conflict between the supremacy and subsidiarity principles is regularly superimposed on a different conflict between Community and national law. The principle of Community law supremacy militates generally for whatever piece of Community legislation is, in an individual case, in conflict with national law. Inversely, the subsidiarity principle militates for the conflicting piece of national legislation. It is from this underlying conflict that the competing principles of subsidiarity and Community law supremacy gain their respective weight in the individual case. In the final analysis, the weight of both principles will depend on the national law involved: on the one hand, on the closeness of the national law against which Community law supremacy is invoked to the targets of the Community enumerated in Art. 3 ECT, and, on the other hand, on the importance of the national value protected by the subsidiarity principle [232].

One further consequence of this superimposition of two levels of conflict should be noted. The reduction of Community law supremacy to a principle, where appropriate, is by necessity reflected in a parallel (partial) reduction of the Community law provision protected by it. This provision must be seen as a principle, or partly as a rule and partly as a principle, too. The whole reasoning above is founded on the underlying Community legislation. It is, in the last analysis, the relationship between that legislation and the conflicting national law that has led to the conclusion that Community law supremacy must be reduced, in appropriate cases, to the status of a principle. It would even be possible to sustain the whole argument without recourse to the principle of Community law supremacy. The argument can validly be based on the subsidiarity principle used as an instrument of interpretation [233]. As such, the subsidiarity principle mandates a restrictive interpretation of sweeping Community law provisions. Once a substantive provision of Community law is, by means of such an interpretation, itself reduced to the status of a principle, there is nothing left that, even if read together with a strong form of Community law supremacy, could prime contradicting national law [234]. However, it is felt that it is more convincing to posit two fundamental but contradictory principles of EC law against one another than to put the subsidiarity principle against a host of different individual provisions. In addition, this is, it is submitted, the only way an overarching argument can be made without getting fettered down by minute legislative details. But the relationship between these two levels of conflict must be borne in mind when we now start discussing the application of the principle of Community law supremacy.

3. The application of the principle of Community law supremacy

a) The balancing of the subsidiarity principle against Community law supremacy in the sphere of absorption is best demonstrated by a brief discussion of the pivotal absorption case, Casagrande [235]. In this case an educational grant [236] had been refused, by a German authority, to an Italian national, according to a provision in a German statute providing such grants, in principle, to German nationals only. On the other hand, according to Art. 12 of Council Regulation 1612/68 [237], "the children of a national of a Member State who is ... employed in the territory of another Member State shall be admitted to that State's general educational ... courses under the same conditions as the nationals of that State, if such children are residing in its territory". Regulation 1612/68 was based on Art. 48 and 49 EECT concerning migrant workers. There was no Community competence on education as such [238]. However, it does not need a lot of phantasy in interpretation to accept (a) that Art. 12 of the Regulation did cover educational grants and (b) that Art. 12, thus interpreted, could be validly based on Art. 48 and 49 EECT. Nevertheless, the point remains that this construction of Community law impinged on Member State competences in areas in which there was no direct Community competence under Art. 3 EECT. Hence, the conflict between Community law and national law was situated in the sphere of absorption, and it is a case in which Community law supremacy would have to be regarded as a principle to be balanced, after the introduction of the subsidiarity principle, against that latter principle.

What, then, are the respective interests here at stake? They can best be identified, on the one hand, as the Community interest to guarantee national treatment to migrant workers and their families and, on the other hand, as a national budgetary interest in not extending national treatment to those workers under the circumstances of the case [239]. Particularly, it cannot be claimed that the national interest at stake is of overriding importance. Although educational policy as such may thus qualify, under the circumstances of the case the determination of that policy is not tempered with by Community law which only extended the circle of possible beneficiaries of this policy [240]. A balancing of the identified interests at stake therefore clearly leads to the conclusion that, under the circumstances of the case, Community law supremacy outweighs the national interests protected by the subsidiarity principle. Hence, there would be no reason to decide Casagrande differently today than it was decided in 1974.

b) Other cases might have to be treated differently. One of the most contentious areas of confrontation between Community law and national law in the recent past and still at present is the Community claim to a competence to regulate aspects of television [241]. There is no doubt that television policy can be a part of cultural policy [242]. According to Art. 128 ECT, the Community power on "Culture" is severely restricted [243]. The cultural aspects of television are therefore generally within the legislative competence of the Member States. On the other hand, very important economic interests are involved in television, in particular in advertising and property rights in films. Whilst it is clearly within the Community competence, in pursuance of the targets enumerated in Art. 3 ECT, to legislate on these interests, conflicts between such legislation or similar primary EC law and Member States' cultural competences are situated within the sphere of absorption [244]. According to the above analysis, therefore, in this area Community law supremacy is reduced to the status of a principle, and has to be balanced against the subsidiarity principle offering the Member States an Exit possibility in cultural questions and protecting their cultural competences against Community encroachments.

