No secular organization has ever peacefully deprived states of as much sovereignty as has the European Union. National autonomy to regulate the environment, labor, the professions, antitrust, consumer protection, food and product safety, agriculture, advertising and almost any other area one can think of, even highly sensitive ones such as criminal law and civil procedure, has been gradually constrained over the years by rules coming from the EU.
Often the source of these rules is EU legislation, usually in the form of directives, which are laws that contain instructions to the member states to take certain action or implement certain rules. Such directives are often controversial because they are adopted by majority voting in the European Parliament and the European Council, meaning that states may well have measures imposed on them that they did not support. The directives also often lack public support because they originate in relatively distant EU institutions and come into being in a way that does not involve national parliaments, which are still the legislators with the most public acceptance and respect.
Yet despite all this, there is usually a grudging acceptance of EU legislation, which does at least emerge from a process in which elected parliamentarians, in the European Parliament, and national ministers, in the European Council, play a key role. By contrast, the decisions of the European Court of Justice are sometimes met with a degree of shock and disbelief. In the past five decades the European Court of Justice has been the most powerful and influential supreme court on the planet, playing a central role in creating the EU legal system and giving it teeth, and it has done this on the basis of a rather freewheeling and goal-oriented judicial style that is quite alien to most European legal systems.
The primary role of the court is to interpret EU law. This includes the two treaties that are the legal basis of the EU—the Treaty on European Union and the Treaty on the Functioning of the European Union—and all the secondary legislation adopted on the basis of those treaties. The court provides its interpretations at the request of national courts, in a process called the preliminary reference procedure. Any national judge who is hearing a case in which EU law may be relevant is entitled to ask the court a question about how that EU law should be interpreted. The court will answer, and the national judge will then apply the answer to rule on the case.
This sounds at first as if the court was expected to have a relatively marginal role in European integration. Following continental European legal traditions, the intention would seem to be that written texts would provide the basis for EU integration, and the court would simply help clear up ambiguities and ensure that coherence did not get lost in translation into the many official EU languages. The court has understood its role quite differently, however, and it has become in substance one of the primary legislators of the EU, laying down far-reaching rules and principles without which the EU would be quite a different creature today.
The story conventionally begins with two early cases, van Gend en Loos and Costa v. ENEL. In these references in the 1960s, from a Dutch and an Italian court respectively, the court laid down two fundamental, game-changing doctrines of EU law: direct effect and primacy. The first states that in most cases—there are some relatively technical exclusions—EU law may and must be directly applied by all national legal authorities, whether courts or the government or the administration. In other words, the doctrine holds that EU law is not just some distant international agreement, but a part of the national legal system that should be integrated into the daily legal life of national judges and the organs of the state.
The second doctrine, primacy, holds that where there is a conflict between national and EU law, EU law should always be given precedence and the national law set aside. The court has even been so bold as to claim that EU law should take precedence over national constitutions, although several member state supreme courts have not accepted this. The German Constitutional Court, the Bundesverfassungsgericht (BvG), has been the European Court’s most important interlocutor on primacy, partly because of the importance of Germany to the EU, but also because of the importance of the German constitution to Germans, and because of the quality and depth of the judgments that the BvG provides. In response to repeated challenges to the constitutionality of EU law, often brought by disgruntled German constitutional law professors, the BvG has taken lengthy and nuanced positions that reserve for the BvG itself the very last word on issues of constitutional importance, while also making sure that constitutional conflict never arises in practice and EU law is always applied.
However, the fact that there is still some fundamental dispute about “ultimate” primacy and the relationship between the treaties and national constitutions highlights what is far more remarkable and important: the absence of dispute about primacy in other, more mundane situations. The court’s two founding doctrines are now almost universally accepted and applied. Courts throughout the EU have become used to setting aside national laws, even parliamentary legislation, if those laws conflict with a directive, a treaty article or a judgment from Luxembourg.
This is perhaps even more surprising given the style of the judgments in Costa v. ENEL and van Gend en Loos. Rather than persuasive law-based argument, the judgments contain a series of claims that are essentially self-fulfilling judicial legislation: that the EU is not merely a creature of international law, but a new and autonomous legal order; that it creates enforceable rights for individuals; and that it is an integral part of the legal system of the member states. All these propositions appear now to be true, but it is thanks to the court that this is the case.
