Can We Make the European Fundamental Freedoms Less Constraining? A Literature Review

Abstract This article reviews the legal and political science literatures on the extensive interpretation of the European fundamental freedoms and on possible ways out. The common market rules were originally laid down in an international treaty, the Treaty of Rome. In functional terms, this treaty became a de facto constitution, implying that its content, including the fundamental freedoms, were constitutionalised. We review how this constitutionalisation constrains legislators at the Member State and European levels. In order to identify possible ways out, we also review several reform options: institutional reforms of the European judicial system; the de-constitutionalisation of the fundamental freedoms; counterbalancing these freedoms with further strengthened social rights; and contestation of over-constitutionalisation within the given primary law framework. We conclude that reform options are available that could gradually free the legislators from the over-constitutionalisation of the common market rules. Such options should become part of the ‘Conference on the Future of Europe’ process and debates about EU reforms in general, as more flexibility is warranted in a heterogeneous EU.


I. INTRODUCTION
With the lost referenda in France and the Netherlands in the year 2005, the idea to provide the European Union with a formal constitution died. In functional terms, however, there had been a constitution anyway. 1 In declaring the supremacy and direct effect of European law in 1963/64, the European Court of Justice ('ECJ', now * We are most grateful for very detailed comments that Nik de Boer gave us, which helped to significantly improve our argument. Further thanks are due to Gareth Davies  are the degrees of freedom of the European policymaker and its judicial and procedural capacity to partly or entirely correct the fundamental freedoms jurisprudence of the CJEU. With regard to potential countermeasures, we review a number of reform options: institutional reforms of the European judiciary, the transfer of the fundamental freedoms to secondary law, defining the fundamental freedoms as bans on economic discrimination rather than as bans on restriction of the Common Market, counterbalancing the fundamental freedoms with further strengthened social rights, and defining areas that shall remain outside the scope of the common market freedoms.
Since far-reaching institutional reforms of the European judiciary are particularly difficult to achieve, we argue, the Member States are better off to write a more autonomy-protective reading of the fundamental freedoms into the treaties if they can agree on such a reading. Going back to the fundamental freedoms as bans on transnational economic discrimination and identifying policy areas that shall not fall into the reach and scope of the market freedoms are the most promising reform strategies. Such strategies, however, require treaty changes. If no respective window of opportunity opens up in the short to mid run, judicial and political reformers should focus on more contestation by the means of oversight by national courts and a more courageous political testing of the 'fundamental freedoms rigidity' of European secondary law.
In the following, we will proceed in two steps. First, by reviewing analyses from jurisprudence and political science, we will describe how the Court's broad interpretation of the common market rules as constitutional individual rights has moved policy matters towards the discretion of judges (II.A-II.C), providing European integration with a particular liberal economic bias. In a second step, we will discuss reform options aiming at more autonomy-protective market freedoms (III.A-III.C). 9 As we will demonstrate, the over-constitutionalisation of the European market rules cannot be changed easily. But reform is not impossible either. Strategies to limit the impact of the fundamental freedoms and thereby to enlarge the room for manoeuvre of democracy at the European and Member State levels should become part of the 'Conference on the Future of Europe' process.

II. OVER-CONSTITUTIONALISATION OF THE COMMON MARKET RULES
When the EEC was founded in 1958, the Treaty of Rome included several aims for cooperation among the six founding Member States. Until the late 1960s, the premier cooperation goal was building the common market, relying on the four freedomsrelating to the free movement of goods, services, people (of workers and of establishment), and capital-as well as cartel law and state aid, next to realising a common agricultural and transport policy. Later treaties added further policy aims, such as citizenship rights and monetary union at Maastricht. Establishing direct effect and supremacy in the 1960s gave constitutional status to these policy goals.
In this Section, we will revisit the literature on how this constitutionalisation of the Treaty of Rome's market freedoms frees individual economic activity from state regulation (II.A), imposing a liberal bias (II.B), and far-reaching constraints on majoritarian policymaking (II.C).

A. Economic Freedoms as Constitutionalised Individual Rights
The CJEU's case law on the fundamental freedoms builds the core of the single market. A large literature has analysed the dynamism that this case law provides to the integration process. Originally, the freedoms were interpreted as prohibitions of discrimination. Member States were barred from imposing additional restrictions on goods, services, or persons from other Member States, but market participants had to comply with the domestic regulatory setting. With its famous Dassonville decision, the Court extended the scope of the freedom of goods in today's Article 34 Treaty on the Functioning of the EU ('TFEU') by arguing for a much more farreaching prohibition of restriction. All 'measures likely to prohibit, impede or render less attractive' 10 the exercise of the market freedoms, even if equally imposed upon domestic and foreign market participants, may violate European law.