This analysis shall be tested by applying it to the facts of the Veronica case [245]. According to Dutch law, companies having received "sending slots" by the competent public authority were forbidden to exercise activities not authorized by that authority. Income created by permitted activities had to be employed in realizing broadcasting programs [246]. It was the obvious aim of the law to prevent broadcasting companies from using public money, or the income thus created, for commercial purposes and not for those cultural purposes for which that money had been granted [247]. The sender Veronica had participated in the setting up of a commercial sender in Luxemburg whose programs were directed towards the Netherlands. The media authority prohibited that participation. In doing so, it prohibited Veronica in effect to offer its services to a company based in another Member State, and that company to accept Veronica's offer.

There exists therefore a conflict between the principle of free provision of services on the one hand and national cultural interests on the other. The Dutch law's target was clearly not connected to one of the targets enumerated in Art. 3 ECT [248]. In other words, the internal market interest in the free provision of services clashed with the entire Dutch system of cultural mandates to television companies. The weight of the subsidiarity principle in protecting this national interest, in this context, is great. It is the content, even the possibility of cultural policy in this area which is at stake, not only, as in Casagrande, some budgetary interest. On the other hand, the internal market interest in the free provision of services is only marginally affected. It is only a very specialised, and very restricted, type of services that is concerned. The balance, therefore, here is clearly in favour of the subsidiarity principle and the national interest it protects.

The ECJ arrived at the same result. It held that Veronica abused the free provision of services provisions of the EC Treaty in relying on them to get rid of the Dutch restrictions [249]. It is respectfully submitted that this reasoning raises the unanswerable question of the distinction between use und abuse of such a freedom [250]. The action of Veronica could only be considered as abuse once it had been decided that the Dutch legislation, i.e. the setting up of a pluralistic, non commercial broadcasting system, should prevail in the case before the Court. As to this decisive question, the Court claims that the objectives pursued by the Dutch legislation are objectives of general interest which a Member State may legitimately pursue [251]. In one former decision taken in reference by the Court [252] this interest had been defined as connected with the freedom of expression "which is one of the fundamental rights guaranteed by the Community legal order". By that means, the legitimacy of the Dutch cultural policy was based on its "connection" with a fundamental freedom supposedly protected by the Community legal system. It is doubtful whether this approach will allow for predictable results as some undefined "connection" with some fundamental freedom will exist in most cases but will not necessarily be decisive. More to the point, this approach completely disregards the new dimension of the subsidiarity principle which had, it is true, not yet expressly been enacted at the time of the judgment discussed. Since this enactment, it is submitted, conflicts like that in Veronica are best solved by applying the subsidiarity principle [253].

c) The by now classic example of a conflict between the internal market freedoms and highly valued national concepts, avoided by the Court of Justice [254], was, of course, the conflict between the Community freedom to provide services, guaranteed in Art. 59 (1) ECT, including medical services legally provided in any Member State [255] and including, therefore, according to national law, abortion services [256], on the one hand, and the human right of the fetus not to be aborted, specifically protected, on the basis of a constitutional amendment, by the Irish constitution [257]. Here, the internal market interest in the free provision of services clashed with a human right protected by national law. This right, clearly, had nothing to do with any of the targets enumerated in Art. 3 ECT; the clash was in the sphere of absorption. The weight to be given to the subsidiarity principle protecting this national choice is obviously great. The Irish constitution's prohibition on abortion was of primordial importance for the Irish legal system. On the other hand, the internal market interest in the freedom to provide services could only be marginally affected by the Irish prohibition. This remains even true if one looks only at abortion services. The business of British abortion clinics that originated in Ireland could only be a tiny percentage of their overall business [258]. Here again, the balance is clearly in favour of the subsidiarity principle and the national interest it protects. As a consequence, if subsidiarity is to be taken seriously, as it must, it is obvious that the decision on the admissibility of abortion, and its limits, and of advertising for abortion facilities must be left to the national level [259], and cannot be taken over by the Community level under the pretext that the decision affects the freedom to provide services to foreign residents [260]. By causing the Community law's supremacy to be balanced against the subsidiarity principle in those areas where there is only a reduced rationale for the former, the latter allows for a solution which, as established by the Protocol quoted above [261], appeared only natural to the High Contracting Parties when concluding the Maastricht Treaty, and which offers them, under a different perspective, an additional Exit possibility [262].

