ARENA Working Papers
WP 00/23



Supranationality and National Legal Autonomy in the EEA-agreement

Hans Petter Graver

1. Introduction

The object of the EEA-agreement is ” to promote a continuous and balanced strengthening of trade and economic relations between the Contracting Parties with equal conditions of competition, and the respect of the same rules”. For this purpose, the aquis communitaire of the European Community regarding the four freedoms, the competition rules and rules regarding the Community's activities in the fields of:

- Research and technological development,

- information services,

- the environment,

- education, training and youth,

- social policy,

- consumer protection,

- small and medium-sized enterprises,

- tourism,

- the audio-visual sector, and

- civil protection,

are included into the agreement. [1]

The object of homogeneity is dynamic in the way that the agreement not only seeks to establish a regime of common rules, but also to maintain this, in light of the development of legislation and jurisprudence within the Community. For this purpose, the agreement establishes a mechanism for revising and supplementing the agreement following the development of new legislation within the Community. [2] Article 102 states that: ”In order to guarantee the legal security and the homogeneity of the EEA, the EEA Joint Committee shall take a decision concerning an amendment of an Annex to this Agreement as closely as possible to the adoption by the Community of the corresponding new Community legislation with a view to permitting a simultaneous application of the latter as well as of the amendments of the Annexes to the Agreement. To this end, the Community shall, whenever adopting a legislative act on an issue which is governed by this Agreement, as soon as possible inform the other Contracting Parties in the EEA Joint Committee.”

The agreement also establishes a system to ensure as uniform an interpretation as possible of the Agreement. [3] A system of exchange of information concerning judgements by the EFTA Court, the Court of Justice of the European Communities and the Court of First Instance of the European Communities and the Courts of last Instance of the EFTA States is set up by the EEA Joint Committee according to article 106. There are provisions to provide for that the jurisprudence of the European Court of Justice is given due account in the interpretation of provisions of the EEA-agreement that are ”identical in substance to corresponding rules of the Treaty establishing the European Economic Community and the Treaty establishing the European Coal and Steel Community and to acts adopted in application of these two Treaties”, namely art. 6 of the EEA agreement and art. 3 of the Surveillance and Court Agreement.

The formal status of the agreement, however, is that of an ordinary international treaty. The EFTA states have not ceded formal sovereignty to international or supranational bodies, be they of the EFTA, EEA or the EU. [4] The negotiations for the EEA agreement were based on the fact that the Constitutions of most EFTA countries give the national parliaments an exclusive right to legislate, so that direct effect of provisions under the agreement into national law would be unacceptable as part of the agreement. [5] The EEA-agreement therefor sought to establish common rules without a common supranational power. This has been characterised as mixing oil and vinegar [6] and as trying to reach for the “nearly impossible”. [7]

In a situation where EU membership is undesirable or unobtainable, the Agreement can be seen as a project of risk-reduction. It ensures equal access for economic operators to the enlarged European market. The legal obligations ensure that this is not suspecting to political whim, thus reducing economic risks by removing or at least reducing political and social risks. From this point of view the Agreement can be seen as a bargain where legal obligation and reduced autonomy are traded for reduced risks to economic operators.

Experience shows that the early observations were correct when pointing to dilemmas. On the surface, all may seem well. The agreement constitutes an agreement under international law, which has to be transposed into national law according to the rules and procedures of the specific national legal system. As legal conflicts arise, paradoxes appear. The agreement itself requires more of the national legal system than a “normal” international treaty. The EFTA Court poses challenges to the point of departure in international law. The fact that the main part of the agreement has been transposed into the law of the EFTA states as national legislation gives rise to the question of what actually has been transposed into national law. Finally, the ordinary legislative process within the Community is not the only challenge to the dynamics of the agreement. Such challenges are posed also by changes of the treaties that constitute the European Union, and therefor also the Community.

2. Provisions in the agreement beyond public international law

There is no duty in public international law to give treaties direct effect into the national legal system. Public international law is concerned with the state as a legal entity under international law, not with the different organs of the state under national law. The EEA agreement differs to a certain extent from this general starting point. Although the agreement does not reach as far into the internal workings of the national legal order as Community law, it has obligations and rules that have as their subjects different organs of the state.