Under this interpretation, Member States' market regulations, when potentially restricting cross-border economic activity, have to be justified with a proportionality test, originally developed in the Cassis de Dijon decision and subsequently applied to all market freedoms following the ruling in Gebhard. 11 According to this test, restrictions are justified if they do not discriminate against foreigners, if they are justified by imperative requirements of the general interest, if they are suitable for securing the attainment of the objective, and if they do not go beyond what is necessary. Reynolds 12 summarises the vast literature on the fundamental freedoms by arguing: 'Most free movement lawyers would agree that the transformation of free movement has occurred across four main axes, covering the expansion in the material scope of the market freedoms, the widening of their personal scope, the extension of their direct effect, and the strengthening of a decentralised system of enforcement'.
As a consequence of this broad interpretation, democratic choices of Member States are significantly constrained even in areas in which they have not provided 10  the European legislator with competences. 13 In theory, the application of the fundamental freedoms is restricted to transnational constellations. In practice, however, it is difficult to come up with economic and social regulations which do not at least potentially also affect foreign exchange. 14 The literature agrees that it is difficult to make out with clarity where Member States retain authority to regulate their markets. 15 Private litigants requesting the preliminary reference procedure of Article 267 TFEU in national courts as well as the Commission with the infringement procedure under Article 258 TFEU can use the re-interpreted market freedoms as an effective weapon against unwanted regulations at the Member-State level. 16 The dynamism that European integration can draw from the interpretative shift to a non-restrictions approach can hardly be over-stated.

B. The Liberal Bias
The literature has not only identified dynamism, undermining Member States' regulatory competence, it has also identified and often criticised the resulting liberal bias which Scharpf has famously described as 'negative integration'. 17 Once the fundamental freedoms are interpreted as prohibitions of restriction, the common market can be realised without agreeing on common rules, merely by less and less accepting different national market regulations as legitimate barriers to trade. The dynamic that is unleashed is one of considerable liberalisation. 18 Conflicts with the market freedoms only emerge where regulations, from the point of view of the litigant, intervene too much. Where national liberalisation measures have gone too far, undermining public policy objectives, European law offers no lever for regulatory reform. 19 Thus, the balance of power between proponents and opponents of market-correcting measures is changed, because only those opposing 13 F Fontanelli and A Arena, 'The Charter of Fundamental Rights and the Reach of Free Movement Law' in The Reach of Free Movement, note 10 above, p 296; Horsley, note 8 above. In the words of Azoulai, the market freedoms have a 're-programming function' even in areas in which the European level lacks political competences: L Azoulai, 'The Court of Justice and the Social domestic market-restricting measures may turn to European law. In fact, Garben argues that the primacy of economic freedoms 'gives governments a tool to eliminate social rules that they do not support, without having to face the normal political contestation at the national level. This leaves social rules systematically vulnerable to attack at both the European and the national level'. 20 Litigants depend, however, on the willingness of national courts to apply EU law and to demand clarification from the CJEU, when in doubt. In deciding for a reference to the CJEU or against, Member-State courts can either safeguard their national legal order from European law influence, or push legal development in a way that judges regard as preferable. Disagreement in the national court hierarchy on legal development fuels this process. The overlapping incentives of litigants, lower national courts, and the CJEU can set in motion a self-dynamic process that pushes market integration further. 21 The reference to litigants makes clear that Member States are not unitary actors, and that liberalisation policies being furthered via the European level find important interlocutors at the national level. In his important book 'Eurolegalism', Daniel Kelemen has shown that the European multi-level system offers incentives for strategic litigation that resemble those in the US context. 22 The provisions of EU law can be regarded as an opportunity structure for actors in search of a more favourable regulatory environment. 23 Supremacy and direct effect allow them to litigate in domestic courts for the application of EU law, if this appears more beneficial, often resulting in preliminary questions to the CJEU, providing it with the opportunity to develop its case law.
The fundamental freedoms not only potentially challenge Member States' laws. They also apply horizontally. 24 29 These rights bind EU-institutions, and national institutions when executing EU law. 30 Viking and Laval not only became important signposts for the extent to which the market freedoms endanger social rights, they also put a halt on hopes that a further strengthening of fundamental rights at the European level could sufficiently tame the interpretation of the common market rules (compare Section III.C): while recognising the right to strike as a European fundamental social right for the first time in these rulings, 31 the Court curbed its exercise as a violation of the fundamental freedoms right away.
The Laval ruling can serve as a good illustration of Kelemen's argument of strategic litigation. Laval had a deep impact on Swedish industrial relations. The Confederation of Swedish Industry financed the legal dispute and used the ruling to put pressure on the unions. 32 Within the national context, the employers would not have had the political clout to enact these changes that could be realised via European law.