4. In particular: Art. 30 ECT and the principle of subsidiarity as principle

a) A special case, to be considered separately because of its central importance to the internal market, is presented by Art. 30 ECT. Art. 30 ECT provides that "[q]uantitative restrictions on imports and all measures having equivalent effect shall ... be prohibited between Member States". This provision appears to contain a rule prohibiting "[a]ll ... rules enacted by Member States which are capable of hindering, directly or indirectly, actually or potentially, intra-Community trade" [263] (the Dassonville formula). According to this interpretation, the prohibition in Art. 30 ECT should cover any national provision concerning the production or the sale of goods as well as secondary costs, e.g. for responsibilty or warranty, and that irrespective of the targets at which the national provision may aim. Any such regulation may somehow, and does therefore potentially restrict imports [264]. Under such an interpretation, read together with the concept of Community law supremacy, national provisions of any kind would have to conform to Art. 30 ECT, or would only be possible as limitations to Art. 30 ECT insofar as such limitations are admissible [265]. However, there is general agreement on the point that such a result has somehow and in some measure to be avoided [266], even if there is agreement neither on the method to avoid it nor on the result aimed at. As the result to be avoided appears, at first sight, to be contrary to the subsidiarity principle as principle [267], it is proposed to analyze the question whether, since the adoption of the Maastricht Treaty, the application of that principle can render a convincing solution to the problem [268], and to outline the resulting solution. To streamline the presentation, the legal fact that Art. 30 ECT can play its traditional rôle only if read together with the concept of Community law supremacy will not specifically be repeated in this analysis [269].

Art. 30 ECT must, of course, be the starting point. Under the reign of the subsidiarity principle, Art. 30 ECT, which forms part of the title on the free movement of goods, should not be construed more sweepingly than is strictly necessary to protect that free movement [270]. The sweeping prohibition found in this provision must therefore be construed as applying in its strictest sense only to such national measures of which it can be said with a reasonable degree of certainty that they hinder the free movement of goods, i.e. those measures intended to restrict imports [271]. These measures encompass evidently, but not exclusively, all quantitative restrictions. Such measures, as has been argued above more generally [272], must be taken as automatically forbidden by a rule contained in Art. 30 ECT. Other measures which in fact affect imports do so quite often only incidentally while aiming at different ends. Under a restrictive construction of Art. 30 ECT, such measures should not automatically be considered as forbidden measures of equivalent effect [273].

On the other hand, such measures cannot automatically be exempt from Art. 30 ECT, either [274]. Indeed, if this were the end of the analysis, free trade between Member States would be wide open to attacks by national measures restricting imports, maybe severely, but not intended to do so. There is here a clear case for the application of the concept of absorption. Conflicts between Art. 30 ECT and national measures not intended to restrict imports but pursuing other public policy aims and restricting imports only incidentally fall into the sphere of absorption [275]. It follows that whilst it must be accepted, and, as stated, has been accepted by the Court of Justice and academic writers [276], that the all-or-nothing solution of a rule prohibiting all restrictions on imports of whatever kind is not adequate in such cases, it also must be accepted that it is not compatible with the idea of an internal market to let such measures go completely unchecked [277]. Therefore, in addition to a rule under Art. 30 ECT prohibiting quantitative restrictions and measures of equivalent effect but restricted to measures intended to restrict imports, a principle must be postulated according to which import restrictions should be avoided even if they are the unintentional consequence of measures pursuing different policy aims (the Art. 30 ECT principle) [278].

This Art. 30 ECT principle must be balanced, in the individual case, against the subsidiarity principle as principle [279] according to which decisions should be taken as closely to the citizen as possible. This principle enclines the decision of a conflict between Art. 30 ECT and national provisions towards upholding the latter. Indeed, it is obvious that a decision taken by a national legislature to deal with a national public policy issue is taken more closely to the citizen, in respect of time, of the level of government acting and of the closeness to the facts, than the EEC's Founding Fathers' decision embodied in Art. 30 ECT. The subsidiarity principle protects, in this respect, basically the national legislatures' power to choose, within the limits of their competences, between different policy alternatives. Indirectly, it protects the choice made by them and, by the same token, the values or principles underlying the chosen rule. It generally does not permit the Community institutions to second-guess the respective national legislature's decision; to hold differently would amount to denying the very basis of the subsidiarity principle [280]. It follows that, in the case of a contradiction between a national provision reasonable in itself, enacted in the public interest as perceived by the national legislature, not intended to restrict imports, and doing so only potentially or very slightly, the subsidiarity principle weighs heavily in favour of upholding the national provision. In such a case, the Art. 30 ECT principle's weight is comparatively light so that the national provison should prevail in accordance with the subsidiarity principle. In the reverse case of a contradiction between a national provision enacted mischievously and evidently not in the public interest, or restricting imports heavily, the balance will be in favour of an application of the Art. 30 ECT principle which has, of course, a lot of weight against any legal device employed by a Member State ultimately restricting imports in an unacceptable way. The hard cases as always will be those in which a provision is enacted in the bona fide public interest but is at the same time restricting imports heavily [281].

