A. Introduction
The European founding treaties and their subsequent amending instruments have fundamentally changed the European conflict of laws landscape over the past fifty years, with the last ten years in particular witnessing a surge in legislative activity.
The Treaty of Amsterdam not only introduced notable institutional and material modifications to both the Community and the European Union, it radically reformed the position and status of private international law, bringing it within the ambit of first pillar competence. In contrast, the adopted though as yet unratifiedFootnote 1 Lisbon TreatyFootnote 2, will treat conflict of laws matters with a comparatively light touch.
This article traces the historical development of European conflict of laws, analysing the issues of material internal and external competence, as well as of geographical scope.
B. Overview of the Development of European Private International Law until the Reform Treaty
European private international lawFootnote 3 long predates the invention of concepts such as the areas of ‘Justice and Home Affairs’ and ‘Freedom, Security and Justice’. As soon as the founding treaties were concluded, the question arose as to the extent to which the objectives and requirements of the European Communities could impact on the conflict of laws.Footnote 4 The fact that the European institutions had no specific competence to enact private international law rules until the Amsterdam Treaty did not prevent the development of European conflict of laws provisions either within directives and regulations (2), or in the form of conventions (1).
1. Intergovernmental Cooperation
a) 1957–1993
Private international law issues did not occupy a substantial place in the Treaties of the European Communities. The only relevant provision in the 1957 Rome Treaty was Article 220Footnote 5. Several agreements were negotiated on this basis (the 1968 Recognition of Companies ConventionFootnote 6 and the 1995 Insolvency Convention),Footnote 7 but only one, the 1968 Brussels Convention,Footnote 8 entered into force.
Member States also cooperated in the field of private international law in the context of the so-called European Political Cooperation (EPC) from the 1970s onwards (long before the EPC was formalised by the Single European Act). Consultations between Member States took place with a view to taking common positions in international organizations, which led to the ratification by all Member States of the Luxembourg and Hague Conventions of 1980 (both of which are open to non EU Member States). However, the (only) two intra-European instruments adopted through the EPC (the 1987 European Legalisation ConventionFootnote 9 and the 1990 European Maintenance Convention)Footnote 10 have achieved little.
There has though been success for an important instrument negotiated during this period: the 1980 Rome Convention,Footnote 11 although it was neither based on a European Treaty provision, nor dependent on the involvement of European institutions.
b) 1993–1999
The Maastricht Treaty complemented the EC Treaty with two additional ‘pillars’ (Common Foreign and Security Policy and Justice and Home Affairs).Footnote 12 It formally introduced (and expanded) the possibility of ‘judicial cooperation in civil matters’, although in this area, as any other third pillar matters, the form of action remained essentially inter-governmental. Under Title VI (‘Provisions on Cooperation in the fields of Justice and Home Affairs’) cooperation in the field of private international law was governed in particular by Article K1Footnote 13 and K3.Footnote 14
These provisions had the potential to lead to an acceleration of changes in (European) private international law. First, the Treaty on European Union created legislative structures at the European level in the form of a coordinating CommitteeFootnote 15 and through the involvement of the Council, the European Parliament, and the Commission.Footnote 16 In addition, unlike Article 220 EC Treaty (which was maintained), Justice and Home Affairs provisions did not expressly restrict the areas of private international law where international instruments could be adopted, simply referring to ‘judicial cooperation in civil matters’.Footnote 17 The Maastricht provisions, although pragmatically and symbolically important,Footnote 18 have been generally criticised as ineffectiveFootnote 19 both in terms of the working methodsFootnote 20 used, and instrumentsFootnote 21 adopted. Only two agreements resulted and neither entered into effect: the 1997 Service Convention and the 1998 Brussels II Convention.Footnote 22 However, the pre-Amsterdam period did see private international law norms adopted in other forms.
