top of page


Updated: May 8

The establishment of direct effect and the supremacy of EU law was a mistake. The various rationales employed by the Court of Justice to justify their development have been seriously flawed. The principles governing the supremacy of EU law are far too strict while the principles governing direct effect are an incoherent mess. Yet these problems can be resolved easily.” Critically discuss this statement with reference to relevant treaty articles, decided cases and academic opinion.


Part 1 will discuss Supremacy. It will examine the rationale of the doctrine. It will further examine if the doctrine of supremacy is too strict. It will then examine if the doctrine can be reconciled? Part 2 will be an examination of the “direct effect doctrine” which will determine how it has been treated by the Court of Justice of the European Union (“CJEU”). The analysis of the doctrine will be followed by a study of restrictive limitations, the vertical and horizontal direct effect arbitrary distinction (“VDE” and “HDE”) and the reasoning behind the Marigold case will also be looked into.


The supremacy of the European Union (“EU”) and its law, which is often referred to as primacy of EU is the principle that EU law in its entirety supersedes the domestic laws of the Member States (“MSs”). Thus EU law prevails over domestic law. This means that the individuals of the member states can rely and cite EU law in their national courts, i.e. EU law has direct effect in national courts.

Doctrine of Supremacy

The European Economic Treaty did not have any provision on supremacy of EC law. However, a primacy section was included in the Constitutional Treaty, Article 1-6, and consequently a declaration on primacy was included in the Lisbon Treaty (Declaration 17).But prior to the introduction of primacy the ECJ had pronounced its visualization of supremacy in their earlier jurisprudence. In Van Genden Loos the ECJ as they were then held:

“the Community constitutes a new legal order of International law for the benefit of which the States have limited their sovereign rights, albeit in limited fields, and the subjects of which comprise not only member states but also their nationals”.

Moreover, in Costa v ENEL the ECJ as they were then held:

“the Member States have limited their sovereign rights, albeit within limited fields, and have thus created a body of law which binds both their nationals and themselves...Such a measure cannot therefore be inconsistent with that legal system… the law stemming from the treaty, an independent source of law, could not because of its special and original nature, be overridden by domestic legal provisions, however framed, without being deprived of its character as Community law and without the legal basis of the Community itself being called into question.”

The point to note is that supremacy of the EU law was created by ECJ and is not something which is in Treaties. ECJ talked about “spirit” of Treaty, this was not the drafters’ intentions. Commentators argue that EU law should be accorded superiority because it stems from the treaty that has been made by the MSs when they joined the EU and formed a new legal order. Yet, no reference to the constitution of MSs was made. This perhaps should have been done. Functional commentators argue the aims of the Treaty would not be achievable unless EU law was accorded supremacy. The security of uniform application of EU law and it effectiveness is the only way the EU can achieve its purpose. Dougan has argued that the doctrine of primacy produces exclusionary effects within national legal systems, in that it sets aside domestic laws that are inconstant with a “hierarchically superior norm of Community law”. This is distinguished from the substitution effects where the direct application of Community law is concerned.

Is the doctrine of supremacy too strict

An instance of where the doctrine has been seen as too strict is where the French disagree with EU law being supreme over all national law. France believes that “EU law ranks above statute, but below their constitution”. Additionally, Germany claims that EU law is not absolutely applicable. In the case of Internationale Handelsgesellschaft, the impact of EU law on the German Constitution was considered. The highest German Court ignored the supremacy of EU law on the grounds that it would detrimentally impact the fundamental rights in the German constitution. Of course, the ECJ disagreed. German Courts claim that they will review EU laws for “compliance with fundamental rights” until they are satisfied that EU law gives “equivalent protection”. In Solange II , following developments in EU fundamental rights jurisprudence the German Court now accepts that equivalent protection is provided and that it will no longer review EU laws for “compliance with fundamental rights”. Frowein argues that the courts did not “surrender their jurisdiction over fundamental rights”, but simply claimed that they would not exercise that jurisdiction while the ECJ continues the present circumstances. In Honeywell,the German courts claimedthat the EU law was being used beyond its competence and hence was ultra vires.

In Gauweiler v Lisbon Treaty the German court has authority to engage in ultra vires review. The Union is not a democracy which means that integration cannot go so far as to affect the ability of the democratic state to put itself together (i.e. “the state must take a role in criminal law, war and peace, public expenditures and taxation, welfare, and culture and religion”). The EU cannot take on powers such as military, revenue, key social powers. Moreover, under extraordinary conditions the “Federal Constitutional Court may declare EU law inapplicable in Germany in violation of Germany’s constitutional identity as a democratic federal state”, this is referred to as the identity lock.

Can the doctrine be reconciled?

The UK accepts supremacy not on the basis of authority of EU law, but the intention of Parliament in passing the ECA 1972, which is a “constitutional statute”. The Conceptual basis for acceptance of supremacy was seen in the case of Factortame II. Lord Bridge accepted the principle of supremacy relied upon by the ECJ and connected the acceptance of the principle by the UK to the UK statute 1972 European Communities Act. Moreover in the case of Thoburn v Sunderland CC, acceptance of supremacy by domestic courts was considered. Laws LJ said that “common law had modified the traditional concept of parliamentary sovereignty”, which created exceptions to the “doctrine of implied repeal”, also known as “constitutional statutes” (of which ECA 1972 was one) which could only be repealed by express wording.

MacCormick argues that there is a constant interaction between the two legal systems- the legal systems of the member states on one hand and the EU law legal system on the other. The national courts have no authority to assume dominance over the international order and in case there is a conflict between the two legal systems, there is always the alternative possibility of recourse to adjudication or international arbitration.

