Jump to Content Jump to Main Navigation
Signed in as:
Oxford Law Citator
Sign up for alerts

Part IV The Law of Obligations, 19 Contracts

Paul Torremans

From: Cheshire, North & Fawcett: Private International Law (15th Edition)

Uglješa Grušić, Christian Heinze, Louise Merrett, Alex Mills, Carmen Otero García-Castrillón, Zheng Sophia Tang, Katarina Trimmings, Lara Walker
Edited By: Paul Torremans, James J. Fawcett

Content type:
Book content
Product:
Private International Law [PRIL]
Published in print:
28 September 2017
ISBN:
9780199678983

From: Oxford Legal Research Library (http://olrl.ouplaw.com). (c) Oxford University Press, 2023. All Rights Reserved. Subscriber: null; date: 06 June 2023

Subject(s):
Contract — Rome Convention — Rome I Regulation and choice of law — Recognition and enforcement of foreign judgments – traditional rules

(p. 681) 19  Contracts

1.  Introduction

(a)  The nature of the problem

The problem of ascertaining the applicable law is more perplexing in the case of contracts than in almost any other area. There are three reasons for this. First, there is the diversity of connecting factors that can be raised by the facts of the case: the place where the contract is made; the place of performance; the domicile, nationality or place of business of the parties; the situation of the subject matter and so on. In most areas of private international law the decisive connecting factor on which ascertainment of the applicable law depends is reasonably clear. There is general agreement, for instance, that it is the place of celebration which indicates the law to govern the formal validity of a marriage. But with contracts the sheer multiplicity of connecting factors makes it hard to identify one single connecting factor as the determinant of the applicable law. Secondly, contracts are planned transactions and the parties may well have considered the question of what law should govern the contract in the event of a dispute arising between them. They may have made provision in the contract, choosing the applicable law. Thirdly, a wide variety of different contractual issues can arise. For example, there can be a problem over whether a contract has been validly created, concerning how it should be interpreted, about whether it has been discharged. This raises the question whether the same law should govern all of these issues. Moreover, there are many different types of contract. A sale of goods contract has different features from an insurance contract or a contract for carriage of goods by sea. Should the same law govern regardless of the type of contract involved, or do the special features of particular contracts necessitate special choice of law rules?

(b)  Various solutions to the problem1

As one would expect from the complex nature of the problem, a wide variety of different solutions have been tried in different countries over the years. In the USA a preference was (p. 682) formerly shown for a rigid and inflexible test, represented by the place of contracting in some of the states but by the place of performance in others. However, the choice of law revolution in that country2 has affected not only tort cases but also contract cases, and a wide range of modern approaches is now used in this area.3 Most of the countries of the European Continent rejected a rigid test and, instead, adopted the doctrine of autonomy under which the parties were free to choose the governing law, though divergent views obtained on the question whether their freedom was absolute or was restricted to the choice of a law with which the contract was factually connected.4 In the absence of choice by the parties, most of these countries adopted a flexible approach, leaving the judge to select the decisive connecting factors from the various elements of the contract and the circumstances of the case.5

English law formerly applied the “proper law of the contract”,6 which was a succinct expression to describe the law governing many of the matters affecting a contract. The doctrine of the proper law was of common law origin and a vast case law developed to take account of the difficulties outlined above. It was both sophisticated and flexible in its approach. The key features of the doctrine were as follows. The parties could choose the proper law, with very little restriction on this right. If the parties did not express a choice, and one could not be inferred by the courts, an objective test was applied. This sought to localise the contract by looking for the system of law with which the transaction was most closely connected. The twin theories which underlay the proper law were therefore the subjective theory, which looked to the intentions of the parties, and the objective theory, which sought to localise the contract. Special rules were adopted for particular issues. The proper law was usually relevant, but the court was required to go beyond the proper law when considering certain issues. Thus, for example, with the issue of illegality the courts were concerned not only with illegality by the proper law but also with illegality by the law of the place of performance. There were also special rules for particular contracts, such as insurance contracts. These rules either made special provision for ascertaining the proper law or departed from the proper law altogether.

Choice of law in contract was later put on a statutory footing. The Contracts (Applicable Law) Act 1990 largely replaced the common law rules and the doctrine of the proper law of the contract. The Act implemented the EEC Convention on the law applicable to contractual obligations of 1980 (the Rome Convention).7 For contracts concluded since 17 December 2009,8 the Rome Convention has been replaced by the Rome I Regulation,9 and any reference (p. 683) to the Convention shall be understood as a reference to the Regulation.10 However, in relation to Denmark and those territories of the Member States to which the Rome I Regulation does not apply, the Rome Convention remains in force.11 The Regulation builds on the acquis of the Rome Convention, with some important modernisations and amendments. The future of the Rome I Regulation after Brexit—as that of all other EU regulations—is uncertain. However, as lawyers and courts in England and elsewhere have grown familiar with the Rome Regulations, which also served as a model for law reform outside the EU, there are good reasons to preserve these Regulations as part of domestic UK law in order to enhance legal certainty after Brexit.12

2.  The Rome Convention13

(a)  The history and purpose of the Convention

As early as 1967 there was a proposal from the governments of the Benelux countries to the Commission of the European Communities for the unification of private international law rules, particularly in the field of contract law. Experts from the then six Member States of the Community prepared a preliminary draft Convention before the United Kingdom, Ireland and Denmark joined the European Community. At that time, this draft also covered non-contractual obligations. In negotiations between experts from the then nine Member States of the European Community there was extensive revision of the draft Convention, culminating in a final Convention (the Rome Convention) in 1980.14 This was only concerned with contractual obligations. By the end of 1981 the Convention had been signed by all of the then Member States.15 However, ratification of the Rome Convention was delayed whilst (p. 684) problems in relation to what powers, if any, the European Court of Justice should have as regards interpretation of the Convention were resolved. In 1988, two protocols on interpretation of the Convention were signed,16 which did not enter into force until 2004.17 The Convention eventually received the requisite number of ratifications and came into force on 1 April 1991.18

In general terms, the Convention has been seen as a continuation of the work on unification begun by the Brussels Convention on jurisdiction and the enforcement of judgments in civil and commercial matters (replaced by the Brussels I Regulation and now the Brussels I Recast).19 More particularly, since the law would be the same wherever trial takes place in the Community, it inhibited the forum shopping that the Brussels Convention (now Brussels I Recast) allowed.20 It was also said that such a Convention would increase legal certainty and make it easier to anticipate the law to be applied.21

(b)  The Contracts (Applicable Law) Act 1990

The introduction of the Rome Convention into English law was a matter of considerable controversy with enthusiasts22 and critics23 taking polarised positions. The controversy ranged (p. 685) over both the need for harmonisation of contract choice of law rules and the nature and form of the new law. Nonetheless, when the 1990 Act was presented to Parliament it was said24 that the Convention would produce benefits in terms both of harmonisation and improved certainty in the law.