Both article 7 and protocol 35 concern directly the national legislator. Article 7 states that an act corresponding to an EEC regulation shall as such be made part of the internal legal order of the Contracting Parties. An act corresponding to an EEC directive shall be, or be made, part of their internal legal orders, but shall leave to the authorities of the Contracting Parties the choice of form and method of implementation. Protocol 35 states that for cases of possible conflicts between implemented EEA rules and other statutory provisions, the EFTA States undertake to introduce, if necessary, a statutory provision to the effect that EEA rules prevail in these cases. The EEA agreement in other words obligates the EFTA states to implement all parts of the agreement, that is to give them effect in the national legal system, and to give these rules primacy over national legal rules. This may be characterised as “a qualified obligation under international law to secure the primacy of EEA provisions in analogy to Community law.” [8] Direct effect and primacy may not follow from the agreement itself, but the agreement dictates that it shall follow from implementing provisions of national law. The question of course is what happens when national law fails to perform according to these obligations. [9]

There are also provisions addressed directly at the national courts. The most obvious is article 6 which states that ”the provisions of this Agreement, in so far as they are identical in substance to corresponding rules of the Treaty establishing the European Economic Community and the Treaty establishing the European Coal and Steel Community and to acts adopted in application of these two Treaties, shall, in their implementation and application, be interpreted in conformity with the relevant rulings of the Court of Justice of the European Communities given prior to the date of signature of this Agreement.” Recital 15 of the preamble also addresses the Courts and states: ”in full deference to the independence of the courts, the objective of the Contracting Parties is to arrive at, and maintain, a uniform interpretation and application of this Agreement and those provisions of Community legislation which are substantially reproduced in this Agreement and to arrive at an equal treatment of individuals and economic operators as regards the four freedoms and the conditions of competition”. It can even be argued that article 3 and the obligation of loyalty to the agreement, also refers to the courts.

So much for the obligations of the agreement. Also on the level of rights entailed in the agreement, it can be seen that it reaches further into the realm of national law than traditional international treaties. Individual rights under international agreements are no longer uncommon, and the EEA agreement falls into this tradition. This can be seen already by the eighth recital of the preamble, which states ”the important role that individuals will play in the European Economic Area through the exercise of the rights conferred on them by this Agreement and through the judicial defence of these rights”. The rights hereby referred to, are of course the rights entailed in Community legislation that are ”substantially reproduced” in the agreement. The content and nature of these rights have lead commentators to characterise the basic rules of Community law as a Wirtschafftsverfassung. [10]

It is well known that rights for the individual to access national courts on the basis of Community rules form an integral and important part of Community law. The EEA agreement is ambiguous on this point. On the one hand the agreement presupposes national transformation for its rules to take effect if that is required by the national law of an EFTA state. On the other hand, its specific provisions substantially reproduce Community rules in such a way that they confer rights upon individuals. This follows from the rules themselves read in connection with article 6, which states that these rules be interpreted in conformity with the relevant rulings of the Court of Justice of the European Communities given prior to the date of signature of the agreement. The dividing line here is 2 May 1992.

Therefore, despite the fact that the agreement has been characterised as a treaty under public international law, it contains clear elements that go beyond most such treaties towards elements of supranationality. It may lack institutions with powers to legislate such as the European Parliament and the Council in the European Union. The EFTA Court does not have the same monopoly and status regarding the agreement as the ECJ has towards community law. The agreement on the other hand does contain elements that in the Community legal context have been important in the characterisation of Community law as a “a new legal order of international law for the benefit of which the states have limited their sovereign rights, albeit within limited fields, and the subjects of which comprise not only the member states but also their nationals”. [11]

3. Developments by the EFTA Court

Taking the text of the agreements as a starting point, one can see further developments in the direction of supranationality in the jurisprudence of the EFTA Court. In the Sveinbj�rnsdottir case, the Court stated that the agreement is "an international treaty sui generis which contains a distinct legal order of its own" that "is less far-reaching than under the EC Treaty, but the scope and the objective of the EEA Agreement goes beyond what is usual for an agreement under public international law". [12] The premises for this conclusion are the objectives of homogeneity and to ensure individuals and economic operators equal treatment and equal conditions of competition, as well as adequate means of enforcement.

In Sveinbj�rnsdottir, the Court found that the homogeneity objective and the objective of establishing the right of individuals and economic operators to equal treatment and equal opportunities are so strongly expressed in the EEA Agreement, that the EFTA States must be obliged to provide for compensation for loss and damage caused to an individual by incorrect implementation of a directive. The Court did not go so far as to state that such an obligation is directly effective in the national law of the Contracting Parties. On the contrary, the Court stated that the Agreement does not entail a transfer of legislative powers. From a strict technical point of view, it is a different matter whether such a rule has been implemented into national law along with the other parts of the agreement more expressly provided for in the wordings of its different articles.