Kirchhof argues that if EU law is made autonomous it will lose its authority and power to develop since the ECJ and domestic courts both have adjudicatory functions and responsibilities of their own, it must be a question of balance and collaboration not subordination, domination or supremacy.

Eeckhout argues that pluralism in this case is based on the highest courts, namely the CJEU on one hand and the national high courts on the other, making contradictory claims about the final authority. The Court of Justice has to identify that its authority is also bound by the questions of EU law. It does not have ultimate authority over national courts’ acceptance of EU law. Moreover, the jurisdiction of national courts must also be respected, which means that there must be no unnecessary interference in the questions of national law.

Direct Effect EU Law

In order to make EU law effectual and reliable it must be given precedence over the laws of the member states. The case of Van Genden Loos laid down the “doctrine of supremacy”. In this case, a customs duty imposed under a Dutch law was challenged on the grounds that it did not comply with the EC treaty Article 12. The ECJ (as they were previously known) held that the EU law superseded the Dutch law and the Dutch had limited their sovereign powers. Therefore, the “doctrine of direct effect” is an EU provision that can be invoked by any individual of any member state which will have a binding effect. Most of the EU treaties , regulations and decisions can be relied upon.

However, the EU directives are not applicable directly. Since, a directive does not immediately become ‘law’, it highlights the objectives which the member states need to accomplish, which can be done in any way the member states deem fit. These directives may be applicable to some or all member states.

The aforementioned case of Van Gend laid out few conditions that have to be met before an EU law can have “direct effect”. The Reyners test states that the provisions of the law in question must be clear, precise and absolute. For directives, the doctrine is based on the principle of estoppel: they are capable of “direct effect” to increase their efficacy and make certain that Member States do not take advantage of their own inactions . However, “direct effect” is only possible after the “deadline for implementation of the directive expires.”

The CJEU has placed another restriction on the “direct effect” of directives: the directives can only have VDE. Invoking Article 288 paragraph 3 of the Treaty on the Functioning of the European Union (“TFEU”), the CJEU stated that the directives were incapable of HDE. Unlike a regulation which becomes a part of the national law instantaneously, a directive requires a nationally implemented legislation to become operational. The reasoning behind distinguishing between HDE and VDE is that Article 288 does not make directives binding on private individuals. In another case of Dori , it was held that the regulation imposes obligations of individuals which a directive does not do.

This arbitrary difference has led to many problems, for instance the cases of Duke and Marshall. Both the cases dealt with a dispute over retirement age but the outcomes were different as one was considered to be a function of the state and VDE was made applicable, where the other was not. Clearly there are two rules, one for the emulation of the state and the other for non- state bodies, with community rules being ignored. Moreover, political agendas have limited the scope of directives which makes bad law. Therefore, the CJEU has diluted the “direct effect doctrine” through the use of “arbitrary distinctions” and “faulty reasoning.”

The CJEU has supported this claim that as stated in Marshall and Dori , a directive cannot be relied upon by private individuals in national courts. In these cases, the CJEU utilized a treaty provision which would be directly effective , or ignored the possibility of HDE by claiming that the action was not covered by the directive , hence clearly discounting the “direct effect doctrine”. The Court has extended the doctrine of VDE for directives in cases like Unilever and CIA Security , using terms such as “exclusionary effect”. These aforesaid cases point out that these directives can be enforced between individuals horizontally.

The CJEU tried to expand the “direct effect doctrine” to overcome the absence of HDE, this extended their inconsistent principles further. Therefore, in the case of Foster the Court examined the same issue as in the Marshall case . In Foster the definition of “state organ” was stretched out to include a “nationalised utilities company”, which was another “arbitrary distinction”.

The ECJ held in the case of Van Colsson that in order to meet the treaty requirements, the aim of the directives must be kept in mind by the domestic courts while interpreting domestic law. This obligation is called the “indirect effect”.

In the case of Marleasing which dealt with private individuals, the Court claimed that the domestic courts must keep the wording and aim of the directives in mind while interpreting the directives, thus confirming that directives have an impact even on cases involving private individuals. In its quest to bar HDE and expand VDE for directives, the CJEU has made the difference almost invisible. Hence, this has led to more confusion.

The ECJ in the case of CIA Security which involved private individuals gave direct effect to a directive which had not been implemented, pointing to the possibility of “horizontal direct effect” in case of directives as well. When this discrepancy was pointed out by Advocate- General Jacobs , the ECJ held that the emphasis was on the objectives of the directive which does not create any rights or obligations on an individual but involves the failure of a state to apply it. This decision is clearly arbitrary.

The CJEU in another case of Mangold, extended the HDE doctrine to a directive involving equal protection and making the domestic courts responsible for ensuring the effects of such rights, even though the deadline for implementing the directive had not passed yet. After the Lisbon Treaty (2009) which gave more rights to the CJEU, the Court further emphasised the decision in Mangold by reinforcing the principle in Kücükdeveci. It was claimed that the directive clearly “gave expression” to the principles of EU law. Therefore, it was within the domestic courts’ jurisdiction to provide legal protection under EU law to the private individuals seeking relief. This time marked the use of both HDE and VDE for directives. . But this raises a lot of questions due to its inconsistency with Dori . With such clear discrepancies between similar cases, the inconsistent and arbitrary principles employed by CJEU need an explanation.

The unnecessary use of restrictions on direct effect of directives and arbitrary distinction between HDE and VDE has seriously diluted the doctrine. Although the doctrine has been extended horizontally by cases such as Mangold , the CJEU has continually discounted the doctrine which has led to a lot of disparity between similar cases. Even though the extension of VDE and HDE are steps in the right direction, it shows that the CJEU has always engaged avoidable limits and then tried to spread out the workings of the doctrine.

Direct effect EU law


Recent Posts

See All
bottom of page