Jump to Content Jump to Main Navigation
Max Planck Encyclopedia of Public International Law [MPEPIL]

Portugal v Council Case

Piet Eeckhout

From: Oxford Public International Law (http://opil.ouplaw.com). (c) Oxford University Press, 2015. All Rights Reserved.date: 20 May 2019

Subject(s):
Tariffs — Services — Intellectual property — Specific trade agreements

Published under the auspices of the Max Planck Foundation for International Peace and the Rule of Law under the direction of Rüdiger Wolfrum.

A.  Introduction and Antecedents

This judgment is the European Court of Justice (‘ECJ’)’s most fundamental ruling on the status of the law of the World Trade Organization (WTO) in European Community law (European Union, Court of Justice and General Court European Community and Union Law and International Law). The European (Economic) Community is a founding member of the WTO, together with the European Union Member States (European Community and Union, Mixed Agreements). The ECJ recognizes that international agreements concluded by the EC form an integral part of EC law (Case 181/73 Haegeman v Belgium [1974] ECR 449, paras 2–6). That does not however mean that the provisions of such agreements are always justiciable. There are several factors which determine whether that is the case, such as the type of proceedings in which reference is made to an agreement and the type of claim which is made. The concept most commonly employed to describe whether the provisions of an agreement can be invoked in judicial proceedings is that of direct effect (Treaties, Direct Applicability). The ECJ is favourably disposed towards recognizing that an agreement has direct effect, but nevertheless considers that the nature and structure of an agreement may preclude such effect. To date, the ECJ has denied direct effect only to the General Agreement on Tariffs and Trade (1947 and 1994) and WTO agreements.

In International Fruit Company, the ECJ held that the General Agreement on Tariffs and Trade (‘GATT’) of 1947 did not confer rights on individuals (Joined Cases 21–24/72 [1972] ECR 1219). This lack of direct effect of GATT was based on the agreement’s flexibility: its scope for negotiations, derogations, and safeguard measures. In later years, the ECJ developed certain exceptions to this principle. In Fediol, the ECJ reviewed a European Commission decision in the light of GATT because the decision was based on a regulation which implicitly referred to GATT (Case C-70/87 Fediol v Commission [1989] ECR 1781). In Nakajima, the ECJ reviewed an anti-dumping regulation on the basis of the GATT Anti-Dumping Code because the EC’s basic anti-dumping regulation was adopted to comply with GATT obligations (Case C-69/89 Nakajima v Council [1991] ECR I-2069). In Germany v Council, on the legality of the 1993 banana regime, the ECJ considered that the above features of GATT also precluded it from taking provisions of GATT into consideration to assess the lawfulness of a regulation in an action for annulment brought by a Member State. The special features of GATT showed that the GATT rules were not unconditional and that an obligation to recognize them as rules of international law which were directly applicable in the domestic legal systems of the contracting parties could not be based on the spirit, general scheme or terms of GATT. In the absence of such an obligation following from GATT itself, it was only if the Community intended to implement a particular obligation entered into within the framework of GATT, or if the Community act expressly referred to specific provisions of GATT, that the ECJ could review the lawfulness of the Community act in question from the point of view of the GATT rules (Case C-280/93 Germany v Council [1994] ECR I-4973, paras 103–12). The judgment was especially noteworthy for extending the direct effect requirement—even if the ECJ did not refer to the concept of direct effect—to an action brought by a Member State.

Shortly after the judgment in the bananas case, the WTO Agreement entered into force (1 January 1995). The establishment of the WTO, as a successor to, but at the same time a significant expansion and upgrade of, the GATT of 1947, clearly injected a much stronger legal dose into the multilateral trade agreements. Not only was the scope of such agreements expanded to include trade in services and the protection of intellectual property, the latter, of course, an inherently rights-oriented subject (Agreement on Trade-Related Aspects of Intellectual Property Rights [1994]; General Agreement on Trade in Services [1994]), but the GATT itself, some of whose provisions are indeed general and vague, employing complex and economically-oriented concepts, was given much more precision through a number of further agreements or ‘understandings on interpretation’. Two of those agreements subjected the important international trade sectors of agriculture, and textiles and clothing, to a process of liberalization which stands in stark contrast to the preceding lack of real disciplines (agriculture) or even organized protectionism (textiles and clothing). The scope for exceptions, derogations, and safeguard measures under GATT was reduced by other agreements and understandings (Safeguards). Most importantly, however, the WTO’s institutional basis is much stronger than that of the GATT of 1947. That institutional basis includes an advanced, sophisticated, and in many respects judicialized system of dispute settlement (World Trade Organization, Dispute Settlement).