The literature states an increasing overlap between fundamental freedoms and fundamental rights. 33 Yet, the conflict between market freedoms and fundamental rights is not solved on an equal footing in the way of 'practical concordance' 34 as practiced by the Bundesverfassungsgericht when balancing different fundamental rights. To the contrary, the Cassis de Dijon procedure, developed for identification and abandonment of illegitimate protectionisms, is intentionally asymmetric in favour of the fundamental freedom. 35 While Trstenjak and Beysen argue for a symmetric double proportionality test, where 'the realisation of a fundamental freedom constitutes a legitimate objective which may limit a fundamental right and vice versa', 36 there appears broad agreement in the literature as to the continued primacy of the fundamental freedoms. 'Social fundamental rights and economic freedoms have not been recognised as having the same legal force in EU law', Robin-Olivier writes. 37 The diagnosis is one of a 'constitutional asymmetry' in favour of economic rights. 38 One may suspect that liberal market economies asymmetrically profit from the judicially driven liberal bias. But reality is more complex. Since all kinds of socioeconomic orders rely on certain balances between market and non-market features, there are no clear winners and losers among Member States. 39 Brexit can be taken as an illustration. Not only did criticism about the powerful role of the Court in the EU policy process play an important role in the British discussion about EU membership. Far reaching social rights under the free movement of labour were also in the centre, although one should have expected the UK with its rather liberal type of capitalism to be a central beneficiary of the liberalisation process.
In summary, the constitutionalisation of the fundamental freedoms imposes a significant liberalising bias on integration.

C. Constraining EU Policymaking
The common market rules constrain, as we have seen above, democratic discretion at the Member State level. This part of the story is not controversial in the literature reviewed here. In contrast, the effects on the political system of the EU-that is, on European secondary law-are much more controversially discussed. Let us start from a purely legal point of view. Most European law students or political scientists without particular expertise on the common market would spontaneously argue that the secondary lawmaker is bound by the market freedoms, simply because secondary law is bound by primary law. But whether this holds true for the fundamental freedoms is more contested than one might expect.
Von Bogdandy, 40  legislator may use secondary law to correct CJEU jurisprudence interpreting the fundamental freedoms. Other authors such as Kingreen, 41 Mortelmans, 42 Sørensen,43 and Sypris 44 argue that the fundamental freedoms bind the European legislator in a less rigid way than they bind Member State legislators. Yet others, such as Frenz, 45 argue in favour of a neat and strong binding. It is fair to conclude from this part of the literature that only the extreme positions, if anything, can be ruled out: it is neither very likely that secondary law operates completely outside the range of application of the fundamental freedoms, nor that the common market rules legally constrain European secondary law precisely as much as they constrain Member State law. Everything in between these extremes, however, is ambiguous (we will come back to this problem under Section III.D). 46 The legal dimension, however, covers only part of the problem. There are, in addition, political and procedural repercussions of case law. Even if the fundamental freedoms bind the European legislator rather loosely in legal terms, the CJEU may nevertheless overshadow and guide the democratic process. In the words of Garben: 'The effect of the case law on the bargaining conditions in the Council makes it difficult for the legislator to deviate from that case law in practical terms'. 47 Scharpf emphasises that consensus requirements put measures of positive integration at a significant disadvantage. 48 Procedurally, the Commission is the only European institution that can start legislative processes, and it almost always bases its proposals on the existing interpretation of the Treaty. It is therefore challenging to get diverging readings of the fundamental freedoms in. Then, a double majority of 55% of Council members representing at least 65% of the population in the Council and a majority in the European Parliament ('EP') are needed. Thus, if the legislator aims at correcting a CJEU ruling, the unwanted CJEU ruling, rather than the status quo ante, will be the fall-back option in the legislative process. 49 How far-reaching are these constraints? In the political science literature, we find again two positions-one that tends to play them down, and one that emphasises the 41 T Kingreen, 'Allgemeine Lehren der Grundfreiheiten' in C Calliess and M Ruffert (eds), EUV/AEUV. Das Verfassungsrecht der Europäischen Union mit Europäischer Grundrechtecharta. Kommentar strictness of the constraints that CJEU case law imposes on the EU policy process. The literature that argues in favour of rather weak constraints can be summarised with a view to two foci: one part of the literature argues that the CJEU aligns its rulings closely with the preferences of Member States. Thus, from this perspective, rulings do not matter much for EU legislation because they broadly lay down what Member States want anyway. Another line argues that the EU legislature is well able to depart from rulings, implying that the power of the Court has been overrated by Scharpf and those arguing in his direction. According to this view, politics rather than judicial reasoning have the upper hand. Naurin and colleagues 50 as well as Carrubba and Gabel 51 have analysed Member States' observations to cases at the CJEU. Member States can join proceedings when they see their interests implicated in any way. In this case, they can tell the Court how they regard the legal question at hand. The Commission normally joins all procedures and Larsson and Naurin find that the Court follows the Commission in 75% of all cases. 52 Larsson and Naurin see judges respond to potential legislative override. Similarly, Carrubba and Gabel argue that judges are concerned about potential noncompliance of Member States.