The Contracts (Applicable Law) Act 199025 provided that the Rome Convention and the different Accession Conventions plus the first Protocol shall have the force of law in the United Kingdom.26 There was a provision dealing with interpretation of the Conventions and Protocol,27 and the Act reserved the right not to apply Articles 7(1) (mandatory rules of foreign countries) and 10(1)(e) (the consequences of nullity of a contract) of the Rome Convention.28 Moreover, for intra-United Kingdom disputes, for which the Convention did not apply,29 section 2(3) provided that “the Convention shall apply in the case of conflicts between the laws of different parts of the United Kingdom”.

The effect of implementation of the Rome Convention was that, for contracts made after the Convention came into force, the traditional common law rules on contract choice of law were largely replaced by the rules contained in the Convention.30 It was not possible for the parties to contract out of the Convention, for this would have defeated its purpose.31 Nevertheless, the traditional common law rules continued to be applied to contracts made after the Convention came into force in situations which fell outside the (substantive) scope of the Convention;32 and the same holds true for situations which fall outside the scope of the Rome I Regulation.33 English courts therefore still have to operate two different regimes for contract choice of law: for Convention (now Regulation) cases (ie cases coming within the scope of the Convention or Regulation) there are the Convention/Regulation rules; for non-Convention cases (ie cases outside the scope of the Convention, now Regulation) there are the traditional common law rules.34 This undoubted complication in the law could have been avoided if the 1990 Act had provided that the rules in the Rome Convention were to be applied to all contracts made after the Convention came into force, even in non-Convention cases, thereby assimilating the two sets of rules.35

(p. 686) 3.  The Rome I Regulation36

(a)  Preliminary remarks

(i)  History

After a consultation process started by a Green Paper in 2002,37 the European Commission put forward in 2005 a proposal that the Rome Convention should be converted to the Rome I Regulation and its provisions modernised.38 Harmonisation of choice of law rules was regarded as helping to facilitate the mutual recognition of judgments,39 and the proper functioning of the internal market was said to create a need for such harmonisation.40 For the same reason there was said to be a need to achieve harmony between three key instruments: the Brussels I Regulation (now Brussels I Recast) on jurisdiction; the Rome I Regulation on contractual obligations and the Rome II Regulation on non-contractual obligations.41 The legal basis for the Regulation was Title IV, in particular Article 61(c), of the Treaty on European Union (now Title V, in particular Article 81(2)(c) of the Treaty on the Functioning of the European Union [TFEU]), which authorised the adoption of measures aimed at ensuring the compatibility of the rules applicable in Member States concerning the conflict of laws. The final version of the Regulation is based on a proposal contained in a European Parliament legislative resolution of 29 November 2007.42 Prior to this there were informal contacts between the Council, the European Parliament and the Commission with a view to avoiding the need for a second reading and conciliation. The Council formally adopted the Regulation on 6 June 2008.43 The Regulation applies to contracts concluded as from 17 December 2009.44 Thus, contracts concluded before that date continue to be governed by the Rome Convention, even if the parties agree after 17 December 2009 on a (minor) variation of this contract. The Rome I Regulation applies only to a contract concluded before 17 December (p. 687) 2009 where this contract is subject, on or after that date, to a variation of such magnitude that it gives rise not to the mere updating or amendment of the contract but to the creation of a new legal relationship between the contracting parties, so that the initial contract should be regarded as having been replaced by a new contract.45

The United Kingdom, after stakeholder consultations, elected initially in 2006 not to opt in to the proposed Rome I Regulation.46 Among the critics of the proposed Regulation47 was the Financial Markets Law Committee, which took the view that it would cause significant uncertainty in the financial markets.48 The prospect of having to consider giving effect to overriding mandatory provisions of foreign countries generated the greatest concern in this regard, although other provisions also caused concern. Nonetheless, the United Kingdom government saw clear benefits in a coherent European Union choice of law regime and attended the subsequent negotiations, without voting rights. After the negotiations on Rome I were complete, the United Kingdom has conducted a full public consultation before deciding whether to opt in or not. In its Consultation Paper,49 the Ministry of Justice has taken the view that the Articles that were of greatest concern to the United Kingdom stakeholders during negotiations have either been removed, substantially revised or returned to their Convention form subject to later review. The Ministry of Justice has also argued that the Regulation improves upon the Convention in a number of respects, in particular in terms of improved drafting. All of this has led to its conclusion that the United Kingdom should opt in to the Rome I Regulation. In July 2008 the United Kingdom government expressed the wish to opt in to the Rome I Regulation and sought the consent of the Council of the European Union and of the European Commission to the United Kingdom’s participation which was duly granted.50

(ii)  Interpretation51

(a)  Referrals to the Court of Justice

Under Article 267 of the Treaty on Functioning of the European Union (TFEU), the Court of Justice may give preliminary rulings on the validity and interpretation of acts of the institutions of the Union, including Regulations and thus the Rome I Regulation. National courts normally have discretion whether or not to request a preliminary ruling on interpretation from the Court of Justice.52 However, where the question of interpretation is raised in a case pending before a national court against whose decisions there is no judicial remedy under national law and the question cannot be regarded as acte clair,53 that court or tribunal must bring the matter before the Court of Justice.54