In line with reasoning developed already in Restamark, the Court, however, pronounced a rule that may be characterised as quasi direct effect, when it stated that “the principle of State liability must be seen as an integral part of the EEA Agreement as such. Therefore, it is natural to interpret national legislation implementing the main part of the Agreement as also comprising the principle of State liability. [13] In Restamark, the Court, commenting upon the national implementation of protocol 35, stated: ”Protocol 35 EEA on the Implementation of EEA Rules stipulates that the EFTA States are under an obligation to ensure, if necessary by a separate statutory provision, that in cases of conflict between implemented EEA rules and other statutory provisions the implemented EEA rules prevail. It is inherent in the nature of such a provision that individuals and economic operators in cases of conflict between implemented EEA rules and national statutory provisions must be entitled to invoke and to claim at the national level any rights.” [14]

In my view, the Court here goes far into the tasks for the national courts namely the interpretation of national law implementing the EEA agreement. In Sveinbj�rnsdottir the Court directly refers to the interpretation of national law and expresses its opinion on how it is “naturally” interpreted. I Restamark, the Court when referring to “such a provision” that must be interpreted to the effect that implemented EEA rules have primacy over national rules, is referring to the national rule implementing protocol 35. The Court in this way blurs the boundaries between international law and national law thus giving the Agreement a certain effect in national law directly and independently of national law itself.

It may be argued that the Court does not go very far in this direction. It is, after all, tied to national rules that implement the agreement. On this basis one can say that there are no direct effects independent of national implementation if national law requires such. To this one may counter that the Court is referring to the basic implementing measures of the main parts of the Agreement. It is unthinkable for an EFTA state not having implemented the main parts. When the EFTA Court endeavours upon giving these measures an independent life of their own, this for all practical purposes amounts to influencing not only the interpretation of the Agreement itself, but also on the way the Agreement takes effect into national law.

It may also be true that the interpretations of national law, which the Court finds �inherent� or �natural�, are not very far-reaching or controversial. It is not very arguable that a rule implementing protocol 35 that in cases of conflict between implemented EEA rules and other statutory provisions the implemented EEA rules prevail, will entail a rule that implemented EEA rules have primacy over national rules. But does this also extent to national statutes that are of a more recent date than the implemented EEA rule? Protocol 35 clearly by its wording covers this situation, also in the case where it is clear that the national provision is intended as a deviation from the implemented EEA rule. But does protocol 35 require a national provision to the effect that EEA rules cannot be unimplemented? And if that is the case, is it “inherent” in a national provision implementing protocol 35? It is clearly arguable whether it is for the EFTA Court to pronounce upon what is “inherent” in provisions of national law. The interpretation of Sveinbj�rnsdottir is perhaps even more arguable. The Court here includes in the national implementation liability of the state, a rule that is not expressly included in the agreement. It is a form of “implementation without implementation”. For the time being, this is limited to state liability, since this principle “must be seen as an integral part of the EEA Agreement as such”. But what about other principles of Community law such as the duty of loyalty of the national courts, administrative principles of proportionality and protection of legitimate expectations, protection of human rights etc.? Are these also “integral parts” that must be seen as implemented as a consequence of the implementation of the Agreement itself? [15] It is tempting to ask how this differs from the principle of direct effect of treaty provisions as found in Community law. If the primacy entailed in the implementation of protocol 35 includes the wider scope of implementation suggested in Sveinbj�rnsdottir, then a hierarchy of norms is introduced into the EEA agreement. The main parts of the agreement, including judge made principles, have a status in national law that amounts to direct effect and primacy.

The jurisprudence of the EFTA Court in my view shows that the court is ready to approach the dilemmas inherent in the conflict between homogeneity and sovereignty by way of further extending the agreement beyond the scope of a normal international treaty.

