Direct effect and indirect effect

Direct effect is the doctrine by which a provision of EU law may be invoked by an individual before a national court. Indirect effect is the related, but distinct, interpretative obligation requiring national courts to construe domestic law, so far as possible, consistently with EU law. Together they are the practical machinery through which the constitutional claims made in Week 2 on supremacy acquire forensic force. Supremacy says that EU law prevails in case of conflict. Direct effect and indirect effect tell the lawyer when, and by what route, an individual can put EU law to work in litigation.

For Durham Year 1, the point is not merely to memorise the formula from Van Gend en Loos. The real issue is remedial architecture. A claimant may rely directly on Treaty provisions, regulations, certain decisions, and—within limits—directives. Where direct reliance fails, the court may still be obliged to interpret domestic law compatibly with the directive. Where both direct and indirect effect fail, state liability may provide a further remedy. The doctrines are therefore best understood as a sequence of judicial strategies designed to preserve the effectiveness of EU law while formally respecting the limits of the Treaties.

The difficult cases are not the easy vertical ones. A clear Treaty provision prohibiting discrimination by the state is straightforward. The problems arise with directives, because Article 288 TFEU states that a directive is binding as to the result to be achieved but leaves national authorities the choice of form and methods. The Court of Justice nevertheless recognised vertical direct effect of directives once the implementation deadline has passed. It refused, however, to recognise horizontal direct effect of directives against private parties. This created the notorious triangle of substitute techniques: a broad conception of the state or emanations of the state; indirect effect; and state liability.

Post-Brexit, the doctrine must be handled carefully. The United Kingdom is no longer a Member State, and new EU law does not operate domestically by reason of the European Communities Act 1972. For historic and transitional purposes, however, direct effect remains essential to understanding how EU law operated in the UK before exit, how pre-exit rights were saved, and how the current category of assimilated law emerged after the Retained EU Law (Revocation and Reform) Act 2023. Durham assessments increasingly expect students to distinguish EU constitutional doctrine in its own legal order from the present domestic status of EU-derived norms. That distinction is examinable: do not write as if the ECA 1972 were still in force; but do not treat Brexit as erasing the constitutional logic of Van Gend, Marshall, Marleasing and Francovich.

A strong answer will therefore proceed in three stages: first, identify the EU norm and its formal source; secondly, ask whether it satisfies the conditions for direct effect and against whom it is invoked; thirdly, if direct effect is unavailable or incomplete, consider indirect effect and then state liability. That is the disciplined structure which avoids most first-year errors.

Direct effect was not expressly written into the EEC Treaty. Its creation was one of the foundational constitutional moves made by the Court of Justice in the early 1960s. The Court was confronted with a Community legal order which depended on national courts for enforcement, but whose Treaties would have been fragile if enforceability depended solely on inter-state proceedings or Commission enforcement. Van Gend en Loos transformed that position by treating Treaty obligations as capable of conferring rights on individuals enforceable in national courts. The doctrine converted the Treaty from an agreement between states into a source of law used by private litigants.

The historical significance of Van Gend lies in its connection with the preliminary reference procedure. National courts were made the ordinary courts of EU law. Individuals could litigate EU points domestically, national courts could refer questions of interpretation to Luxembourg, and the Court of Justice could articulate uniform constitutional principles. Direct effect therefore helped create what Joseph Weiler famously described as a system of legal integration: private enforcement became a central mechanism of supranational governance.

The sequence with supremacy matters. Week 2 dealt with Costa v ENEL and the claim that EU law could not be overridden by domestic legal measures. But supremacy without direct effect would often be inert. If a private trader, worker, consumer or taxpayer cannot invoke the EU norm, the conflict with domestic law may never be judicially addressed. Conversely, direct effect without supremacy would allow national courts to recognise EU rights only until a contrary national measure appeared. The doctrines therefore developed symbiotically: direct effect opens the courthouse door; supremacy determines which rule wins once the claimant is inside.

The Court’s early direct effect cases concerned Treaty provisions. The doctrine then extended to regulations, certain decisions, and directives. Regulations presented little conceptual difficulty because Article 288 TFEU says that a regulation is directly applicable. Direct applicability and direct effect are not identical, but regulations are designed to enter national legal systems without implementing legislation. Directives were harder. They are not directly applicable in the same way, and they are addressed to Member States. Yet the Court reasoned that a Member State should not benefit from its failure to implement a directive. Thus, where the implementation period had expired and the relevant provision was clear, precise and unconditional, an individual could invoke the directive against the state.

The refusal to permit horizontal direct effect of directives marks a constitutional compromise. In Marshall, and later Faccini Dori, the Court held that a directive could not of itself impose obligations on an individual. This respected the wording of Article 288, but it created practical inequality: a claimant’s remedy could depend on whether the defendant was a public authority or a private employer. The Court responded not by openly reversing Marshall, but by expanding adjacent doctrines. It defined the state broadly in Foster; required national courts to interpret domestic legislation consistently with directives in Von Colson and Marleasing; recognised incidental effects in some contexts; and developed state liability in Francovich.