2. Supranational Measures
a) EC measures before the Amsterdam treaty
Various types of isolated private international law rulesFootnote 23 are contained in a number of scattered directives and (although on a smaller scale) regulations, based on various sectoral provisions of the Treaty (the most important being Article 95 EC Treaty).Footnote 24 These were adopted in the context of measures harmonising substantive law. They generally aimed at extending or restricting the scope of European private international law norms,Footnote 25 specifying the territorial scope of the application of directives,Footnote 26 or protecting the legal framework of the internal market in certain areas against the choice of law of third States.Footnote 27 This fragmented approach not only rendered European private international law particularly opaque, but was detrimental to the coherence of its sources.Footnote 28
b) The communitarization of private international law under the Treaty of Amsterdam, and the Treaty of Nice
The Treaty of Amsterdam,Footnote 29 which revised the European Community Treaties as well as the Treaty on European Union, opened new perspectives for European private international law: not only was the Maastricht Union's objective of the development of ‘close cooperation on justice and home affairs’Footnote 30 replaced by the more precise aim of maintaining and developing ‘the Union as an area of freedom, security and justice, in which the free movement of persons is assured’Footnote 31 but, unlike its predecessor, the Amsterdam Treaty took concrete steps for the implementation of this redefined goal. Competence in the field of judicial cooperation in civil matters was transferred from the third to the first pillar, thus affording the Community power to legislate in the field of private international law.
Introducing a Title IV on ‘Visas, asylum, immigration and other policies related to the free movement of persons’, the Treaty of Amsterdam indeed enabled in Article 61 the Council to adopt ‘measures in the field of judicial cooperation in civil matters as provided for in Article 65’, with this latter provision qualifying and enumerating various domains in which such measures could be taken.Footnote 32 As is well known, this new regime which has given rise to very strong reactions in the private international law world,Footnote 33 has been fraught with uncertainties.
The new arrangements were designed to increase the efficiency of JHA cooperation as they provided for greater involvement of the EC institutions in the law making process and indeed ended many disadvantages of the previous system—protracted ratification procedures, the necessity to renegotiate accession conventions with each enlargement, the absence of inherent ECJ jurisdiction. Yet, the Community regime of private international law under the Amsterdam Treaty remained attenuated: first for a transitional period of five years, the Council's decisions were to remain unanimous and the European Parliament only had a consultative function. Second, by exception to Article 234, the ECJ had under Article 68 a reduced power of interpretation in that preliminary rulings in the private international law field could only be requested by national courts of final appeal. In addition, specific protocols provided by Article 69 established derogations in respect of the UK & Ireland as well as Denmark.
Further, the EC Treaty provisions impacting on private international law were complex and technically imperfect: Article 220, by then renumbered Article 293, continued to coexist with Article 65 (as well as the sectoral approximation of laws provisions); the location of Article 65 within Title IV linked judicial cooperation in civil matters and the free movement of persons and the uncertain drafting of this provision led to numerous contentions as to the exact scope of the new competence.Footnote 34
Whilst the Treaty of Nice brought about some changes in this field, these uncertainties remained. However an important amendment was introduced in Article 67, which provided for the immediate transition of Article 65 measures, with the exception of family law, to the Article 251 procedure (co-decision with QMV in Council).
In less than 10 years since the entry into force of the Treaty of Amsterdam, facilitated by the changes in voting requirements brought about by the Nice Treaty, there has been a dramatic increase of European Private international law sources. The scope and pace of these developments can be explained not only as a reaction to the meagre achievements under the third pillar approach, but also as a response to ‘negative’ and ‘positive’ ‘integration logic’:Footnote 35 the harmonization of private international law in Europe is seen (notably by the Commission) as responding to the imperious need to remove all barriers to free movement in the internal market, ensuring certainty and efficiency, and as well as preventing discrimination. Article 65 has thus been used to achieve the ‘reformatting’ of Article 220 and Article K3 conventions which had never entered into force;Footnote 36 to communitarize and amend Conventions already in force,Footnote 37 as well as to bring new initiatives to fruition, thus responding to every single one of the eight objectives set forth in Article 65.Footnote 38
Some of the questions initially raised by the new regime were in effect answered by the practice which progressively emerged: for instance, although Article 65 did not specify the type of Community action to be used (and instead referred to the notion of ‘measures’), and although several subparagraphs of the provision mentioned the ‘promotion of compatibility’ (which could have been interpreted as not referring to unification but simply a minima harmonization realisable through directives), the EC involvement in private international law has so far generally taken the form of precise (and by definition binding in their entirety and directly applicable) regulations.Footnote 39 These were systematically based on Article 65. Private international law provisions though did continue to be included in measures aimed at harmonising certain areas of substantive law,Footnote 40 with all the problems this entails.