4 Responses in national law

There are two sides of the coin regarding the effects of the EEA agreement in national law. I will approach this question from the perspective of dualist states, that is a state where national implementing measures are necessary for rules of international law to take effect in the national legal system. This is the case of Norway and Iceland, but not of Liechtenstein, or for that matter, the European Community. [16]

The starting point in a dualist state is that rules of international law have to be implemented by national legal measures. This does not totally exclude that such rules have any effect in national law. International law may be drawn upon when interpreting national law, and national law is often interpreted under a presumption that it conforms to the international obligations of the state. A legal conclusion cannot, however, be drawn exclusively from international sources, and international legal rules cannot have primacy over domestic rules without express provisions of national law.

From the point of view of Norwegian law, the core of the dualist principle has constitutional status. This follows from � 1 of the Constitution, which states the independence of the nation, and several articles that endow the legislative power to the Parliament. This entails that it is not possible to state that all international law shall be directly effective in Norwegian law without changing the Constitution. [17] It may also entail that adopting principles that amount to direct effect and primacy for non implemented rules of EEA law, may be barred by the constitution. This was the argument of the Norwegian State in a recent case for the Norwegian Supreme Court and the Court in a preliminary ruling on the right of the state to intervene in the case left the question open. [18] The case concerns the relationship between national provisions and a directive that is incorrectly implemented. The Court has not yet heard the case. It therefore remains to be seen whether the Court will pronounce anything to this question, and in that case, what will be pronounced.

The Icelandic Supreme Court stated in its ruling in Sveinbj�rnsdottir that it is natural that the act implementing the main part of the EEA agreement is interpreted in such a way that an individual has a claim to that Icelandic legislation is brought in harmony with EEA rules. Insofar this is not accomplished, it follows from the act and the fundamental principles of the EEA agreement that the state undertakes a liability under Icelandic law. [19] The court here clearly regards the implemented main part of the agreement as national Icelandic law, and in this includes the fundamental principles of the agreement. It also announces its will to abide by advisory opinions from the EFTA Court unless there are special circumstances that require exemption from it.

An interesting point in this judgement is that state liability is not based on an interpretation of internal legal sources. Liability follows from the specific EEA-implementing act and the agreement itself. It is equally clear that state liability is not part of the agreement in such a way that it can be said that the state openly undertook this obligation upon signing the agreement and implementing it into national law. The clear fact of the matter is that the EFTA Court, based on a construction advised the Icelandic Supreme Court to find the state liable, and the Supreme Court, unable to find special circumstances to justify a deviation from this advice, felt obliged to impose this EEA-specific liability. The legal mumble jumble differs, but the realities certainly resemble the Community situation where the right to compensation follows from community law, and that individuals are entitled to invoke this right before their national courts.

5 Beyond the Common Market

The EEA Agreement was negotiated and signed by the European Communities prior to the establishment of the European Union by the Maastricht treaty. The treaty of Amsterdam has further developed the treaties. Treaty reforms are an important part of developing the economic co-operation between the member states. Illustrating examples are the Single European act from 1986 facilitating the establishment of the single market and the Maastricht treaty revising the freedom of capital movements and establishing the monetary union. [20]

Treaty reforms not only add new aspects to the co-operation and concern institutional developments. Several of the reforms are also within the scope of the common rules of the EEA Agreement. The new rules on movement of capital introduced in the Maastricht treaty affect the Agreement directly, where the rules on capital are based on the old directive 88/361 on liberalisation of capital movements. [21] The economic and monetary policy, whereby Member States shall conduct their economic policies with a view to contributing to the achievement of the objectives of the Community, affect potentially interpretation and application of all four freedoms. The union citizenship may affect the free movement of persons. Several of the revisions from the Amsterdam treaty possibly affect the content and interpretation of the four freedoms and competition rules entailed in the EEA Agreement. This is notably the case for the new article 16 EC on services of general interest, the protocol on public broadcasting and for amendments in article 2, 3 and 6 giving potentially more weight on the cross-sectional obligation to protect the environment, to promote employment and to promote equality between men and women. [22]

Changes in the objectives and policies of the European Community Treaty may affect the homogeneity of the EEA. It is well established in case law that differences in objectives may lead to differences in the interpretation of provisions that are identical or similar in their wording. [23] To the extent that changes in the treaties of the EU challenge the homogeneity of the EEA, the question arises as to by which mechanisms homogeneity may be re-established, if at all.