These are important findings. The quantitative analyses, however, do not allow differentiating between more relevant and hardly relevant cases, those that steer the integration process and those that do not. In path breaking cases like van Gend, Costa, Dassonville, Cassis, and others, the Court often decided against Member States' opinions. That many other cases were more closely aligned with Member States' preferences does not matter so much, once the Court with its own rulings constitutionalised the Treaty, allowing it to change the integration path of the single market by making the requirement of full harmonisation obsolete, for instance. In fact, in a highly interesting network analysis, Larsson and colleagues show how some rulings establish new precedent and stand in the centre of case law development in an area. 53 By not being able to weigh the qualitative importance of rulings, the quantitative analyses do tell us that the Court does not fully disobey Member States' preferences, but not more. As Karen Alter argues, a court that would consistently rule against political preferences would be odd. 54 There is ample scope for case law to alter the status quo, and therefore the bargaining dynamics in EU policymaking. We are back here at Kelemen's insight about strategic litigation and the use of EU law to bring about European or domestic institutional change that would not receive sufficient political Martinsen, in her important book, An Ever More Powerful Court?, 55 combines a quantitative analysis of all social policy legislative proposals between 1958 and 2014 as to whether they respond to case law with a qualitative analysis of three cases: the Posted Workers Directive and the Patient Mobility Directive, both influenced by case law on the freedom to provide services, and the Working Time Directive, where case law related to secondary law. Martinsen argues that the importance of the Court has been overrated. She distinguishes four reactions of the legislature to case law of the Court: codification, modification, non-adoption, and override, and hardly finds examples for instances of override, where the EU legislature overturns a ruling. Rather, she bases her assessment on the strength of the legislature on frequent instances of 'modification', where legislation responds to case law but adds onto it.
'Where the "judicialisation of politics" theoretical approach assumes politics to be the receptive part of judicial dialogues, we find that politics also influences the judiciary', 56 Martinsen points out. This surely makes sense, also given the insights by Naurin, Carruba and Gabels, and others. But also remember Alter's remark here, cited above: a court consistently deciding against the Member States' preferences would be strange. Note in addition that Martinsen adopts very strict criteria for 'codification', based on the assumption that case law could give full-fledged descriptions of policy, so that the enactment of any additional legislative rules is used to categorise the case as 'modification'. The weakness of the Court that the author finds is therefore also due to very high expectations concerning the extent to which case law could in principle shape policy, expectations that even scholars who argue in favour of a strict binding of the legislator may not share.
Other political science research emphasises the constraints that case law poses on policymaking. Bulmer had shown already in 1994 that the adoption of the merger regulation needs to be explained with the pressure of case law. 57  interpreting the Treaty strongly constrained the EU legislator. The failure of the EU legislature to respond to the Laval and Viking cases with the so-called Monti II regulation as well as to codify the Court's case law on direct company taxation show the limits of driving integration via the Court. At times, then, a very fragmented regulatory picture results as case law describes limits to national policies but no consensus is possible on how to embed this case law in a common EU policy. In sum, the research literatures of both legal and political sciences discuss the extent to which the CJEU's fundamental freedoms jurisprudence binds the secondary lawmaker controversially. It is fair to conclude that case law never fully dictates the content of secondary law. But it nevertheless changes the rules of the political game and often steers secondary law in certain directions. For us, the importance of the CJEU lies in its mandate to allow for European or domestic institutional change that could not be enacted by the means of politics alone. That said, we close this part of the review and turn to potential countermeasures.

III. REFORM OPTIONS
In the following Sections, we will consult those parts of the literature that deal with reform options. We will distinguish options targeting institutional procedures of the Court (III.A) from attempts to change the status and content of material EU law, in order to end the privilege of economic freedoms in EU jurisprudence (III.B-III.D).

A. Institutional Reforms
Suggestions for institutional reform of the European judicial sphere have a long tradition and are still part of the EU reform debate. For example, given the breadth of the CJEU's jurisdiction and the lack of appeal in all those areas where the General Court has no mandate, 60 there have been considerations of institutionalising a greater judicial specialisation, either through specialised courts, or through more specialised chambers. The hope is that greater specialisation would make the Court more attentive to the negative repercussions of its case law. A chamber specialised in social matters, for example, might be prone to more sensitivity with regard to the autonomy of social protection and the wage-bargaining regimes of the Member States.
Specialised chambers would reform the CJEU without fundamentally altering the logic of the European judicial system. The same holds true for the Rasmussens' famous call for permitted and published dissenting opinions on the side of the judges. 61 Much more fundamental is the idea, first voiced by Weiler, of complementing the 60 The General Court is mainly responsible for individuals' judicial review against EU administrative or legislative acts, competition law, as well as for employment matters regarding EU employees. 61  The debate about such ideas revived after the German Constitutional Court's ruling on the European Central Bank's asset purchase programs in 2020. 65 In this context, together with Sarmiento, Weiler updated his proposal. 66 The authors propose a new appeal procedure within the province of the CJEU, with a 'Mixed Grand Chamber' being composed of six CJEU judges and six judges from the highest Courts of the Member States, presided by the CJEU president. As in the earlier proposal, the new chamber would only deal with conflicts over the distribution of competences, and a decision validating a contested European measure would have to be supported by at least eight or nine judges.