C. Union Competence in Private International Law under the Reform Treaty
1. Internal Competence in Judicial Cooperation in Civil matters
Under the Reform Treaty Arts 61 and 65 EC Treaty are replaced by Arts 67Footnote 41 and 81Footnote 42 which are inserted in a Title V TFEU renamed ‘Area of freedom, security and justice’. Arts 67 and 81 are to be read in conjunction with new provisions on competence. According to these, Union competence is limited by the principle of conferralFootnote 43—which is in no way affected by the fact that the Union has a legal personalityFootnote 44—and Union competence in the area of freedom, security and justice is expressly shared between the Union and Member States.Footnote 45 Measures on judicial cooperation in civil matters can therefore be taken by the latter only to the extent that the Union has not exercised its competence or has decided to ‘cease exercising’Footnote 46 its competence,Footnote 47 bearing in mind that ‘when the Union has taken action in an area, the scope of this exercise of competence only covers those elements governed by the Union act in question and therefore does not cover the whole area’.Footnote 48
The use of competence by the Union further continues to be subject to the (now better defined) principles of subsidiarity and proportionality.Footnote 49
From a procedural point of view, Article 81 largely maintains the status quo: the ordinary legislative procedure (ie codecision & QMV) applies, except in the field of family law, where a passerelle remains available. The latter allows the Council to decide to transfer some aspects of family law to the ordinary procedure, but for this to happen the European Parliament must be consulted and there must be unanimity within the Council.Footnote 50 Additionally, the Treaty of Lisbon, in accordance with its objective to enhance the ‘democratic legitimacy of the Union’, now gives national parliamentsFootnote 51 the power to veto the use of the family law passerelle,Footnote 52 thus introducing a better safeguard for States' interests in this sensitive field.
From a substantive point of view, the changes introduced by the Reform Treaty appear limited. Aside from the fact that judicial cooperation in civil matters is now detached from the goal of free movement, the most obvious change is the ending of the requirement that measures to be adopted are necessary for the proper functioning of the internal market. Once the Treaty of Lisbon enters into force, the latter will simply be an illustration of the type of measures that the Union may take. Through a substitution of words (‘particularly when’ instead of ‘in so far as’), the legal basis of Union action in judicial cooperation in civil matters is thus formally expanded. However, given the broad interpretation so far given to Article 65, this terminological amendment is unlikely to add much impetus to any further expansion in practice. The removal of the necessity of a link with the internal market in turn has the advantage of clarifying the existence of legal basis for the Union competence, notably as regards to the adoption of measures in the field of family law.Footnote 53 In contrast with Article 65, Article 81 also contains a longer but fixed list of areas for potential action. Again, this particular amendment will not in practice lead to a great extension of the scope of Union action: some of these measures (such as measures for the development of ADR or effective access to justice), had already been taken under Article 65.Footnote 54
Article 81 maintains the requirement of cross-border implications, and as such, an important limitation to the legal basis of Union competence in private international law.Footnote 55 This is particularly significant given the recent (failed) attempts made by the Commission to legislate on the basis of Article 65 in domestic as well as cross-border cases.Footnote 56 Nevertheless, the TFEU does not elaborate on what constitutes ‘cross border implications’; definitional difficulties will therefore remain: while it may seem logical that the understanding of the cross-border requirement may vary depending on the particular aim of the measure considered, the lack of uniform definition may raise important difficulties in the practical application of related instruments.