(p. 688) (b)  The principles and decisions laid down by the Court of Justice55

Where the meaning of the Regulation is not referred to the Court of Justice, it should be determined in accordance with the principles laid down by, and any relevant decision of, that Court.56 This means that the English courts have to act in accordance with two different types of authority: first, any relevant decisions of the Court of Justice; secondly, the principles laid down by the Court of Justice. If the Court of Justice has previously given a decision on the provision in issue, this must be followed. A relevant decision for these purposes could include one of the decisions of the Court of Justice discussing the Rome Convention57 in a jurisdiction case on the Brussels Convention or the Brussels I Regulation. However, the provision which is in issue may not have been previously discussed by the Court of Justice. In this situation, the English courts must act in accordance with the principles of interpretation laid down by the Court of Justice. In this respect, the Court is likely to apply the same general principles of interpretation to the Rome I Regulation as it applies to other areas of EU law. Thus, in particular three factors should be considered for interpretation: first, the wording of the provision, secondly, the systematic context in which the provision is found, including its relation to other provisions in the Regulation or in other EU instruments, in particular the Brussels I Regulation (now Recast) and the Rome II Regulation,58 and thirdly, as a very important factor,59 the objectives of the provision in question and the objectives and scheme of the overall Regulation.60 These objectives are, in particular, the predictability of the outcome of litigation, legal certainty and foreseeability as to the law applicable and the uniform application in all Member States.61 In ascertaining the meaning of concepts used in the Regulation, regard should be had to the meaning of cognate concepts to be found in the European Union Treaties or in secondary legislation (in particular in neighbouring instruments such as Brussels I or Rome II), unless the European Union legislature has, in a specific legislative context, expressed a different intention.62 It is encouraging to note that English judges when construing the Rome Convention have adopted a purposive approach63 and have given the scope of the Convention an autonomous meaning.64

(p. 689) (c)  The principle of uniform and autonomous interpretation

One principle that the Court of Justice and national courts are bound to follow is the principle of uniform interpretation, which was set out in the Rome Convention itself. Article 1865 provided that “In the interpretation and application of the preceding uniform rules, regard shall be had to their international character and to the desirability of achieving uniformity in their interpretation and application.”66 The same concept applies also to the Rome I Regulation: It follows from the need for uniform application of EU law and from the principle of equality that the terms used in the Regulation which make no express reference to the law of the Member States must normally be given an autonomous (independent) and uniform meaning by reference to their wording, scheme and purpose, rather than being understood as a reference to national law.67 This means that courts should not define concepts by reference to their national law, but instead give independent EU meanings to the terms used in the Convention and now the Regulation. This also has important consequences when it comes to aids to interpretation, which will now be considered.

(d)  Aids to interpretation68

(i)  The Giuliano and Lagarde Report69 and other legislative materials

The Rome Convention was accompanied by the Giuliano and Lagarde Report, which is a commentary by members of the Working Group responsible for drafting the Convention. Section 3(3) of the Contracts (Applicable Law) Act 1990 allowed this Report, and the Tizzano Report on the 1988 Protocols on interpretation,70 to be considered by the English courts in ascertaining the meaning or effect of any provisions in the Rome Convention or first Protocol on interpretation. This follows the pattern of the Civil Jurisdiction and Judgments Act 1982, which allowed the Jenard and Schlosser Reports to be considered by English courts when interpreting the EU law on jurisdiction and enforcement of judgments.71 The latter Reports have been constantly referred to by the Court of Justice and national courts when interpreting the Brussels Convention and also the Brussels Regulation (where the Regulation builds on the Convention). As many of the provisions of the Rome I Regulation are derived from the Rome Convention, the Giuliano and Lagarde Report is likely to be treated as being of the same high authority when it comes to interpretation of the Rome I Regulation.72 Also the other legislative materials of the Rome I Regulation73 should be consulted, even if their contents are not strictly binding for the interpretation of the Regulation.74 In particular the Explanatory Memorandum contained in the Commission Proposal of 200575 may provide helpful insights in interpreting the Rome I Regulation.

(p. 690) (ii)  The decisions of other European courts

Whatever the normal attitude of English judges towards the decisions of Continental courts, in this context foreign decisions on interpretation of the Rome Convention and the Rome I Regulation are of persuasive authority. Article 18 Rome Convention (uniformity of interpretation) enabled parties to rely on foreign decisions.76 This included the decisions of Continental courts in Member States which have applied the Convention in anticipation of its coming into effect, as well as decisions subsequent to this. The same holds true under the Rome I Regulation, where national courts should strive for an autonomous EU interpretation and thus take the views of other EU courts into consideration.77

(iii)  Texts in other languages

The 1990 Act gave the force of law78 to the Convention in its different language texts, all of which are equally authentic,79 rather than just to the English text, which is merely set out in the Act for ease of reference. Moreover, there is Court of Justice authority to the effect that courts of Member States should always be prepared to consider the texts of EU legislation in other languages.80 As a consequence, the Rome I Regulation has to be interpreted also in the light of the versions existing in the other official languages. Any divergences between the different languages are to be resolved by reference to the purpose and general scheme of the Regulation.

(iv)  The Brussels I and Rome II Regulations

It has already been mentioned81 that English courts should follow decisions on interpretation of the Brussels Convention and the subsequent Brussels I Regulation which are relevant to interpretation of the Rome Convention82 and the Rome I Regulation.83 Going beyond this, some of the provisions in the Rome I Regulation use the same concepts as,84 or are even lifted word for word from,85 the Brussels Convention (now the Brussels I Recast). In such a situation, it is only right and proper that the earlier interpretation of this concept or term under the Brussels Convention and recast Regulation should be looked at, and, unless there is a good reason to the contrary, followed.86 More generally, Recital (7) Rome I states that “[t]he substantive scope and the provisions of this Regulation should be consistent” with the Brussels I Regulation and the Rome II Regulation.87 As a consequence, it seems safe to assume that the interpretation of parallel terms and concepts (p. 691) under these Regulations is to be applied also to the Rome I Regulation, unless there are cogent reasons not to do so.

(v)  The traditional common law rules

Much of the Convention appears familiar to English lawyers and there may be a temptation to resort to the old common law rules when interpreting the Convention88 or the Rome I Regulation. However, this is not usually justified and would be a dangerous habit to get into.89 The provisions in the Convention and Regulation are not normally based on English law or on that of any other country’s national law but on a common core of ideas used in EU countries.90 A rule which may appear at first sight to be the same as the common law rule it has replaced may turn out, on closer examination, to be different in some respect.91 The aim of uniformity of interpretation throughout the EU will not be achieved if English courts interpret the Convention or Regulation as a codification of the proper law of the contract. Article 18 (uniform interpretation) served as a reminder that national courts should not act in this way;92 and the same holds true under the Rome I Regulation as a consequence of the principle of uniform and autonomous interpretation of EU law.93

(b)  When does the Regulation apply?