In the context of our review on the fundamental freedoms, a decisive question would be whether a narrow (empowerment to legislate) or a wide definition of competence conflicts would apply. If Sarmiento and Weiler's 'Mixed Grand Chamber' would only deal with the supposed overstretching of the right to legislate, fundamental freedoms conflicts of the Laval kind would be out of its scope; if it would deal with all kinds of authorisation conflicts, its case load may become huge. Another question would be the purpose of the Chamber if European and national constitutional law clearly provide different answers to a problem, given that a 'meta law' above them does not exist. In such constellations, as Sölter has convincingly argued, 67 the compromise to be found can only be of a political kind-and we may ask why judges should be the architects of such compromises anyway. In line with such considerations, Scharpf has presented an alternative institutional solution by arguing for a political rather than constitutional check of CJEU decisions. He suggested that if Member States were seriously negatively affected by a ruling, they should be allowed to turn to the European Council and should have to accept the ruling only if a qualified majority of Member States supports the interpretation of the Court. 68 Given the rigidity of the constitutional structure of the EU, institutional reforms of the CJEU are, on the one hand, difficult to accomplish. On the other hand, reforms have indeed occurred in the past. The establishment of the General Court in 1989 as a Court of First Instance was the result of such a reform, responding to the increase in case load. The saga of reforming the General Court, however-ending in the doubling of the number of judges in order to achieve the backing of Member States, while abolishing the Civil Service Tribunal as a specialised court-surely limits the Member States' enthusiasm concerning further negotiations about reforms of the court system. 69 Would institutional reforms of the judicial system effectively curb the overconstitutionalisation of the common market rules? Both Weiler and Scharpf suggest (partly) intergovernmentally composed appeal bodies. This idea has a comprehensive logic: while intergovernmental and supranational institutions balance each other out in the European legislative and executive branches, the supranational CJEU lacks an intergovernmental counterbalance. 70 A more balanced European judiciary may consequentially prevent or slow down further extensive interpretations of the market freedoms. But it would, at the most, lead to a very incremental (if any) retrospective correction of the CJEU's interpretation of the fundamental freedoms that has emerged over decades. Even the introduction of an Appeal Court might therefore enact few changes to the over-constitutionalisation of the common market rules.
Recently, Scharpf suggested a more limited, procedural change. He proposed tackling the extensive development of case law by letting the preliminary references and infringement procedures apply to cases of potential non-compliance with secondary law only, but not to cases that concern the policy goals of the TFEU (Titles I-IV, VI, and VII). 71 If litigation was only allowed to ask whether national law conforms to EU secondary law, the interpretation of the fundamental freedoms could no longer dynamically undermine national regulations, and push European integration along.
We conclude from this Section of the review that solid ideas for institutional reforms exist. But the reforms would rather not turn back the overconstitutionalisation discussed in this literature review. They could, however, protect the status quo of Member States' regulatory competence to legislate. If Member States can agree on a more autonomy-sensitive reading of the market freedoms, they may be better off to write it directly into the treaties, a possibility to which we turn now. 69

B. Re-defining the Fundamental Freedoms
A peculiarity of the next literature parts is that they largely originate from German authors. This may be due to the prominent role of the German Constitutional Court ('GCC') within the national polity and within the EU. Its series of the so-called Solange rulings 72 on the limits of European supremacy were an important stimulus for the CJEU to develop its own fundamental rights jurisprudence. The GCC seems to have stimulated a more lively academic debate on the legitimate protection of legislative autonomy than other constitutional courts have done elsewhere. Along with his normative criticism of over-constitutionalisation and its impact on democracy, Grimm suggests de-constitutionalising European law by transferring the TFEU from primary into secondary law. 73 Although Grimm's suggestion does not specifically target the market freedoms, they would be part of the transfer. Let us therefore consider the pros and cons of that idea. The advantage would be that any re-interpretation of the fundamental freedoms would, from now on, be nothing else than a re-interpretation of secondary law. Consequently, whenever the way the CJEU interprets the fundamental freedoms becomes contentious, the EU legislator would undisputedly have the mandate to correct this case law, irrespective of the extent of the 'fundamental freedoms rigidity' of secondary law discussed under Section II.C. The enlarged probability of correction alone should encourage less expansive common-market jurisprudence. 74 However, all problems which derive from the Commission's monopoly of legislative initiation and from the veto points inside the political system of the EU would remain in place. 75 Clearly, changing the Treaty to such a significant extent has a low prospect to be realised. But are there valid counterarguments against the substance of the idea? The proposal would place the protection of the common market entirely into the hands of the European legislator. Because most of the CJEU's fundamental market acquis has been codified in European or national secondary law or both, 76 it would apply until the (EU or Member States') legislators impose changes. A wave of legal uncertainty among market participants needs not be feared. For the moment, the reform would only imply that national market regulations could no longer be individually 72  challenged. Remembering that trans-border rights are very comprehensive in the EU, particularly in comparison to the US, 77 and that the enlarged Union of 27 Member States requires more rather than less regulatory diversity, there are good arguments for this solution. But less invasive alternatives could be available. We have seen in Section II.A that the CJEU re-interpreted the fundamental freedoms by replacing the ban on the discrimination of foreign-market participants with a ban on the restriction of the common market. This re-interpretation allowed the Court to tackle a much wider range of potential protectionisms among the Member States, however, at the cost of blurring the line between protecting the common market and liberalising matters internal to the Member States with only modest (if any) significance for trade. We may therefore ask whether Pandora's Box may be closed today the same way as it was opened back in the 1970s: by correcting the interpretation of fundamental freedoms as bans on restriction and going back to their normative meaning as bans on the discrimination of foreign market participants.