2. Role of the European Court of Justice
As indicated above, the Treaty of Amsterdam, in Title IV, curtailed the broad preliminary reference jurisdiction afforded to the ECJ by Article 234 EC Treaty.Footnote 57 From the perspective of private international law this change could be seen as particularly retrograde given that the uniform interpretation of the Brussels I Regulation risked being less extensive than that of its ‘predecessor’, the 1968 Convention. Under the protocol to the Convention appellate courts had the possibility to request a preliminary ruling.Footnote 58 By contrast only courts of final appeal can currently avail themselves of this possibility in relation to the Brussels I Regulation.Footnote 59 Such a restriction could appear particularly inappropriate as experience shows that references from lower courts have led to very important ECJ rulings on the Brussels regime.Footnote 60 Interestingly, although Article 67(2) EC Treaty requires the Council, after May 2004, to take a decision with a view to adapting the jurisdiction of the Court, no initiative was taken on this basis by the Council.Footnote 61
Once it enters into force, the Treaty of Lisbon will end this situation: Article 68 is repealed and the whole area of Freedom, Security and Justice will fall within the scope of general jurisdiction of the ECJ under Article 267 TFEUFootnote 62 (which replaces Article 234 EC Treaty): preliminary references may be requested by any national court, even one of first instance,Footnote 63 if a ruling is necessary to enable judgment to be given, with courts of final appeal being under a duty to bring the matter before the Court. This is a significant change, which will improve the consistency of application and interpretation of European private international law rules without forcing litigants to ‘exhaust’ domestic remedies thereby wasting time and resources. The court's enhanced jurisdiction nevertheless has practical implications which will require action,Footnote 64 most notably the risk of overload.Footnote 65 The impact of these changes on the duration of the procedure may however be limited: As per Article 23aFootnote 66 of Protocol 3 on the Statute of the Court of Justice of the European Union recognises, the rules of procedure now provide for an urgent procedure for references for a preliminary ruling relating to the area of freedom, security and justice.Footnote 67
D. Geographical scope of the measures adopted—special position of the UK, Ireland and Denmark
The realisation by the Treaty of Amsterdam of the area of Freedom, Security and Justice has come at the price of a ‘protocolisation’. With respect to judicial cooperation in civil matters, this was in the form of Protocol (No 4) on the position of the United Kingdom and Ireland (1997), as well as Protocol (No 5) on the position of Denmark.Footnote 68
The arrangements negotiated by the UK and Ireland are flexible, allowing both countries to control the level of their involvement in Title IV measures. According to Arts 1 & 2 of the Protocol No 4, the UK and Ireland do not, in principle, take part in the adoption of, and are not bound by, Title IV measures. However, Article 3 allows both countries, within 3 months from the presentation of a specific proposal, to decide to participate,Footnote 69 and Article 4 specifies that both countries also have the opportunity to decide to accept Title IV measures after these have been adopted.
In the new protocol ‘on the position of the UK and Ireland in respect of the area of freedom, security and justice’ (Protocol No 21) the flexible mechanism put in place by the initial agreement is maintained,Footnote 70 and clarifications are introduced to facilitate its implementation. However a number of issues remain uncertain, notably in the light of recent UK experience.
What impact should the decision not to opt-in have on the participation in negotiations? Protocol No 4 does not expressly indicate that the decision not to opt-in is to be accompanied by the loss of the ability to be involved in the elaboration of the measure. Practice has shown that refusing to opt-in does not preclude participation in the shaping of a measure.Footnote 71 The Treaty of Lisbon does not, however, formalise this possibility. Could the UK or Ireland opt out after initially opting in to a measure? Given that this is not expressly permitted by the protocols (whether in their Amsterdam or Lisbon version), the answer has to be in the negative. Could the UK or Ireland be seen as bound by their initial decision to take part in the event of a dramatic change of substance of the planned measure during the negotiations? Again, the Treaty of Lisbon remains silent on this particular point. But it is to be noted that there is one possible way out in this case: in both versions of the protocol, the Council can, under Article 3(2), decide after a reasonable period of time to exclude the UK or Ireland if either opposes the adoption of a measure (to which they had initially opted in) to the extent that a blocking majority is achieved.