b) A distinction along these lines appears to be workable. The delimitation between measures intended to restrict imports and other measures affecting intra-Community commerce is a function of the purpose of the national measure which can be found out by interpreting that measure. Only if the purpose of the measure is to restrict imports, the measure is automatically forbidden as measure having equivalent effect. This test appears to be simple and, it is submitted, can be applied easily and objectively. It is felt that, if taken together with the balancing test advocated for conflicts situated in the absorption sphere of Art. 30 ECT, it can explain most of the ECJ's decisions, and can do so generally more convincingly and in a more coherent way than the reasons proferred by the Court itself. It is felt, indeed, that this model, or ideas akin to it, underlies those decisions of the Court of Justice in which national provisions were upheld [282] although, in fact, there could be no serious doubt that they would lead to some minor restrictions of commerce [283], and although the Court denied the applicability, in those cases, of a rule de minimis [284]. In the last analysis, it is felt that the above model generally explains the law as applied by the Court. For demonstration, the test will be applied to the measures underlying some cases decided by the Court [285].

The Court sometimes states that a certain national measure is, or is not, a regulation of commerce which concept appears to be similar to that, used in Art. 30 ECT, of import restrictions. The standard it applies in doing so is not altogether clear [286]; it does not appear to be related to the intention behind the national measure [287]. Among others, the Court has considered the French prohibition on resale at a loss [288] not to be a measure regulating commerce. It has seen these national provisions, among others, as being such measures: the Dutch prohibition of the offering of free gifts for sale promotion purposes [289]; the French ban on canvassing at private homes [290]; the Luxemburgish and German prohibitions on publication of the duration of a special offer or the price previously charged [291]; the English prohibition on Sunday trading [292]. Of these provisions, the Court declared the prohibition on the offering of free gifts [293] and the canvassing ban [294] to be justified, and the prohibition on Sunday trading to be proportionate [295]. Efforts of academic writers to find a coherent rationale behind these decisions were ultimately not successful [296].

On application of the standard advocated here, none of the national measures considered can be seen as intended to restrict imports [297] whether or not the Court considered it a regulation of commerce. The French prohibition of resale at a loss is based on considerations of fair competition; therefore, the rule-core of Art. 30 ECT is not affected. The Sunday trading ban's target is the national market, and it only marginally affects some factual preconditions of imports by prohibiting the sale of (among others) imported goods on Sundays [298]. As to the provisions prohibiting certain kinds of publicity [299], it is evident at the outset that they are indiscriminately applicable to imported and home products and traders, and therefore not intended to restrict imports [300]. It follows that these measures are not forbidden by the rule in Art. 30 ECT as measures of equivalent effect, and do need neither to be justified before this provision nor to pass the proportionality test. However, the national measures discussed, although not intended to restrict imports, do affect intra-Community commerce. Conflicts between them and Art. 30 ECT fall therefore in the sphere of absorption. The provisions restricting or prohibiting certain kinds of publicity permitted in other Member States [301] must be taken as indirectly restricting imports from those Member States, and affecting intra-Community commerce, by requiring, on the national territory, a different pattern of publicity [302]. Indeed, it can be said that all national rules on "selling arrangements" work somehow to the advantage of home products [303]. Concerning more specifically the prohibition of resale at a loss, were the French importers subsidiaries of a company situated in a country permitting such resales [304], they might have intended to base their whole marketing conception on such "selling arrangements" [305].