The establishment of the WTO triggered a great debate in EC law circles on the direct effect of WTO law, at the political, judicial, and academic levels. The importance of the question could hardly be overestimated. In light of the broad scope of WTO law, covering trade in goods (Goods, Free Circulation of), trade in services (Services, Trade in), and intellectual property law (Industrial Property, International Protection; Intellectual Property, International Protection), and in light of its many connections with EC law, the recognition of the direct effect of WTO law provisions would have opened up large avenues for claims based on WTO law, claims which would no doubt often be directed at invalidating or setting aside either EC or national legislation. At the political level, the Council and the Commission clarified from the outset that they were opposed to the recognition of direct effect by inserting a provision in the preamble of the Council Decision concluding the WTO Agreement stating that the Agreement, by its nature, was not susceptible to being directly invoked in Community or Member State courts (Council Decision 94/800 [1994] OJ L 336/1).

B.  The Judgment

In Portugal v Council the Court grasped the nettle. The case concerned the legality of a Council decision concluding agreements on trade in textile products with Pakistan and India. Portugal claimed that the decision constituted a breach of certain rules and fundamental principles of the WTO (World Trade, Principles), in particular those of GATT 1994, the Agreement on Textiles and Clothing, and the Agreement on Import Licensing Procedures.

The Court effectively maintained its existing case-law, excluding direct effect of WTO law, but based it on a different reasoning. The judicial analysis consists of two parts. In the first part (paras 34–41), the Court started by referring to its earlier Kupferberg judgment (Case 104/81 Hauptzollamt Mainz v Kupferberg [1982] ECR 3641): contracting parties were free to agree on what effect the provisions of an agreement were to have, but if the question was not settled by the agreement it was for the Court to decide, as any question of interpretation; also, according to the general rules of international law there had to be good faith (bona fide) performance of every agreement. And although each Contracting Party was responsible for executing fully the commitments which it had undertaken, it was nevertheless free to determine the legal means appropriate for attaining that end in its legal system, unless the agreement, interpreted in the light of its subject-matter and purpose, itself specified those means. On that basis, the Court examined the WTO Agreement. It acknowledged the strengthening of the system of safeguards and the mechanism for resolving disputes, but nevertheless considered that the resulting system accorded considerable importance to negotiation between the parties. To demonstrate that, the Court turned to the dispute settlement system, and in particular to the scope for compensation under the provisions of Art. 22 (1) and (2) of the WTO’s Dispute Settlement Understanding (‘DSU’). It conceded that Art. 22 (1) showed a preference for full implementation of a recommendation to bring a measure into conformity with the WTO agreements in question, but pointed out that under Art. 22 (2), in the event of failure to implement, a member was required to enter into negotiations with a view to finding mutually acceptable compensation. Consequently, the Court continued, to require the judicial organs to refrain from applying the rules of domestic law which were inconsistent with the WTO agreements would have the consequence of depriving the legislative or executive organs of the contracting parties of the possibility afforded by this provision to enter into negotiated arrangements even on a temporary basis. It followed that the WTO agreements, interpreted in the light of their subject-matter and purpose, did not determine the appropriate legal means of ensuring that they were applied in good faith in the legal order of the contracting parties.

The Court added a second part (paras 42–49), which can only be understood against the background of its previous case-law on direct effect. Indeed, the Court readily recognized the direct effect of other international agreements, even though most of them do not ‘determine the appropriate legal means of ensuring that they are applied in good faith etc.’ either. So there needed to be further reasons for denying WTO law direct effect.