The Regulation applies to matters coming within its scope, and it has universal application, ie it applies equally to contracts having no connection with a European Union State and to contracts with such a connection. Before turning to examine in detail these two aspects of the application of the Regulation, four general points need to be made. First, neither the Rome Convention nor the Rome I Regulation have retrospective effect.94 The Rome Convention only applies in a Contracting State to contracts made after the Convention has entered into force in that state (1 April 1991 in the case of the United Kingdom), and the Rome I Regulation applies only to contracts concluded as from 17 December 2009; the traditional common law rules will continue to apply to contracts made before the Convention has entered into force. Secondly, neither the Rome Convention nor the Rome I Regulation prejudice the application of other international conventions to which a Contracting or Member State is a party.95 This means, for example, that, as far as the United Kingdom is concerned, carriage of goods by sea will still be dealt with by the Hague-Visby Rules, implemented by the Carriage of Goods by Sea Act 1971, and not by the Rome Convention or Rome I Regulation.96 Thirdly, acts of the institutions of the European Union, namely Regulations and Directives, and national laws implementing such acts, laying down choice of law rules relating to contractual obligations in relation to particular matters, take precedence (p. 692) over both the Convention and the Regulation.97 Fourthly, the possibility under the Rome Convention for Contracting States to unilaterally introduce choice of law rules inconsistent with those contained in the Convention98 has been abolished under the Rome I Regulation.

(i)  The scope of the Regulation

(a)  Contractual obligations in situations involving a conflict of laws

Article 1(1) Rome I states that: “This Regulation shall apply, in situations involving a conflict of laws, to contractual obligations in civil and commercial matters.”99 There are three separate requirements under this provision. First, the obligation must be contractual. Secondly, there must be a conflict of law problem. And thirdly, the obligation must concern a civil and commercial matter and not, in particular, a revenue, customs or administrative matter. While the meaning of the third requirement is identical to the parallel exclusion under the Brussels I Recast to which may be referred here,100 the other two requirements deserve closer consideration.

(i)  A contractual obligation

This concept should be given a European autonomous meaning that is not blinkered by national conceptions such as consideration101 or even privity.102 As the Rome I Regulation does not contain a definition of a “contractual obligation”, but requires consistency with the Brussels I Recast and Rome II Regulation,103 the case law of the Court of Justice on Article 7(1) Brussels I Recast can give guidance to define the “contractual obligation” and to draw the line between contractual and non-contractual obligations.104 Drawing from an analogy to the meaning of “matter relating to a contract” in Article 7(1) Brussels I Regulation, the Court has held that the concept of “contractual obligation” within the meaning of Article 1 Rome I Regulation “designates a legal obligation freely consented to by one person towards another”.105 Recently, the Court of Justice added that the mere fact that one contracting party brings a civil liability claim against the other is not sufficient to consider that the claim concerns “matters relating to a contract” within the meaning of Article 7(1) Brussels I Recast. Rather, that is the case only where the “conduct complained of may be considered a breach of contract”, which will be the case “where the interpretation of the contract which links the defendant to the applicant is indispensable to establish the lawful or, on the contrary, (p. 693) unlawful nature of the conduct complained of against the former by the latter”.106 A non-contractual obligation, on the other hand, can be defined as a claim which seeks to establish the liability of a defendant and which is not related to a “contract” within the meaning of Article 7(1) Brussels I Recast.107 For the purposes of the Rome II Regulation, such a non-contractual obligation “must be understood as meaning an obligation which derives from one of the events listed in Article 2 Rome II Regulation (tort/delict, unjust enrichment, negotiorum gestio or culpa in contrahendo)”.108

Moreover, the process of characterisation can be guided by the subject matter and wording of the Rome I Regulation itself. A provision in the Regulation, such as Article 12 on the scope of the law applicable or Article 14 on voluntary assignment, may show a clear intention to embrace a particular issue.109 Applying these principles it has been held110 that the issue of whether, following an assignment (of the benefit of an insurance policy), the obligor (the insurer) had to pay (the proceeds of the insurance policy) to the assignee (a bank) rather than the assignor (a vessel owner) fell within the contractual rather than the proprietary umbrella.111 Also falling within this contractual umbrella are such issues as whether a contract has been novated, and whether a third party may enforce a right conferred on him from the outset under a contract.112 Novation and making new contracts with third parties are instances of the parties’ freedom to contract and it is this party autonomy that is the dominant theme influencing the modern international view of contract.113 A contractual obligation “is by its very nature one which is voluntarily assumed by agreement”.114 Where, on the other hand, a tortious or equitable duty of care is imposed entirely independent from the existence of a contract, this arises not from agreement and therefore the Regulation will not apply.115 Not only do tortious obligations and property rights116 fall outside the scope of the Regulation but so also do intellectual property rights117 and claims that arise as a matter of (p. 694) company law and company regulation.118 The same holds true for obligations directly linked to dealings prior to the conclusion of a contract (culpa in contrahendo; eg violation of the duty of disclosure and the breakdown of contractual negotiations) which fall in the scope of the Rome II Regulation,119 unless a tacit contractual relationship existed between the parties.120

The position in relation to quasi-contract, ie the consequences of an invalid contract, and other restitutionary remedies is more complicated. Both the Rome Convention and the Rome I Regulation contain a provision dealing with “the consequences of nullity of the contract”,121 which under English law is regarded as being a quasi-contractual issue.122 However, the Convention allowed a reservation not to apply this provision of which the United Kingdom made use.123 Under the Rome I Regulation, this reservation has been abolished. Thus, the consequences of nullity of a contract and any (restitutionary) remedies arising from this situation are to be classified as a contractual matter to be governed by the Rome I Regulation. For other restitutionary remedies, the Rome II Regulation on the law applicable to non-contractual obligations establishes separate provisions for two different types of restitutionary claim, which supports a classification as non-contractual for these types of claims.124 One provision deals with non-contractual obligations arising out of unjust enrichment, the other with non-contractual obligations arising out of an act performed without due authority (negotiorum gestio). Moreover, in the context of jurisdiction, a claim in restitution has not been regarded as a matter relating to a contract.125 Issues in restitution should therefore be regarded as being outside the scope of the Rome I Regulation and falling in the scope of the Rome II Regulation, unless they concern the consequences of nullity of a contract. In practice, however, it will not make a huge difference whether to apply Rome I or Rome II: Where a non-contractual obligation arising out of unjust enrichment or negotiorum gestio concerns a relationship existing between the parties, such as a contract, that is closely connected with that unjust enrichment, it shall be governed by the law that governs that relationship.126 Therefore the law governing the contract will normally prevail both under Rome I and Rome II.