It follows that in these cases the Art. 30 ECT principle has to be balanced against the subsidiarity principle protecting the national values at the base of the publicity restrictions which can generally be seen as considerations of consumer protection to be realized by fair competition as conceived by the national legislature. It is submitted that in all these cases, in balancing the Art. 30 ECT principle against the subsidiarity principle protecting the choice of the respective national legislature and their view of consumer protection, the subsidiarity principle should, since its enactment, prevail. In those cases in which the Court upheld the respective national measure [306], the reasons given in those decisions can be applied also in the different legal context in which these cases are here discussed. In the two other cases concerned respectively with the duration of a special offer and the publication of the price previously charged, the Art. 30 ECT principle was, it is submitted, only marginally affected; it would have been sufficient, for the foreign trader to comply with the national consumer protection provisions, in the one case, to have printed separate promotion flyers for Luxembourg [307], and, in the other case, to adapt the French marketing conception of the trader concerned further to the German provisions [308], a translation of the conception being necessary anyhow. The subsidiarity principle, protecting here the national legislature's choice of the mode of consumer protection, on the other hand, was gravely disturbed. National consumer protection measures of long standing, which had formed deeply engrained habits of industry and consumer, would have to be changed. The respective legislatures' choice was evidently based on a bona fide evaluation of the consumer interests and could, under the subsidiarity principle, not be disregarded for the reason that, in both cases, only one other Member State had made the same choice. It is therefore submitted that these two cases, under the reign of the subsidiarity principle, should be decided differently, and that the national provisions previously condemned should now be upheld [309].

c) It was the aim of this part of the present article to apply the new dimension of the subsidiarity principle, i.e. the subsidiarity principle as principle, to conflicts between national law and Art. 30 ECT in order to enhance the Member States' freedom of action in the face of that sweeping provision. The application to the facts of cases decided by the ECJ has shown that the Member States' freedom on the basis of this reasoning does not reach much further, insofar as these cases are concerned, than the freedom already conceded, for apparently different reasons, in the ECJ's case law [310]. What remains is, it is submitted, a more stable fundament for a more coherent sphere of freedom which finds its basis in the subsidiarity principle as one of the general principles of Community law. It is further submitted that this is the most adequate basis for the said freedom. It is quite obvious that the question has something to do with the respective legislative powers of the Community and the Member States. As the relationship between those two levels of government is, under the Maastricht Treaty, governed by the subsidiarity principle, it is appropriate to apply that principle to the question discussed. The proposed test, i.e. the balancing of the Art. 30 ECT and subsidiarity principles, further allows for a structured discussion of the few hard cases [311]. The interests to be balanced are, on the one hand, the integration of the internal market, protected by Art. 30 ECT and, on the other hand, the integrity of the legal systems of the Member States, protected by the subsidiarity principle. The result of the balancing is, it is felt, quite independent of the question whether the national provision in possible conflict with Art. 30 ECT, is a restriction or prohibition of certain "selling arrangements" [312] or a requirement as to certain product characteristics. It is true that the distinction between those two types of provisions [313] may allow a certain approximation to the interests involved because the integrationist interest will generally militate more strongly against requirements concerning product characteristics than against "selling arrangements". However, it falls short of taking those interests fully into account, and that in two respects. It does not allow for the Member State interest in the case of product characteristic requirements, and it does not adequately allow for the Community interest in the case of non-discriminatory restrictions on "selling arrangements". Though the said distinction will render adequate results in easy cases [314] it is bound to prove inadequate in hard cases. The broader test of balancing the Art. 30 ECT and subsidiarity principles appears, therefore, in the final analysis more adequate, and preferable.

V. Conclusion

Under the Maastricht Treaty, subsidiarity has to be taken seriously. Taking subsidiarity seriously implies basically not to restrict the subsidiarity principle to the rôle of a rule, as embodied in Art. 3b (2) ECT, but to give it an additional dimension by holding that there exists, alonside this rule, a subsidiarity principle in its technical sense. This principle must be given the large meaning laid down particularily in the second penultimate preamble of the Maastricht Treaty. It serves to defend the field of application of the rule in Art. 3b (2) ECT by restricting the concept of "exclusive competence" which excludes the application of the subsidiarity principle as a rule. Additionally, it pervades the whole field of the relationship between Community and Member State competences. It restricts the absolute supremacy of Community law to conflicts with those national laws which are intended to regulate a subject matter dealt with in Art. 3 ECT, for instance to restrict imports. But it acknowledges a relative Community law supremacy, based on a balancing with the national interests at stake, in a sphere of absorption extending beyond the area described. It opens for the Member States an additional Exit possibility in all cases in which the national interest at stake outweighs the Community interest and in which a direct transfer of competences to the Community has not taken place. By these means, it promotes a balance between Community and Member State interests which it is thought will be to the lasting advantage of Community coherence.



Questions or comments about this site?

Top of the page