The Treaty of Lisbon does remove an important uncertainty concerning the possibility for the UK or Ireland to refuse to opt-in to a measure amending a regulation it had previously accepted. This difficulty had first appeared in the context of the Rome III dossier, where the silence of the Amsterdam version of the protocol proved particularly problematic.Footnote 72 Article 4a of the new protocol now clarifies that the UK & Ireland keep the right to decide not to opt-in to a measure amending an existing measure by which they are bound. But, where the non-participation of the UK or Ireland in the amended version of an existing measure makes the application of that measure ‘inoperable’ for other Member States or the Union, the CouncilFootnote 73 may urge UK or Ireland to make a notification that they wish to take part. If they maintain their position, not only will the existing measure cease to apply to them but they may have to bear the financial consequences incurred as a result of the cessation of their participation in the existing measure.
The introduction of such potent sanctions is obviously likely to change the negotiating dynamics each time ‘amending’ initiatives are put forward after the entry into force or the Treaty of Lisbon. Further as the mass of the European private international law acquis expands so will the likelihood of amending proposals, and so too will the pressure to participate. However the threat of exclusion from the operation of an existing measure is dependent on the existing measure becoming ‘inoperable’ as a result of the non-participation. This choice of words implies a very high threshold which will probably not often be reached in practice in the field of private international law.Footnote 74 In addition, as any other Council decisions, the determination of inoperability by the Council could be challenged before the ECJ.Footnote 75
Unlike the arrangements made under Protocol No 4 on the Position of the UK and Ireland, the position secured by Denmark in Protocol No 5 was inflexible.Footnote 76 However, individual agreements were concluded with a view to extending the provisions of two Title IV Regulations to Denmark.Footnote 77 Protocol (No 22) maintains the status quo, but an annex is added, which essentially reproduces the terms of Protocol No 21 on the position of the UK and Ireland.Footnote 78 Whilst the annex will not be applicable until and unless Denmark decides to substitute it for a part in the original Protocol, it is the State's declaredFootnote 79 intention to avail itself of the option to participate in the adoption of Title V of the TFE. Consequently the blanket opt-out in this area might soon come to an end.Footnote 80
E. External competenceFootnote 81
The treaties, until the Lisbon agreement, did not contain a provision empowering the Community to conclude international conventions with third States. In the field of private international law the issue of the external competence started being debated with the Treaty of Amsterdam. Given the well established principle foro interno-foro externo,Footnote 82 the question raised by the communitarisation of private international law was not whether the Community had external competence in this field, but revolved around the exact material scope of this power and whether this competence was exclusive or shared with Member States.Footnote 83 The difficulty was to reconcile potentially antagonistic needs: the importance to strengthen the role of the Community on the international plane—Community action being deemed more efficient than isolated Member States' action—and the special interest Member States may have to enter into agreements with third parties.
The renegotiation of the Lugano Convention gave the ECJ an occasionFootnote 84 to clarify how the principles it had developed since the ERTA judgment applied in the particular context of private international law. Although the policy underlying, and the implications of, this complex opinion gave rise to numerous questions, some important points were made by the Court. First a Community competence may implicitly flow from provisions of the Treaty and from measures adopted, within the framework of those provisions, by the Community institutions. Additionally ‘whenever Community law created for those institutions powers within its internal system for the purpose of attaining a specific objective, the Community has authority to undertake international commitments necessary for the attainment of that objective even in the absence of an express provision to that effect’.Footnote 85 This competence is exclusive where ‘the conclusion of the international agreement is necessary in order to attain objectives of the Treaty that cannot be attained by establishing autonomous rules’Footnote 86 and where the instrument would entail obligations which would affect already adopted common rules.Footnote 87 A case-by-case assessment is necessary to find out if the Community has competence to conclude an international instrument and if that competence is exclusive. ‘In doing so, account must be taken not only of the area covered by the Community rules and by the provisions of the agreement envisaged,Footnote 88 insofar as the latter are known, but also of the nature and content of those rules and those provisions, to ensure that the agreement is not capable of underminingFootnote 89 the uniform and consistent application of the Community rules and the proper functioning of the system which they establish’.Footnote 90