We are not aware of academic authors who ask the Member States to write such a correction into the treaties, though they could put down in the TFEU that the market freedoms only target unequal treatment of internal and foreign market participants. 78 Rather, the CJEU judges are the addressees of this idea. Kingreen, 79 for example, argues that the fundamental freedoms are essentially bans on direct and indirect discrimination and that their interpretation as bans on restrictions even if equally imposed upon internal and foreign market participants has always been a misconstruction. He therefore asks the CJEU for judicial self-correction. 80 Concerning the potential effectiveness, first, the acquis would change only with new legislation because of the extent of codification. A wave of uncertainty among market participants would not emerge. The option would mainly curb potential future clashes between Member States' market-correcting regulations and the principle of non-restriction. However, second, the effects of the reform may remain below the reformers' aims because of the concept of indirect discrimination, without which discrimination bans cannot be effective. 81 Valta argues that most cases of restriction can also be modelled as cases of-widely interpreted-indirect discrimination, which could allow the Court to often stick to its extensive interpretation of the fundamental freedoms. 82 The effectiveness of the reform would, therefore, require significant empathy on side of the CJEU. It would have to recognise, accept, and implement the intentions of the reformers-but this caveat is likely true for any reform scenario.

C. Re-balancing Fundamental Freedoms and Social Rights
Another set of reform proposals suggests altering the relationship between fundamental freedoms and social rights. Such a proposal is the Social Progress Protocol that was drafted by the European Trade Union Confederation ('ETUC') in direct reaction to Viking and Laval. Among the proponents in the literature are Iossa, 83 Hayes et al, 84 Bücker, 85 and Bruun. 86 The proposed protocol text underwent some revisions after its first presentation in March 2008, 87 and has been politically supported by the S&D political group in the EP 88 and the Social Democratic Parties of many EU Member States. 89 The primary aim of the proposal is to shield the autonomy of collective bargaining from the European fundamental freedoms. Decisive for this is Article 3: 'Nothing in the treaties, in particular economic freedoms, shall have priority over fundamental social rights. In case of conflict fundamental social rights shall take precedence'. Thus, the proposal assumes an implicit hierarchy among fundamental freedoms and social rights and aims at placing the latter above the former. Critics have argued that the proponents are asking for 'super-fundamental rights', supreme above all other rights that are protected by European law. 90 This is an important point because the CJEU did not, neither in Viking and Laval nor anywhere else, declare the market freedoms to be superior to social or other fundamental rights, although the outcomes resemble such a hierarchy. 91 Rather, in Viking and Laval, the Court had let the exercise of the right to strike fall within the scope of application of the fundamental freedoms, and then proceeded as usual: it applied the Dassonville 92 formula on common market restrictions and performed the Gebhard 93 proportionality test. It is therefore unclear how much a ban on subordination 94 -'shall [never] have priority over'-could really change.
The experience with the Charter of Fundamental Rights is worth considering here. Not only does the Charter honour the right of collective action in Title IV, it also includes several other rights to solidarity. Being binding on all acts of the European institutions since 2009, the Charter could in principle have brought the change to the interpretation of the fundamental freedoms that the Social Progress Protocol aims at-but it turned out that the introduction of the Charter did not alter the relationship between fundamental rights and fundamental freedoms. 95 Robin-Olivier 96 sees an 'inequality of force' and criticises that the CJEU has even further strengthened the pro-market bias by referring to the Charter's 'freedom to conduct a business'. It is, thus, unlikely that additional strengthening of the normative meaning of the social rights could curb the interpretation of the fundamental economic freedoms.