In cases where there is concurrent liability in contract and tort, the position prior to the introduction of the Rome II Regulation was very favourable to the claimant. As under English domestic law, the claimant was, for choice of law purposes, free to frame the action in tort, in contract, or both.127 The introduction of the Rome II Regulation means that this (p. 695) needs rethinking. The analogy should be drawn with jurisdiction under Article 7(1) and (2) of the Brussels I Regulation128 and each obligation should be classified as contractual or non-contractual but not both, so that there would be no question of the claimant being able to choose.129

(ii)  A conflict of laws130

A preliminary draft of the Rome Convention stated that it only applied “in situations of an international character”.131 This requirement was criticised132 for the definitional problems it created, and it was replaced by the more straightforward requirement that there be a situation involving a choice between the laws of different countries (Rome Convention), or, respectively, a conflict of laws (Rome I Regulation). As far as the United Kingdom is concerned this merely makes explicit what was implicit under the traditional common law rules on contract choice of law. However, the fact that this is now spelt out in statutory form means that some attention needs to be given to this point. Under English private international law a conflict of laws problem exists whenever the court is faced with a dispute that contains a foreign element.133 With a contractual dispute, typical examples of a foreign element are as follows: one of the parties to the contract is a foreign national or is habitually resident abroad; the contract is concluded abroad; the contract is to be performed by one of the parties abroad. In such cases the foreign country has a claim to have its law applied, and the uniform rules in the Regulation are intended to apply.

The position is more difficult if the court is faced with a dispute involving a foreign element, but in respect of what is an essentially domestic contract. This can arise in two different types of case. The first is where, for example, there is a purely German contract, which is the subject of trial in England, subsequent to the defendant having moved his business to England after concluding the contract. The situation involves a foreign element in that one of the parties now carries on his business in England. However, what is lacking is any relevant connection with a country other than Germany of the sort which would give that other country’s law a claim to be applied.134 Nonetheless, it is desirable that such cases come within the Regulation.135 The object of the Regulation of achieving harmonisation of choice of law rules in contract is most likely to be attained if the scope of the Regulation is given as wide an interpretation as possible. The above example should therefore be regarded as one involving a conflict between the laws of different countries. The second type of case is where there is, for example, a purely English contract, but the parties have agreed that French law shall govern the contract. It is implicit from the terms of Article 3(3)136 that the Regulation will apply in this situation.137 However, the Regulation will not apply if there is a purely English contract (p. 696) which merely incorporates French law by, for example, setting out verbatim a provision of French law as a term of the contract.138

There is another problem in relation to the requirement that there is “conflict of laws” which is less easily solved. Under English law, if foreign law is not pleaded or proved, the court gives a decision according to English law.139 The courts are free to apply this rule in relation to the Rome Regulation because matters of evidence and procedure are excluded from the scope of the Regulation.140 If the English court is going automatically to apply English law it is arguable that this is not a situation involving a conflict between the laws of different countries. However, the purpose of the Regulation is not going to be met if the English courts allow the parties to side-step the uniform rules contained therein by a simple omission to plead and prove foreign law. It would therefore be better if this sort of case was regarded as coming within the Regulation.141

Although it is not stated explicitly in Article 1(1) Rome I Regulation, the conflict must be between the laws of different countries,142 which includes the laws of non-EU Member States.143 A country is defined under the Regulation in the normal private international law sense as a territorial unit with its own rules of law, in this case relating to contractual obligations.144 A French court, for example, will have to apply English law, or Scottish, or Northern Irish law under the Regulation, even though the United Kingdom is a Member State in the sense of the Regulation. Similarly, an English court may have to apply, for example, Ontario or New South Wales law under the Regulation. Indeed, the Regulation can apply to an interstate dispute involving connections with the “countries” of California and New York, provided that trial takes place in an EU Member State. However, the Regulation makes it clear that it is for the United Kingdom to decide whether it wants to apply the rules in the Regulation to intra-United Kingdom disputes. It is certainly not bound to do so,145 but the obvious inconvenience of having a different regime for intra-United Kingdom contractual disputes from all other cases has led to the decision by the UK legislator to apply the Rome Convention and the Rome I Regulation to such disputes.146 The upshot is that England, Scotland and Northern Ireland are separate countries for the purposes of the Regulation, even in intra-United Kingdom disputes.

(p. 697) (b)  Exclusions147

Article 1(2) to (3) excludes a wide variety of matters from the scope of the Regulation. These matters can be put into three main categories. First and foremost, it excludes certain commercial contracts such as arbitration agreements and certain contracts of insurance. Secondly, it excludes non-commercial contracts, such as agreements to make wills and agreements to pay maintenance. Thirdly, it excludes certain matters which do not involve contract choice of law, such as evidence and procedure, or which under the laws of some Member States do not involve contract choice of law, such as negotiable instruments and the issue of capacity to contract. The matters excluded from the scope of the Regulation, and the reasons for their exclusion, will now be examined, in the order in which they are set out in the Regulation. These are as follows:

(i)  Questions involving the status or legal capacity of natural persons, without prejudice to Article 13148