The EEA-agreement does not seek to curb development of Community law. On the contrary, as set out in the introduction above, the Agreement contains several mechanisms to assure its dynamic development alongside the development of Community law. These mechanisms do not, however include revisions and amendments of the treaties establishing the Community and the European Union. One solution to this is to amend the EEA-agreement itself to reflect amendments in the treaties of the Union. This is of course an option. This option makes clear, however, that a dynamic homogeneity is not achieved by the EEA-agreement as it now stands. This conclusion rather admits that the EEA-agreement fails to meet its most central objective of establishing a dynamic and homogeneous European Economic Area, based on common rules and equal conditions of competition. Dramatic as this conclusion may seem, it is in no way surprising in light of the importance that treaty amendments have had and will have in the development of the European Union. If this is true, then the EEA is eroded a little bit every time treaty amendments of the EU take effect in influencing corresponding rules in the Community.

Another option is to let the interpretation of the EEA-agreement reflect the development in the interpretation of the corresponding Community provisions. To a certain point, at least the amendments of the EC treaty by Maastricht and to some extent Amsterdam were foreseen in the EEA agreement. The objectives of the EEA Agreement as stated in the preamble for this reason differ from the tasks and policies as stated in the EC treaty at the time of the signing of the agreement. Recital 10 states a high level of protection concerning health, safety and the environment as a basis for further development of the rules. Recital 11 states the importance of the development of the social dimension, including equal treatment of men and women, in the European Economic Area. In a joint declaration to the agreement, the EFTA states declared their commitment to the principles and basic rights ensured in the 1989 Community Charter of the Fundamental Social Rights of Workers. This charter has now been incorporated into art 136 of the EC treaty.

A closer scrutiny of the objectives and goals of the EEA Agreement may therefore facilitate a development in the interpretation of its provisions in line with case law from the ECJ based on the renewed treaty basis. It seems to me that the interpretation of article 81 in the Albany case could also be defended in the EEA context. Here the court found that agreements concluded in the context of collective negotiations between management and labour in pursuit of social policy objectives must, by virtue of their nature and purpose, be regarded as falling outside the scope of Article 81(1) of the Treaty. [24] The Court based its conclusion on Article 2 of the EC Treaty that after the amendment provides that a particular task of the Community is to promote throughout the Community a harmonious and balanced development of economic activities and a high level of employment and of social protection. Furthermore, the conclusion was based on the new article 136 and the agreement on social policy. Based on the preamble to the EEA Agreement and the joint declaration of the EFTA states, it could be held that this is in conformity with the objectives of the agreement.

There is no doubt that many of the developments in interpretation of community rules based on an enhanced weight on social policy, environment and equal rights for men and women could be accommodated into the interpretation of corresponding rules in the EEA Agreement. Such an approach does not, however, solve the question of to which extent this development should be reflected into the agreement. To hold that the EFTA states are under an obligation to let their obligations be influenced by revisions of the treaties between the member states of the European Union would certainly be to give a new meaning to the concept of supranationality. This would amount to a situation where the EFTA states were legally bound to developments caused by treaties concluded by foreign states. It is in no way clear what such an obligation should entail. Would it be up to the courts to decide, or would it be an obligation for the legislator of the EFTA states? New treaties and treaty revisions in the EU are not matters that may be brought before the EEA Joint Committee. There is therefore no cause for implementing the dispute settlement provisions of article 102.

There are other changes in the treaties that are not easily reflected in the Agreement by way of its own formulations of goals and objectives. There is for instance scant if any basis in the agreement to combat discrimination based on racial or ethnic origin, religion or belief, disability, age or sexual orientation. There can be no basis in the agreement to obligate the EFTA states to reflect these aims in the interpretation or application of its provisions.

The arguments above may seem to be of a more academic than practical nature. The will to find solutions is presumably high on all sides. The political development of the EU with higher emphasis on environment, social policy and services of general interest is easily supported by mainstream politics in the EFTA Countries. But this situation may alter. Political priorities may change both in EFTA states and in the EU. Instead of convergence one may have divergence. As soon as this happens, the challenges are much greater, and the distinction between law and politics will become clearer. A situation with diverging political priorities is of course a challenge also to a structure like the European Union. In the EU it means that the political co-operation looses pace. Within the EEA where the core of common rules is burrowed by differences in main objectives, more is at stake.

If the EFTA-states are under no obligation to let the changes in Community law be reflected in the agreement, clearly this must also apply to the other side, the Community and the member states of the EU. Little by little homogeneity may then disintegrate. The practical consequence of this may be for instance that measures restricting competition to protect for example the elderly from discrimination in an increasingly risk and cost based society, may be justifiable in member states but not an option under the EEA-agreement. Following the judgement in Polydor, this may open for an interpretation of the EEA agreement by the ECJ, which allows restrictions on the import of goods and services from EFTA states.