Akin to the Social Progress Protocol, another reform suggestion 97 is to exclude areas from the scope of application of the fundamental freedoms. Examples for this already exist. Thus, the free movement of workers in Article 45(4) TFEU does not apply to employees in public services. In the Keck 98 decision, the Court ruled that sales methods are 'outside the scope' 99 of the ban on restrictions of the inner-European products markets. 100 Interestingly, in Keck, the Court explicitly underlined its intention to curb the liberalisation potential of the fundamental freedoms: 'In view of the increasing tendency of traders to invoke [the free movement of goods] as a means of challenging any rules whose effect is to limit their commercial freedom even where such rules are not aimed at products from other Member States, the Court considers it necessary to re-examine and clarify its case-law on 92 Procureur du Roi, note 3 above. 93 Reinhard Gebhard, note 11 above. 94 Bücker interprets the Social Progress Protocol not as the demand for 'super fundamental rights' but as a ban on subordination. Bücker, note 85 above, p 15. 95 Trstenjak and Beysen, note 33 above. 96 Robin-Olivier, note 37 above. 97 Further authors who share the general aim of a rebalancing of common market freedoms and social rights include Garben (note 20 above), who calls for a 'political awakening' (pp 52, 60) in the form of a political re-balancing of the judge-made relationship between the common market rules and social matters in the EU. Similarly, Reynolds, note 12 above, p 644, calls to tackle the 'structural disadvantage'. 98 Keck and Mithouard, C-267/91 and C-268/91, EU:C:1993:905. 99 Ibid, para 17. 100 According to an alternative reading of Keck, the Court argued that 'the legal commitment to free trade appears to entail the abolition of discrimination, or protectionism, but no more than this'. In this interpretation, the ECJ delivered exactly the 'back to non-discrimination' decision which Kingreen had asked for (compare Section III.B). See  Similarly, matters such as collective bargaining-the contested matters in Viking and Laval 102 -could in principle be exempted from the scope of application of the fundamental freedoms. 103 After all, in its Albany 104 decision the Court has already excluded collective bargaining from the application of European competition law; a treaty change, likewise, could exclude it from the scope of application of the fundamental freedoms. Bast et al 105 and Heuschmid,106 taking collective bargaining as an example, have made detailed suggestions regarding the possible wordings of such exemptions as well as their potential positioning within the TFEU. Horsley makes a similar point, asking the Member States to make clearer statements about areas which shall be excluded not only from the usual EU policymaking but also from what he calls the CJEU's 'direct policymaking'. 107 In sum, the essence of the option is to ask the treaty partners to steer the application range of the market freedoms politically rather than to leave the definition of 'Keck exceptions' 108 or 'Albany exceptions' 109 in the discretion of judges. 110 For that purpose, the treaty partners would have to write a negative list of matters that shall remain outside their scope, or perhaps even better, write a positive list of what shall remain within. However, as for all the other reform options discussed so far, a change of the Treaty would be necessary. Given that this is a high barrier, 111 it is useful to end this review with possibilities that leave the Treaty untouched.

D. Contesting Over-constitutionalisation
In principle, three types of actors could challenge the over-constitutionalisation of the common market rules within the existing primary law framework: the CJEU judges themselves, other actors in the multi-level European Court system, and the European legislator.
Focusing on CJEU judges, Horsley asks the Court to recognise the limits of the EU Treaty's framework in its judicial interpretations. 112 In principle, all claims that the CJEU's reading of the fundamental freedoms has become too extensive can be understood as requests for judicial self-correction and more self-restraint after that correction. In questions of detail, such self-correction can always occur, at the CJEU no less than elsewhere. 113 But to reverse the over-constitutionalisation of the common market rules would be such a far-reaching self-correction, decades after Dassonville and Cassis de Dijon, and more than one decade after Laval and Viking, that it would contradict the incrementalism of case-law development. At least, we lack any indication that such a revolutionary judicial U-turn may be forthcoming.
It is more likely that impulses for change occur from within the broader judicial system. The CJEU depends on the integrated European court system, where national courts pass on preliminary references to enquire about the interpretation of European law and where under the 'acte claire' doctrine, national courts independently apply European law alongside their national law. While there is no data on the independent application of EU law in national proceedings, the literature has shown that requests for preliminary rulings differ widely among the Member States. 114 In general, such differences motivate scholars to call for a more comprehensive training of national judges, in order to assure that supreme European law is uniformly applied throughout the EU. 115 This assumes that despite all heterogeneity among the Member States, uniform rules applied throughout and targeting increasing areas of national legal orders can make for a successful integration.
Davies has recently forwarded a very different argument in favour of interpretative pluralism. 116 Because constitutional courts have to be embedded into a political context, and the CJEU is hardly able to recognise and factor in the implications of its rulings in 27 different national jurisdictions, he argues that national courts have a critical duty towards the CJEU. 117 If the case law of the CJEU leads to salient political problems in a Member State, courts need to contest these implications, and have to critically challenge the case law before integration as such becomes at stake. 118 In principle, such challenges could also include the 'nuclear option' of constitutional courts declaring CJEU rulings as domestically non-binding. But such incidents are extremely rare for good reasons, and none of the few cases so far was a fundamental freedoms case. We are also not aware of proponents of the 'nuclear option' in jurisprudence or political science who specifically argue that constitutional courts should declare certain ECJ decisions on the market freedoms as non-binding.