However, both decisions also mention the need to make EU law effective. If EU law were only adjudicated before the European Court of Justice, then it would not be accessible to the individual and in practice would have little social or economic impact. And even direct effect would achieve little if member states could still deviate from EU law when complying with it became inconvenient, making primacy an essential part of the package.
This focus on effectiveness has become a continuing motif in the judgments of the court, which consistently find that EU law is what it needs to be—that the open and often rather vague treaty texts should be understood to contain, or imply, the rules and principles necessary to make those texts effective law. Costa v ENEL and van Gend en Loos were just the beginning of these kinds of judgments. In Francovich, the court discovered a general principle that states should be liable to pay damages when they breach EU law. There is nothing in the treaty that states this, but the court found the principle to be “inherent in the system of the treaty,” since otherwise states would be able to ignore EU law without domestic consequences, making the rights granted in EU law ineffective.
The same logic has taken EU law into areas of national law that had commonly been considered to be outside the EU’s competence, such as civil procedure: matters such as rules of evidence and time limits. This is far from an arcane matter, since the largest part of most lawsuits is procedural argument, and cases are often won or lost on such points. In addition to mattering in practice, rules of civil procedure are also deeply rooted in national legal traditions, whose procedural styles vary a great deal; there is probably less agreement on how justice should be done than on what it actually is. Nevertheless, the court has repeatedly found that wherever some national procedural rule might make it unreasonably difficult to enforce rights granted under EU law, that rule must be set aside.
That national courts and governments do in fact accept all this, albeit with some grumbling, must be partly because they realize its practical merits. It may at times raise the blood pressure of textually fixated lawyers, but it means that the socio-economic agreements at the heart of EU law actually get enforced. For most states, most of the time, that is worth the trade-off in loss of autonomy.
Things could have been different. Had enough national courts or governments found the court’s judgments excessive, they could have found ways to limit those judgments, and there are plenty of reasons why they might have done so. In some member states, the idea that a court would set aside parliamentary legislation violated ancient legal taboos, and in most member states constitutional review of law was only possible under very limited circumstances. Making all national law conditional upon compliance with EU law, without any clear legal basis for doing so, almost invited national rebellion.
That rebellion did not happen to any significant extent, and so the court ushered in a legal revolution whose impact can hardly be overstated, even if that impact emerged slowly rather than overnight. States have been essentially transformed into subordinate jurisdictions very much resembling the regions of a federal state. Their last vestiges of sovereignty are their armies and their right to leave the EU, which is often considered unrealistic. It is sometimes said that what took 200 years and a civil war in the United States happened peacefully in a couple of decades in the EU on the basis of a few court judgments. What the European Court has accomplished is little different from a judicial coup d’etat.
To be fair, the court has never hidden any of this, remarking in Costa v. ENEL that member states, in joining the EU, had permanently surrendered a part of their sovereignty. To be quite clear about this, here the court is saying that member states are not fully sovereign anymore: That is the price they pay for the benefits of deep socio-economic integration. It is the EU’s readiness to impose this price on member states, and member states’ readiness to pay it, that makes the EU fundamentally different from other regional organizations or trading blocs.
How much all this matters in practice depends on what EU law contains. The wider the law’s scope, the greater the loss of sovereignty will be. It is therefore significant that EU law has grown continuously in all directions, under a number of different impulses. This growth has resulted partly from member states’ amending the basic treaties to add new chapters, giving the EU powers to legislate in new fields, such as criminal law and immigration, and to act on behalf of the member states in aspects of foreign affairs. But the policy area at the heart of the EU’s purpose and identity, which also has the greatest impact on the member states, is one where the court’s interpretative role has been central, and where once again the pursuit of effectiveness has been used to take EU law to places few expected it to go.
This policy area is the internal market. The treaty defines the internal market as a geographic area without barriers to the movement of persons, goods, services and capital among the member states, and a series of treaty articles prohibit national measures that restrict such movement. Alongside this, the treaty gives the EU the power to adopt legislation aimed at the establishment and functioning of the internal market. An enormous part of the debate about what the EU does and should do has been about two seemingly innocuous textual questions: What is meant by “restricting” free movement, and what does the “establishment and functioning of the internal market” entail? The court’s interpretation of these terms in the treaty has been a major contributor to social and economic change.