This phrase is a familiar one, and is to be found in the list of exclusions from the Rome Convention and Brussels I Recast.149 Questions of status are clearly outside the scope of a Regulation concerned with contract choice of law, and do not need expressly to be excluded. The exclusion of legal capacity is more controversial. To common lawyers capacity to contract is a matter falling squarely within the ambit of rules on contract choice of law. But to civil lawyers this is regarded as a matter relating to status, hence its exclusion from the Regulation.150 This particular exclusion only relates to natural persons. The exclusion of the legal capacity of corporations is dealt with under a separate provision.151 The result of the exclusion is that national courts are left to apply their traditional rules of private international law to the issue of capacity to contract; in England’s case this will be the traditional common law rules. However, there is one exception to this. The exclusion of capacity to contract is subject to Article 13 of the Regulation, which is a fairly narrow rule designed to protect a party who contracts with a natural person under an incapacity from being caught unawares by this. The English common law rules on capacity to contract, and Article 13, will be examined later on in this chapter in the section on particular issues.152

(ii)  Obligations arising out of family or succession law

Article 1(2)(b) and (c), which adopt the wording of the Rome II Regulation on non-contractual obligations,153 are concerned with non-commercial contracts. They exclude

obligations arising out of family relationships and relationships deemed by the law applicable to such relationships to have comparable effects, including maintenance obligations, [and]

obligations arising out of matrimonial property regimes, property regimes of relationships deemed by the law applicable to such relationships to have comparable effects to marriage, and wills and succession.154

(p. 698) Indeed, many disputes relating to the matters listed above will not even involve contractual obligations. For example, disputes in relation to succession are not normally contractual, but are concerned with issues such as the validity of the will, or concern obligations imposed by law and not by agreement. This provision makes it clear that, in the rare cases which raise contractual obligations, for example an agreement to make a will, the Regulation will not apply.

Family relationships in the sense of Article 1(2)(b) shall include “parentage, marriage, affinity and collateral relatives”.155 Under the Rome Convention, the comparable exclusion of “rights and duties arising out of a family relationship” was intended to ensure that contractual obligations relating to any family law matter were excluded from the Convention.156 In particular, it was intended to exclude maintenance obligations. However, the exclusion of maintenance was not all-embracing. The Giuliano and Lagarde Report157 distinguished between, on the one hand, obligations to pay maintenance which are imposed by law in respect of which there is also an agreement to pay (these were excluded from the scope of the Convention) and, on the other hand, purely contractual obligations to do so (these were within the scope of the Convention). This distinction can also be applied under the Regulation. Thus the case of a father who is under a legal obligation to maintain his children after a divorce, but who also agrees to maintain them, although involving a contractual obligation, is excluded from the scope of the Regulation.158 In contrast to this, if a person who is not under a legal obligation to provide maintenance for a member of the family, nonetheless agrees to do so, as where a child agrees to maintain a parent, this would fall within the scope of the Regulation.

The phrases “wills and succession” and “matrimonial property regime” are to be found among the list of exclusions from the scope of the Brussels I Recast,159 and their meaning has been discussed in that context. Moreover, the EU has now adopted specific instruments for these matters which may be consulted to interpret the scope of the respective exclusions in Rome I.160 What deserves to be mentioned, however, is that the exclusion in Article 1(2)(b) goes further than that in the Rome Convention in that it extends also to “relationships deemed by the law applicable to such relationships to have comparable effects”. According to Recital (8), the term “relationships having comparable effects to marriage and other family relationships” (such as registered partnerships) “should be interpreted in accordance with the law of the Member State in which the court is seised”. This clarifies that the definition of “relationships having comparable effects” is to be determined by the national (choice of law) rules of (p. 699) the forum in order to respect the traditional national boundaries of what is considered to be part of family law.

(iii)  Obligations arising under bills of exchange, cheques and promissory notes and other negotiable instruments to the extent that the obligations under such other negotiable instruments arise out of their negotiable character161

The identical exclusion but in respect of non-contractual obligations is to be found in the Rome II Regulation.162 The reason for the incorporation of the exclusion in Rome I is the same as the reason for its incorporation in the Rome Convention and Rome II Regulation, namely that “the Regulation is not the proper instrument for such obligations, that the Geneva Conventions of 7 June 1930 and 19 March 1931 regulate much of this matter and that these obligations are not dealt with uniformly in the Member States”.163

Under English law, negotiable instruments involve contractual obligations, but have long been subject to special rules, including those contained in the Bills of Exchange Act 1882, rather than being governed by the proper law of the contract.164 The effect of the exclusion of negotiable instruments from the Rome I Regulation is to preserve these special rules. The exclusion applies to bills of exchange, cheques and promissory notes, each of which category is well known to English lawyers. It also applies to “other negotiable instruments to the extent that the obligations under such other negotiable instruments arise out of their negotiable character”.165 “Other negotiable instruments” is not defined under the Regulation, and Member States may have different ideas on whether an instrument is negotiable. However, the Giuliano and Lagarde Report on the Rome Convention166 stated that it is for the private international law of the forum to determine whether a document is to be characterised as being negotiable. If the transfer takes place in England, the instrument is negotiable if English mercantile custom or a statute so provides. Examples of instruments which are negotiable in England include bonds issued by foreign governments and debentures issued to bearer by English companies. On the other hand, a bill of lading which is transferred in England is not negotiable, and is therefore within the scope of the Regulation.

Even if it can be shown that what is involved is a negotiable instrument other than a bill of exchange, cheque or promissory note, the exclusion is limited to cases in which the obligation arises out of the negotiable character of the instrument. This would cover a dispute where, for example, an acceptor of the instrument wants payment but the other party refuses, alleging that the acceptor is not a holder in due course of the instrument. Such a dispute would be outside the scope of the Regulation. On the other hand, contracts for the issue of, for example, Government bonds or for purchase/sale of such bonds are not concerned with the negotiable character of the instrument, and are thus within the scope of the Regulation.167

(p. 700) (iv)  Arbitration agreements and agreements on the choice of court168

The exclusion of arbitration agreements and agreements on the choice of court—which is identical in the Rome I Regulation and the Rome Convention—was probably the most controversial of the exclusions from the Rome Convention, with the United Kingdom delegation arguing unsuccessfully that such agreements should be subject to the rules contained in the Convention.169 The exclusion applies not only to separate arbitration or choice of jurisdiction agreements, ie agreements whose sole or main purpose is to provide for arbitration or a place of trial for a particular dispute, but also to arbitration or choice of jurisdiction clauses contained within a contract, which under English law are themselves regarded as separate agreements. However, when an arbitration or choice of jurisdiction clause is excluded,170 this only affects the clause itself; the remaining clauses in the contract will be within the scope of the Regulation and judges will have to apply the rules under the Regulation to them. This is uncertain for arbitration tribunals, for which it is disputed whether they are bound by the Rome I Regulation at all (and EU conflict of law rules in general).171