Another initiator of change within existing primary law may be the European legislator. Remember that the extent to which the market freedoms legally bind the European legislator is contested in the literature, as discussed under Section II.C. For those who argue that the market freedoms tightly bind the secondary lawmaker, the opinion of Advocate General Sánchez-Bordona in the legal dispute over the reform of the Posted Workers Directive must have come as a surprise in the year 2020: harmonisation directives, he argued, are allowed to impose common market restrictions that would violate the freedom of services if single member states imposed them. 119 The CJEU has, so far, largely avoided declaring European regulations as violations of the ban on restrictions, as long as the regulations were nondiscriminatory. 120 Although the contours of the respective degrees of freedom remain unclear until today, these observations encourage the conclusion that secondary law has significant, but yet to be specified, legal leeway vis-à-vis the fundamental freedoms. As Davies criticises, 'if the legislature fails to punch its weight in the European Union, then that must be at least as much to do with its own choices: there has been little attempt to autonomously steer the course of integration, at least of economic integration'. 121 But remember that procedural forces bind the legislator, too. Practically, political actors who aim at contesting extensive interpretations of the fundamental freedoms on the side of the Court will often find themselves on the defensive, due to the Commission's monopoly to initiate legislation, due to the multitude of supermajority requirements and veto points in the political system of the EU, and due to the fact that the new status quo of the extensive interpretation on the side of the Court, rather than the status quo ante will always be the fall-back option (compare Section II.C). 122 We should nevertheless not confuse the practical rigidity of the political system of the EU with a potential constitutional ban on the political steering of the scope of the market freedoms by the means of secondary law. Political rigidity can be overcome under conditions of exceptional politicisation, as it happened in the case of the reform of the Posted Workers Directive. 123 In sum, the over-constitutionalisation of European law places significant responsibility in the hands of the CJEU. Nevertheless, if the EU legislator and national courts do not blindly accept the supremacy of European case law but critically reflect what it means for the political order at the European and the Member States' levels, the extensive interpretation of the fundamental freedoms can be challenged.

IV. CONCLUSION
With the announcement of the 'Conference on the Future of Europe' process at the end of 2019, the Commission and the European Parliament have opened up a debate about reforms of the EU architecture that shall last for at least two years. A first attempt to start such a debate was the Commission's 'White Paper on the Future of Europe', already published in 2017. 124 It is striking that the discussions which followed the White Paper were almost entirely focused on the distribution of political competences among the EU and its members. The EU, however, does not only act by passing legislation. The discussion remained largely silent with respect to European actions which proceed behind the backs of the European and national legislators. The constitutionalisation and extensive interpretation of the European market freedoms are among the reasons why so many European de facto policies are judge-made. Rather than just pushing for 'more integration', this kind of activism constrains political choices in fields with only modest significance for European integration. It also undermines majoritarian decision making and thereby political participation.
This constellation is neither God-given nor written in stone. But how can the fundamental freedoms become less constraining? The interdisciplinary research literature on European integration may offer help, but the respective literature strands are difficult to access. Specifically, to our knowledge, potential ways out of the overconstitutionalisation of the common market rules have never been systematised. To identify and critically evaluate reform options has been the purpose of this review. As we have shown, reform ideas are available that, if implemented, could correct the ever-more extensive interpretation of the European fundamental freedoms, without 122 Scharpf, note 19 above. 123 What we would nevertheless rule out is that a 'negative list' on the scope of application of the fundamental freedoms, such as the one discussed in Section III.C, could be introduced in secondary law. Such a list would have to be placed where the fundamental freedoms are placed, that is, in the common market chapters of the TFEU. 124 European Commission, White Paper on the Future of Europe: The Way Ahead (European Commission, 2017). being overly invasive and therefore without endangering the transnational access to the inner-European market.
The reform options discussed in this review differ with respect to their invasiveness, persuasiveness, and probability of being realised. Complete overhauls of the European judicial systems are particularly unlikely to succeed. In our view, restricting litigation via infringement procedures or preliminary references to secondary law, narrowing the scope of the fundamental economic freedoms to the equal treatment of national and EU economic actors, and defining areas such as collective bargaining that shall be outside the range of application of the market freedoms are particularly promising reform concepts. But they all require Treaty changes. If no respective window of opportunity opens up, a better oversight by Member State courts and a more courageous political testing of the 'fundamental freedom rigidity' of European secondary law are the only options left. But even raising the academic and public awareness towards the power of the Court's case law would be a step forward.
Reforms of the fundamental freedoms have the potential of enlarging the democratic discretion at both the Member State and the European levels-a positive-sum game that could serve the legitimate call for a better autonomy protection in sensitive policy fields, and that could make European elections more meaningful at the same time. Enlarging the EU reform debate towards such options would therefore also break-up the unproductive split between 'integrationists' and 'EU sceptics'. In general, careful interdisciplinary work at the intersection of jurisprudence and political science can bring about results that offer orientation in the complex debates about the future of the European Union.