The Court opened the second part by indicating that this part concerned, ‘more particularly’, the application of the WTO agreements in the EC legal order. It noted that the WTO was still founded, like GATT 1947, on the principle of negotiations with a view to ‘entering into reciprocal and mutually advantageous arrangements’ (see the preamble), and was thus distinguished, from the viewpoint of the EC, from agreements concluded between the EC and non-member countries which introduced a certain asymmetry of obligations, or created special relations of integration with the EC, such as the agreement which the Court was required to interpret in Kupferberg. The Court added that it was common ground that some of the contracting parties, which were among the most important commercial partners of the EC, had concluded from the subject-matter and purpose of the WTO agreements that these were not among the rules applicable by their judicial organs when reviewing the legality of their rules of domestic law. The Court then acknowledged its statements on reciprocity in Kupferberg, but considered that in the case of the WTO agreements, based as they were on reciprocity, lack of reciprocity as regards judicial application could lead to disuniform application of the WTO rules. It added that, to accept that the role of ensuring that EC law complied with those rules devolved directly on the EC judicature would deprive the legislative or executive organs of the EC of the scope for manoeuvre enjoyed by their counterparts in the EC’s trading partners. The Court concluded that it followed from all those considerations that, having regard to their nature and structure, the WTO agreements were not in principle among the rules in the light of which the Court was to review the legality of measures adopted by the EC institutions.

C.  Commentary

As mentioned, the reasoning of the Court in Portugal v Council is divided into two parts. Part one of the judgment examines whether the WTO Agreement itself provides, expressly or implicitly, that there is to be direct legal effect. The Court’s reasoning has attracted strong academic criticism. It has been suggested that the Court misinterpreted the DSU provisions, in that there is in fact no option for non-compliance under WTO law. Retaliation and compensation are mere temporary alternatives to full compliance, and the latter is absolutely mandated by the DSU. But that critique does not capture the essence of the Court’s reasoning. The Court does not examine whether the WTO Agreement contains a commitment of full compliance. What the Court examines is whether the WTO Agreement mandates direct effect. Those are two distinct and different questions of law. There is much to be said for the view that the DSU, when properly interpreted, requires full compliance, yet that is not to be equated with direct effect. International agreements generally require full compliance, but international law does not generally mandate direct effect.

10  In the second part of the Court’s reasoning, the core meaning of Portugal v Council is revealed. The Court refers to reciprocity in the relations with other trading partners, but ultimately it is not reciprocity as such which leads the Court to deny WTO law direct effect. Rather, it is the impact of direct effect on the EU’s political institutions. If direct effect were granted, those institutions would lose the scope for manoeuvre which they currently have as regards implementation of WTO law, particularly in case of disputes with other WTO members. The hands of those institutions would be much more tied than the hands of their US, Japanese, and other counterparts. Ultimately, the Court is unwilling to take the step of tying the hands of the EU’s legislative and executive organs. This is the clear constitutional dimension of Portugal v Council, and it goes beyond the issue of reciprocity in international trade relations. The Court defers to the EC legislature, in terms of respecting both the statement in the preamble and any specific policies, now or in the future, which may cause WTO friction.

11  In its later case-law, the Court confirmed its reluctance to allow litigants, be they private parties or EU Member States, to invoke WTO law. It has inter alia decided that the logic of Portugal v Council prevails even in cases where, in the WTO, the EC has been found to have breached its WTO obligations, and where it has exhausted the reasonable period for implementation of the relevant WTO ruling (Case C-377/02 Van Parys [2005] ECR I-1465). In other words, the lack of direct effect extends to WTO dispute rulings. On the other hand, the Court does maintain the Fediol and Nakajima exceptions, referred to above, but construes them narrowly. It is also worth adding that the lack of direct effect of WTO law is limited to the provisions of the WTO Agreement which come within EC competence. In Dior, a case concerning the TRIPs Agreement, the Court stated that, with respect to WTO law provisions which do not come within EC competence, EC law neither requires nor forbids that the legal order of a Member State should accord to individuals the right to rely directly on the provision in question or that it should oblige the courts to apply that provision of their own motion (ECJ, Joined Cases C-300/98 and C-392/98 Parfums Christian Dior v Tuk Consultancy [2000] ECR I-11307, paras 41–45).

Select Bibliography