In the case of restrictions, the court has taken what is often called an “effects-based” approach, looking not at formal categorizations, but merely whether some measure might have the effect of inhibiting cross-border movement. As a result, it has found, just to name a few examples, that medical patients have the right to go abroad for medical treatment at the expense of their home state, with significant impacts on the organization of healthcare systems; that students have the right to study abroad and pay the same fees as local students, even if these fees are—as is usually the case in Europe—enormously subsidized by the state, with the result that student migration has become a significant financial burden for some states; that states that have laws on product specifications, such as Germany and its well-known Rheinheitsgebot for beer, are as a general rule not allowed to apply these rules to imports, creating considerable competitive tensions; that government influence in a privatized company is a restriction on the free movement of capital since it may discourage foreign investors; that planning rules limiting the size of shopping centers are a restriction on the movement of companies (legal persons), since companies may wish to build larger shopping centers than are allowed by a given state’s laws; that rules restricting the use of goods—such as restrictions on use of jet skis, which were at issue in the recent case Mickelsson and Roos—can dissuade consumers from buying those goods, and hence reduce imports, and so may constitute restrictions on trade; that a refusal to grant a residence permit to the non-European illegally present parent of a child who is an EU citizen violates that child’s free movement rights, since it effectively forces the child to leave the EU; that the bodies regulating professional sports are obliged to make transfers between teams reasonably accessible for dissatisfied players; and perhaps most controversially among the court’s recent decisions, that a trade union is prohibited from taking industrial action aimed at keeping out low-cost foreign workers from other EU member states.
This small selection of examples shows how almost every area of regulation has been invaded by treaty rules that were once thought to be largely about protectionist trade barriers, a sort of European internal GATT and General Agreement on Trade in Services (GATS). Not only do these interpretations go beyond most conceptions of protectionism, but they do not apply only to states either—they also bind private persons or organizations that have sufficient de facto power to restrict trade or movement, as in the cases of trade unions and football authorities. It is sometimes hard to think of an area of national policy or law that could not be argued to inhibit cross-border movement for someone, somehow.
The other side of the internal market is EU legislation. There are very different beliefs within the EU about how much regulation markets actually need, as well as very different ideas about which level of government is the appropriate authority to provide that regulation, making it inevitable that eventually the European Court will have to rule on these sensitive questions. But what does it actually mean for the EU to have competence to legislate for the establishment and functioning of the internal market?
The court’s answer, most famously in the Tobacco Advertising case, decided in 2000, was that EU measures that claim to serve the internal market must contribute to one of two goals if they are to be valid: Either the measure must help remove barriers to movement, for example by harmonizing product standards or technical specifications or by providing for procedures for mutual recognition, or the measure must remove nontrivial distortions of competition. Such distortions are said to arise where levels of regulation vary between member states, so that businesses in one state enjoy a competitive advantage over businesses located in more heavily regulated states. Since such a situation offends a certain conception of fairness, the court considered it legitimate to adopt common EU regulation in such cases.
This may all sound reassuringly technical, but it imposes serious constraints on national legislators. The court, perhaps very European in its outlook, has produced formulas that mean almost any difference between states can be identified as a problem. Distortions of competition, as understood by the court, are ubiquitous, and removing them all would entail almost a complete transfer of law to the EU level. And what kind of differences between states might tend to inhibit movement between them? Language? Educational systems? The quality of the welfare state? Removing barriers to movement is as open-ended an idea as that of “fair” competition.
The court in Tobacco Advertising did not ensure that such things will be harmonized, for there is no political will within the European Parliament to do so. What the court did do was ensure that the limits to the legislative power of the EU are less legal, and more political, than many Europeans are happy with. In practice, if there is political will within European institutions to adopt a certain law, then it should be possible for it to do so in a way that fits with the competence definitions of the court.
This offends against the idea that the EU is a creature of conferred powers—an organization that can only undertake the tasks specifically entrusted to it, leaving all other matters to the member states. The spread of internal market law had been a source of political discontent for some time in some member states, notably Germany, and after Tobacco Advertising there was considerable unrest within the EU at this apparent approval of what has been called “competence creep.” The lack of meaningful limits to EU powers is often remarked to have been one of the driving forces behind the call for an EU constitution and a revision of the treaties. In some states, such as Britain, such proposals were widely perceived as an attempt to broaden EU powers even more. Meanwhile in Germany the effort was more commonly regarded as an attempt to bring order and certainty to what was becoming in some eyes—as a result of the court, and an EU Commission always enthusiastic to propose ambitious new legislation and now legally free to do so—an uncontainable legal monster.