The exclusion in Article 1(2)(e) obviously relates to any choice of law issues that arise with regard to arbitration agreements and agreements on the choice of court, such as the formation, validity172 and effects of such agreements. It is also said to relate to any procedural questions that arise in relation to the arbitration.173 The result of the exclusion is that national courts will continue to apply their own rules of private international law to arbitration agreements and agreements on the choice of court. In England’s case this means the traditional common law rules.174 Contracts will have to be split up so that a question, for example, of interpretation of a choice of jurisdiction clause will have to be determined under the traditional common law rules, whereas the rest of the contract will be governed by the rules applicable under the Regulation. This can lead to different laws governing the agreement on arbitration/choice of court and the rest of the contract.175 One could end up with a contract (p. 701) which is void according to the rules on the applicable law contained in the Regulation, but which contains an arbitration agreement which is valid according to its proper law. It was in order to avoid such splitting of the contract that the United Kingdom argued that arbitration and choice of jurisdiction agreements should not be excluded from the scope of the Convention (and now Regulation).

(v)  Questions governed by the law of companies and other bodies, corporate or unincorporated, such as the creation, by registration or otherwise, legal capacity, internal organisation or winding up of companies and other bodies, corporate or unincorporated, and the personal liability of officers and members as such for the obligations of the company or body176

This provision—almost identical in the Rome Convention and the Rome I Regulation—clarifies the point that, if contractual matters are raised in a company law context,177 they fall outside the scope of the Regulation. Examples of matters excluded by this provision are the contract which, under English law, is contained in the memorandum and articles of association of a company, and a contract to wind up a company, including mergers or grouping of companies.178 The legal capacity of a company to contract is also excluded from the scope of the Regulation.179 On the other hand, an agreement by promoters to form a company is apparently not excluded from the scope of the Regulation.180 This is presumably on the basis that this is a purely contractual matter and is not governed by company law. Moreover, the merger of a company under which it loses its legal capacity does not affect the law applicable to contracts taken out by the acquired company with third parties.181

(vi)  The question whether an agent is able to bind a principal, or an organ to bind a company or other body corporate or unincorporated, in relation to a third party182

The question whether an agent is able to bind a principal is also excluded from the scope of the Rome I Regulation.183 The original proposal for a Regulation did not exclude this question and contained a proposed new choice of law rule dealing with this issue and other issues arising where a contract is concluded by an agent.184 However, this special rule was subsequently deleted and the original exclusion in the Rome Convention was restored, albeit with slightly modified wording.

(p. 702) The exclusion is only concerned with the relationship between a principal and a third party, and is confined to the specific question of whether the principal is bound vis à vis third parties by the acts of the agent.185 It follows that, for example, a contractual dispute between the principal and agent arising out of the contract of agency is not excluded. The exclusion is therefore a narrow one. However, it does encompass the question whether an organ of a company can bind the company. This raises the question of ultra vires, which under English law is a question of company law. The exclusion has been explained186 on the basis that the principle of freedom of contract, which is deeply enshrined in the Convention’s (and Regulation’s) rules on the applicable law,187 is difficult to accept in relation to the matter excluded. As far as English law is concerned, the effect of the exclusion is the retention of the common law rule under which the proper law of the contract concluded between the agent and third party governs the question of whether the principal is bound vis à vis third parties by the acts of the agent.188

(vii)  The constitution of trusts and the relationship between settlors, trustees and beneficiaries189

The English concept of a trust is said to define the subject matter of this exception.190 This raises the question of why this exclusion of the common law trust was introduced. It is presumably because, under English law, the constitution of trusts and the relationship between trustee/beneficiary and settlor/trustee are not based on contract. The exclusion is for the sake of clarity. There are Continental equivalents of a trust which are contractual in origin and thus appear to come within the Regulation. However, these will also be excluded if they exhibit the same characteristics as a common law trust.191 It is noticeable that the exclusion does not extend to trust property, although this can in fact raise contractual problems. For example, a trustee could invest in property abroad and could then be sued in contract by the vendor of the property. This situation appears to come within the scope of the Regulation.

(viii)  Obligations arising out of dealings prior to the conclusion of a contract192

The Rome I Regulation differs from the Convention by expressly excluding obligations arising out of dealings prior to the conclusion of a contract (culpa in contrahendo, eg violation of the duty of disclosure and the breakdown of contractual negotiations) on the basis that these fall within the scope of the Rome II Regulation,193 which has a separate free-standing provision dealing with non-contractual obligations arising out of dealings prior to the conclusion of a contract.194 If, however, the breach concerns an obligation arising from a tacit contractual (p. 703) relationship existing between the parties, such a dispute may be classified as contractual (and thus fall in the scope of Rome I).195

(ix)  Insurance196

Unlike the Rome Convention,197 the Rome I Regulation covers most (private198) insurance contracts, irrespective of whether an intra-EU or a third-state case is concerned.199 The inclusion of insurance contracts and the adoption of a special conflict rule for these contracts200 is a result of a controversial discussion in the legislative process. Despite a specific question in the green paper201 and academic proposals to include insurance contracts in Rome I,202 the initial proposal of the Commission excluded them from the scope of the Regulation.203 The debate continued204 and soon became one of the most controversial questions of the whole project, prompting a consultation of stakeholders from the United Kingdom205 which opposed a proposal206 for extending choice of law restrictions to the insurance of (third-country) mass (non-consumer) risks.207 As a compromise,208 insurance contracts were included, but any substantial change in critical points was avoided by merely reflecting in Article 7 Rome I Regulation the conflict-of-law rules which were previously spread over several Directives.209 This has led to a consolidation of all insurance conflict rules in one instrument which is probably one of the greatest achievements of the new Regulation.210 Still, a few aspects of insurance contracts are not covered by the Regulation.