In the United Kingdom, these doctrines operated through the European Communities Act 1972, especially section 2(1). That provision was the domestic conduit through which directly effective EU rights were recognised and enforced. It was also the statutory basis for the reasoning in Factortame, although Factortame is primarily a supremacy case. After Brexit, the European Union (Withdrawal) Act 2018 preserved certain EU-derived rules as retained EU law, later renamed assimilated law, but the Retained EU Law (Revocation and Reform) Act 2023 significantly altered the landscape, including by repealing the saving of directly effective rights under section 4 of the 2018 Act at the end of 2023. The historical doctrine remains indispensable, but its present domestic application now depends on the post-Brexit statutory settlement.

The starting point is the source of EU law. Direct effect is not a free-standing label attached to EU law generally. One must ask whether the particular provision invoked is capable of being enforced by an individual in national proceedings. The classic formula requires the provision to be sufficiently clear, precise and unconditional. The content of the obligation must be judicially manageable: a national court must be able to identify what is required without waiting for further legislative or administrative discretion.

Treaty provisions. Treaty articles may have direct effect where they satisfy the Van Gend criteria. Some Treaty provisions are vertically directly effective, allowing reliance against the state. Others may also be horizontally directly effective, allowing reliance in litigation between private parties. Defrenne v SABENA is the orthodox example: the equal pay provision, now Article 157 TFEU, could be invoked by a worker against a private employer. Not every Treaty article will qualify. Broad programmatic provisions, provisions dependent on further political action, or provisions leaving substantial discretion will not usually be directly effective.

Regulations. Regulations are binding in their entirety and directly applicable in all Member States. Direct applicability means that no national implementing measure is required for the regulation to become part of domestic law. Direct effect asks the further forensic question whether a particular provision of the regulation is enforceable by an individual. In practice, many regulatory provisions are directly effective where they are sufficiently clear and unconditional. A student should avoid saying simply that all regulations are directly effective. The better formulation is that regulations are directly applicable, and provisions within them commonly have direct effect if they meet the ordinary criteria.

Directives. Directives are binding as to the result to be achieved, while leaving national authorities the choice of form and methods. The Court of Justice nonetheless permits vertical direct effect of directives once the implementation deadline has expired, provided the provision is clear, precise and unconditional. The rationale is partly estoppel-like: a Member State should not rely on its own failure to implement EU law. The limitation is equally important. Directives do not have horizontal direct effect. A directive cannot of itself impose obligations on a private individual. In problem questions, this is the central hinge: identify whether the defendant is the state, an emanation of the state, or a private party.

The leading cases should be learned as a sequence rather than as isolated authorities. Van Gend en Loos is the origin of direct effect. A Dutch importer challenged a customs duty by relying on what is now Article 30 TFEU. The Court of Justice held that the Treaty created a new legal order whose subjects included individuals. The operative importance of the case is the test: a provision may be invoked before national courts where it is clear, precise and unconditional. Van Gend is therefore both a constitutional case and an access-to-remedy case.

Defrenne v SABENA shows that some Treaty provisions have horizontal direct effect. The claimant relied on the equal pay provision against a private airline employer. The Court held that the principle of equal pay could be invoked directly, including in proceedings between private parties. This matters because horizontal effect is not categorically unavailable for all EU norms. It is directives, not Treaty provisions as such, that encounter the Marshall prohibition.

Van Duyn and Ratti are the essential directive cases. Van Duyn accepted that a directive could have vertical direct effect, despite Article 288’s language, because the binding effect of directives would be weakened if individuals could never rely on them. Ratti added the implementation-deadline logic: after the deadline, a defaulting Member State cannot rely on its own failure to implement. The vertical character of the doctrine follows from that rationale.

Marshall is the doctrinal barrier. Mrs Marshall sought to rely on the Equal Treatment Directive against her employer, a health authority. She succeeded because the employer was a public authority, but the Court stated that a directive cannot of itself impose obligations on an individual. That proposition was reaffirmed in Faccini Dori, where the defendant was a private company and the claimant could not rely directly on the directive against it. Faccini Dori is especially useful because the Court openly acknowledged the gap while pointing to indirect effect and state liability as alternative techniques.

Foster addresses the public-private boundary. The claimant relied on a directive against British Gas. The Court provided criteria for identifying an emanation of the state. The importance of Foster is practical: many bodies that look formally separate from government may still be treated as the state for vertical direct effect. This mitigates, but does not abolish, the horizontal-direct-effect problem.

Von Colson and Marleasing establish indirect effect. Von Colson required national courts to interpret domestic implementing legislation in light of the directive’s purpose. Marleasing extended the obligation to national law generally, whether enacted before or after the directive. The significance of Marleasing is that it allows directives to influence private litigation without formally granting them horizontal direct effect. The court is applying national law, but reading it through an EU-consistent lens.

Francovich, although not always listed under direct effect, is indispensable to the remedial structure. Where a directive has not been implemented and direct effect is unavailable, individuals may recover damages from the state if the conditions for liability are met. In a problem question, Francovich is often the final step after Marshall and Marleasing have been considered.

See practice questions section below.

See further reading section below.