The Treaty of Lisbon partly codifies the ECJ rulings and opinions on external competence. Article 216 TFEUFootnote 91 provides that the Union has competence to conclude an international agreement where the treaties so provide, where the instrument is necessary in order to achieve an objective (either referred to in the treaties or provided for in a binding Union act), or, is ‘likely to affect common rules or alter their scope’. Although framed in terms remarkably similar to those used in the context of exclusive competence by the ECJ, Article 216 does not spell out if the external competence shall be exclusive or shared. This provision however is to be read in conjunction with a new Article inserted by the Treaty of Lisbon in the TFEU, Article 3(2), which specifies that the Union competence for the conclusion of international agreements shall be exclusive when the conclusion of the agreement is ‘provided for in a legislative act of the Union, or is necessary to enable the Union to exercise its internal competence’. In addition, the Union has exclusive competence for the conclusion of an international agreement ‘in so far as its conclusion may affect common rules or alter their scope’. While Article 3 gives a much needed precision that was lacking in Article 216, the similar but not identical wording of the two provisions renders their respective interpretation at times complex. The combined effect of the provisions makes it clear (unsurprisingly) that the Treaties may (expressly) provide for shared external competence in certain areas. It is similarly clear that where the conclusion of the international agreement is provided for in a European legislative act, the Union's competence not only exists but is to be exclusive. The concurrent reading of both provisions further seems to imply that external competence is only shared where the conclusion of an international agreement, although necessary in order to achieve one of the treaty objectives, is not actually necessary to enable the Union to exercise its internal competence. Was this differentiation really intended by the Member States?Footnote 92 And while Article 216 provides that external Union competence exists where the conclusion of an agreement ‘is likely’ to affect common rules and alter their scope, Article 3(2) provides that the external competence of the Union is exclusive ‘in so far as its conclusion may’ affect common rules and alter their scope. Could the use of the terms ‘in so far as’ in Article 3 manifest the idea that the exclusivity of the Union competence is restricted to ‘concrete’ situations where the common rules already exist or are at least foreseeable, the external competence being only shared in situations where the Union has internal competence but where no common rules have yet been adopted or precisely planned?
Further the reference in both provisions to ‘common’ rules raises difficulty in the context of measures which, as a result of the application of Protocols No 21 & 22, are not binding on certain Member States,Footnote 93 or which are adopted under the enhanced cooperation framework.Footnote 94 If the Maintenance Regulation were to be adopted with the UK maintaining its decision not to opt-in, would the rules of the Regulation be considered ‘common’ for the purpose of Arts 3 & 216 TFEU? If so, would an international agreement concluded by the Union and altering the scope of these rules be binding on the UK? Given the rationale of the implied external competence in this case (the importance of preventing an international instrument from undermining the uniform and consistent application of the Community rules), it seems logical to consider that rules binding, say 26 States, are common for the purpose of Arts 216 & 3. None of these 26 States should be allowed to enter into agreements with third States that would affect the common rules: competence in this matter should exclusively belong to the Union. But this does not necessarily mean that the non-participating State itself should be bound by the instrument concluded between the Union and third States. In light of the terms of Article 2 of Protocols No 21 & 22,Footnote 95 one is led to conclude that the non participating State would not be bound by the agreement. The same conclusion should mutatis mutandis apply in the context of enhanced cooperation.Footnote 96
F. Conclusions
Closing a review of the impact of the Treaty of Amsterdam on private international law, a strong proponent of European integration in this field wrote in 2000:
‘A great step forward—private international law recognised as Community task—has been achieved but at a high price on the technical level. Thus on balance it has been a mixed success. One hopes that the next Revision Conference will rectify this major mistake.’Footnote 97
This wishFootnote 98 was only partially acted upon by the Lisbon Treaty. Once reformed, the Treaties will provide a slightly wider and more (but not entirely) clearly defined legal basis for Union action in the European internal sphere. However, while Article 293 has been deleted, the specific competence afforded to the Union to legislate in the field of conflict of laws will continue to coexist with provisions allowing it to enact sporadic private international law provisions in a way detrimental to the overall coherence and transparency of the field. More importantly, the elements of variable geometry which have been introduced by the Treaty of Amsterdam will remain, and continue, despite some changes, to raise many questions. Similarly, while the insertion in the reformed Treaties of provisions on categories and areas of competences will in some ways improves the legibility of the definition of external competences relevant to private international law, it will introduce new uncertainties. Technical perfection, it seems, will have to wait for yet another Revision Conference.