However, as is well-known, the proposed constitution was defeated. What is less well-known, but just as true, is that the proposed constitution failed to introduce any serious changes to the scope of EU powers anyway. The current treaties, adopted at Lisbon shortly after the constitution was finally abandoned, are in fact very similar to the proposed constitution in all but name. They are also very similar to the EU treaties that preceded the proposed constitution. Although there were changes in various areas, the process that began with a constitutional conference—modeled on the one that took place in the U.S. two centuries ago—and ended with the Treaty of Lisbon, ended up being a resounding, if bad-tempered, endorsement of the status quo. Despite concern about the breadth of EU law, the way the court interprets it and the limited space that it leaves to member states, no one was able to come up with an alternative that could attract support.
This is yet another example of how the EU simultaneously attracts the criticism of the states and their passive consent. In particular, the fact that the court has been able to get away with so much disruptive judicial legislation, and the states have not combated it with restraining legislation of a more traditional sort, or with revisions of the treaty, shows that to describe the European Court of Justice as a body imposing radical doctrines is only half the story. The EU and the court are also giving the customer what it wants, providing a market-based medicine that appeals to significant domestic lobbies who perceive that however imperfect the EU and the court may be, they offer a way to save naturally insular and overregulated European states from their own most self-destructive instincts.
This is even more the case in the aftermath of the global financial crisis and the European debt crisis that ensued. The package of measures adopted to enforce member state budget discipline, and to prevent the banking industry from re-imploding, is the greatest transfer of power to the EU in at least two decades. Yet the Euroskeptic British and Dutch prime ministers, David Cameron and Mark Rutte, were among the first to call for strict EU action to prevent states from overspending, rushing to Brussels as if they could not give away power fast enough. The lesson here appears to be that when politicians are faced with a choice between their ideology—in this case fiscal conservatism—and the sovereignty of their state, the former wins hands-down. Given the anti-nationalist, postwar origins of the EU, it is perhaps encouraging that this is the case.
Nevertheless, this European light-heartedness about surrendering power makes the EU more troublesome as a role model for other regional organizations. While they often observe the EU with interest and respect, most states would rather not give away as much sovereignty as the Europeans have done, and would like to bring EU-like effectiveness and depth of economic integration to their region, while avoiding supranational interference in sensitive policy areas and institutions. But nothing in the EU story suggests a way of doing this. Making trade agreements effective, and creating broad-ranging, strict constraints on national autonomy are two sides of the same coin, the latter being what makes the former possible. Perhaps one can have one’s cake and eat it too, that is, have integration as well as sovereignty. But the EU is not instructive on this.
An alternative reading is more encouraging. If the court and EU law have had to disrupt national states so severely in order to create a meaningful shared market, this may be because the intensity of regulation in European states was unparalleled. Centuries of guilds, professions, class-based politics, prickly regions and porous land borders had created uniquely densely regulated countries in which much of the law was designed to stabilize the social order and protect it from internal and external disruption. Such law is, of course, inherently anti-competitive. The work of the EU is in that sense a distinctive product of Europe, and if the court sometimes seems more like a social engineer than an arbiter of a trade agreement, then perhaps this is because the only way to create trade in Europe is to re-engineer its states.
This suggests that other regions may not need to go as far as the EU has done in order to get real benefits from open borders. However, it is hard to know this in advance. Little of the court’s case law would have been predicted in 1957 when the Treaty of Rome was signed. The court’s case law has been largely a reaction to circumstances and lawsuits that raised issues that had to be dealt with. As every lawyer knows, one simply cannot predict all the consequences that new laws will have, certainly not ones as broad and open as the EU treaties. Thus the most serious lesson for states considering their own regional arrangements is that if one creates a powerful court, and gives it a text with serious ambitions, one embarks on a path of integration whose end point is unknown, but with the uncomfortable certainty that the further one goes, and the greater the degree of integration achieved, the higher the costs become of turning back.
Gareth Davies is professor of European law at VU University Amsterdam and on the governing board of the Amsterdam Centre for Contemporary European Studies (ACCESS). He is co-author of one of the leading textbooks of EU law, “EU Law” (Chalmers, Davies and Monti; Cambridge University Press) and author of numerous journal articles on the European Union, its court and its economic law.