The EFTA states voluntary undertaking to follow the development of the EU could mitigate this risk. Voluntary adaptation to Community law has been the road taken by Switzerland in its relation to the Community. For many purposes this may be a viable option to a reciprocal agreement. [25] But the question may then arise of the purpose in having the EEA Agreement at all. The EEA Agreement may even to a certain extent hinder this option. Restrictions on competition and free movement based on later developments in the basis of the Union may technically be contrary to the EEA agreement as it stands. Probably the leeway to protect the environment and equal treatment of men and women is greater under Community law today than it was in 1992. Does that mean that it is also increased under the EEA Agreement? In other relations, member states` room to manoeuvre is decreased. Restrictions on capital movements and the use of safeguard clauses to protect economic policies are harder to justify after Maastricht than before. Clearly that cannot mean increased obligations on the EFTA states?

The EEA Agreement is an attempt to tie the EFTA-states to the internal market of the EU, without the EFTA states having to sacrifice their sovereignty and legal autonomy. As the analysis above has shown, one may question the success of the agreement on both these accounts.

It can be argued that the agreement in these respects fulfilled its goals at the outset. The agreement replicated Community law at the time by copying all relevant provisions in the treaties and secondary legislation, and by adopting the aqui en bloc through article 6. It was recognised that to take effect in the autonomous national legal systems, the agreement had to be adopted, and obligations were provided for this reason in article 7 and protocol 35. Once this was undertaken by the EFTA states, homogeneity was ensured by legal means. Homogeneity was not only a one-sided undertaking by the EFTA-states to adapt to the single market of the Community.

In the time that has passed, this picture has changed. The inherent difficulties in reconciling homogeneity with autonomy have become more acute. On the one hand, the development is towards less autonomy for the national legal systems. The EFTA Court seeks to influence national implementation by demanding compensation to individuals for failures in the national legal systems, and by developing what amounts to direct effect and supremacy for the main parts of the agreement, independent of the national implementation. So far this seems to have been accepted by the courts in Iceland and Liechtenstein. The question in Norway is still more open. [26]

On the other hand, the question of whether the EEA still is based on the same rules as apply within the single market may be raised. Community law develops in a way that is not legally reflected within the agreement. This being the case, homogeneity rests not upon obligation, but upon voluntary adaptation by the EFTA states to Community law. It may be argued that this is not so different from the case with developments in secondary legislation, which has to be agreed upon by the EEA Joint Committee to become part of the Agreement. In this way it is true that substantial parts of the dynamics of the agreement is always dependent upon political will. There is, however, a main difference. When applying the agreement itself, article 102 more specifically, the EFTA states have a legal claim to homogeneity, and once they invoke this, they become legally bound by it. When it comes to the dynamics of treaty revisions, there is no such legal claim, and the EFTA states are not legally bound by their voluntary adaptation. The alternative, that they were legally bound, would imply that they were bound by the way of dynamic interpretation of the agreement, and not by political decision. This would raise serious questions about sovereignty and legally autonomy, since it would entail that the EFTA states are bound by treaties concluded by a group of states in which they are not included.

So it may be concluded that despite the EEA Agreement becoming increasingly supranational, this does not achieve homogeneity. Instead of reducing political and economic risks by legal obligations, both the obligations and the risks are increased.


[1] For a comparison in scope and content between the EEA-agreement and Community law see Peter-Christian M�ller-Graff, EEA-agreement and EC Law: A Comparison in Scope and Content in M�ller-Graff and Selvig (eds.) The European Economic Area – Norway’s Basic Status in the Legal Construction of Europe, Berlin1997 (M�ller-Graff and Selvig 1997). See also Thomas Bruha, Is the EEA an Internal Market? in M�ller-Graff and Selvig (eds.) EEA-EU Relations, Berlin1999 (M�ller-Graff and Selvig 1999).

[2] For an overview see Fredrik Sejersted, Between Sovereignty and Supranationalism in the EEA Context in M�ller-Graff and Selvig 1997.

[3] See Sejersted in M�ller-Graff and Selvig 1997 and for more sepcific analysis of the relationship between the EFTA Court and the ECJ Hans Petter Graver, The Efta Court and the Court of Justice of the EC: Legal Homogeneity at Stake? And Carl Baudenbacher, The EFTA Court and the European Court both in M�ller-Graff and Selvig 1999.

[4] The one exception to this is in competion law, where ESA has many of the same compences towards operators inh EFTA countries as the Commision has within the EU, see Sejersted in M�ller-Graff and Selvig 1997.

[5] See H�kon B�rde, The European Economic Area, Norway and the European Union, in M�ller-Graff and Selvig 1997 and Henrik Bull, Monetary Union and the Capital Market: The Internal Market for Financial Services, Norwand the EEA, in M�ller-Graff and Selvig (eds.) on p. 181 with refernce to 100 1991-92.

[6] B�rde in M�ller-Graff and Selvig 1997.

[7] M�ller-Graff in M�ller-Graff and Selvig 1997.

[8] M�ller-Graff in M�ller-Graff and Selvig 1997.

[9] The simple answer is that this constitutes a breach of international obligations to be settled under the treaty it self and according to general principles of international law. The EFTA court has, however, stated that the state is liable to provide for compensation for loss and damage cause to individuals by breaches of the obligations under the EEA Agreement in case E-9/99 between Erla Mar�a Sveinbj�rnsd�ttir and The Government of Iceland. It has also been argued that the obligation to give primacy and direct effect does not exclude the possibility of direct effect and primacy without national implementation, see most recently Bruha in M�ller-Graff and Selvig 1999.

[10] Se for example Hans Peter Ipsen, Europ�isches Gemeinschaftsrecht, J.C.B. Mohr (Pauil Siebeck), T�bingen 1972 pp. 565-567.

[11] Case 26-62 van Gend & Loos v Netherlands Inland Revenue Administration 1962 ECR 26.

[12] Case E-9/97 Erla Mar�a Sveinbj�rnsd�ttir and The Government of Iceland paragraph 59.

[13] Case E-9/97 Erla Mar�a Sveinbj�rnsd�ttir and The Government of Iceland paragraph 63.

[14] Case E-1/94 Ravintoloitsijain Liiton Kustannus Oy Restamark paragraph 77.

[15] For a discussion on the possible inclusion of fundamental principles of Community law in the EEA agreement see Sejersted in M�ller-Graff and Selvig 1997 pp. 54-58.

[16] For this reason I do not agree with Thomas Bruha in M�ller-Graff and Selvig 1999 pp. 116-117 in that the ruling of the Court in First Instance in case t-115/94 Opel Austria Gmbh v. Council 1997 ECR II-39 has much relevance to the question of direct effect of the agreement. The Court here pronounces on a rule of community law on the effect of the international obligations of the Community in its own legal system.

[17] See Torkel Opsahl, Noen sider av problemet om overgang til et "monistisk" system i Norge NOU 1972:16 p. 99-112

[18] Ruling of the Supreme Court in plenary session 22 August 2000 in Ankesak nr. 55/1999.

[19] Ruling of the Supreme Court 16 December 1999 my translation from Danish.

[20] For an analysis of the dynamics of treaty reforms based on the case of the Amsterdam treaty, see Ulf Sverdruo, Precedents and Present Events in the European Union: An Institutional Perspective on Treaty Reform in Neunreither and Wiener (eds.) 2000.

[21] See Henrik Bull in M�ller-Graff and Selvig (eds.) 1999.

[22] See Peter-Christian M�ller-Graff, The Treaty of Amsterdam: Content and Implications for EEA-EU Relations, in M�ller-Graff and Selvig (eds.) 1999.

[23] See ECJ case 270/80 Polydor, 1982 ECR p. 329, opinion 1/91 on the EEA Agreement 1991 ECR p. I-6079 and EFTA Court case E-2/97 Maglite, 1997 REC p. 127.

[24] Case C-67/96 Albany International BV v Stichting Bedrijfspensioenfonds Textielindustrie 1999 nyr paragraph 60.

[25] See Stephan Kux and Ulf Sverdrup, Fuzzy Borders and Adaptive Outsiders: Norway, Switzerland and the EU, European Integration, 2000, Vol. 00, pp. 1-34 for a comparison of the two approaches.

[26] The issue is different in Liechtenstein due to its monist approach to international obligations. The Supreme Court of Liechtenstein has, however, expressely affirmed the direct effect of the Agreement, se to this Thomas Bruha in M�lle-Graff and Selvig (eds.) 1999.

[Date of publication in the ARENA Working Paper series: 15.10.2000]