First, Article 1(2)(j) Rome I Regulation excludes insurance contracts arising out of operations carried out by organisations other than undertakings referred to in Article 2 of the consolidated life assurance Directive211 “the object of which is to provide benefits for employed or (p. 704) self-employed persons belonging to an undertaking or group of undertakings, or a trade or group of trades, in the event of death or survival or of discontinuance or curtailment of activity, or of sickness related to work or accidents at work”. This exclusion—modelled on Article 3(3) Life Assurance Directive 2002/83/EC212—has been justified by the proximity of such insurance contracts to the social security system of the country where the insurer is established. Its practical relevance is limited because the exclusion applies only to undertakings established in a third country (non-EU/EEA State)213 which would normally be required to establish an agency or branch in the respective Member State to provide insurance services in that state214 and thus fall under EU law.215

A further exclusion which may be relevant for insurance is Article 1(2)(f) concerning “questions governed by the law of companies and other bodies, corporate or incorporate”. This provision might become relevant if insurance is organised in the form of a mutual association membership such as the German “Versicherungsverein auf Gegenseitigkeit”. Even if the membership in such an association and the contract of insurance are normally regarded as inextricably linked, it may be appropriate to apply Rome I to those questions of the insurance relationship which are not specific to the membership in the organisation, i.e. which could arise in a similar manner if the insurer was not organised as a mutual association and the insurance was based on an insurance contract alone.

Finally, Article 1(2)(i) excludes obligations arising out of dealings prior to the conclusion of a contract (culpa in contrahendo) from the scope of Rome I. Such obligations—eg the obligation to inform or advise the insured—are covered by Article 12 Rome II Regulation. Still, as Article 12(1) Rome II refers back to the law that applies to the contract, which is again determined by the rules of Rome I, a synchronisation of contractual and pre-contractual obligations will normally be achieved despite Article 1(2)(i) Rome I.216 Beyond specific exclusions, some aspects of insurance contracts fall outside Rome I because they have been considered to be non-contractual in nature: According to Article 18 Rome II Regulation, a direct action against the insurer is possible if either the law applicable to the non-contractual obligation or the law applicable to the insurance contract (as determined by the Rome I Regulation) so provides. Article 19 Rome II Regulation refers the question of subrogation to the law which governs the third person’s (i.e. insurer’s) duty to satisfy the creditor, i.e. for the case of insurance to the law governing the insurance contract which is determined by Rome I.217

(x)  Evidence and procedure, without prejudice to Article 18218

This provision excludes two matters, procedure and evidence. The exclusion of evidence is not total, but is subject to Article 18,219 which subjects three specific evidential matters, namely presumptions of law, the burden of proof (in so far as this raises rules of substance) and proving a contract, to the rules of the Regulation. The exclusion of evidence and procedure was said by the Giuliano and Lagarde Report to require no comment.220 Nonetheless, two (p. 705) obvious questions need to be asked. First, why were these matters expressly excluded from the scope of the Regulation? Presumably, this is just for the sake of clarity. Procedural and evidential matters would not appear to come within the scope of a Regulation which is concerned with contract choice of law (a matter of substance) and therefore do not need expressly to be excluded. Secondly, there is the vital question of when a matter is to be classified as being one of procedure. Procedure is a very different matter from the other matters excluded in that it involves a potential escape device, ie if you classify a matter as being purely procedural you escape from the choice of law rules under the Regulation. National courts are likely to resort to their own traditional ideas of what is a procedural matter. However, English courts cannot automatically assume that the classifications which they have adopted in the past will continue to be appropriate under the Regulation. For example, the question of whether a contract has to be in writing was classified at common law as being one of procedure. Under the Regulation, seemingly, it is to be regarded as a matter of substance raising an issue of formal validity of the contract.221 Moreover, certain remedies222 and the rules on prescription and limitation of actions223 are classified as substantive and not procedural, thereby falling within the scope of the Regulation. The danger of different states classifying the same matter differently can be avoided by adherence to the principle of uniform interpretation. Once it has been decided that the issue is one of evidence or procedure, the effect of the exclusion is that this issue is left to be governed by the forum’s rules on private international law. Under English private international law all procedural matters (including evidence) are automatically a matter for the law of the forum.224

(ii)  The universal application of the Regulation

The Rome I Regulation is intended to be of universal or world-wide application, ie it applies in any EU Member State forum,225 regardless of whether the contract has any connection with an EU Member State.226 In particular, there is no need for either party to the contract to be domiciled or resident in a Member State. The only thing that matters is that the dispute is tried in an EU Member State bound by the Regulation. Thus a contractual dispute between a New York resident and an Ontario resident which is tried before the Commercial Court in England will be subject to the Regulation. This avoids the need to distinguish for choice of law purposes between EU Member States and third States, a distinction which would be particularly difficult to apply to contracts which involve connections with both a Member and a non-Member State.227

Article 2228 provides that: “Any law specified by this Regulation shall be applied whether or not it is the law of a Member State.” This provision makes it clear that if the uniform rules under the Regulation point, for example, to Japanese law as the law governing the contract, the courts of Member States will apply that country’s law, even though Japan is not an EU Member State. However, Article 2 only deals with one aspect of the universal application of the rules in the Regulation. It says nothing about whether the situation or the parties must have a connection with a Member State. It is the Giuliano and Lagarde Report229 which makes it clear that the Rome Convention (and now the Rome I Regulation) is intended to (p. 706) have universal application and, in particular, will apply to nationals of third states and to persons domiciled or resident therein.230

(c)  The applicable law

The provisions on the applicable law are at the heart of the Regulation. Two basic distinctions have to be made: First, it needs to be considered whether one of the special rules in Articles 5–8 applies, which is the case for contracts of carriage, consumer contracts, insurance contracts, and individual employment contracts.231 Then, a distinction is drawn between the situation where the law is chosen by the parties and the situation where the applicable law is ascertained in the absence of choice. Choice is concerned with the actual intentions of the parties (either expressed by the parties or inferred by the court) and absence of choice requires reference to objective connections localising the contract. The applicable law under the Regulation, whether chosen or not, refers to the domestic law of the country in question, and there is no place for the doctrine of renvoi.232 It is presupposed that there has to be an applicable law at the time when the contract is concluded.233 It follows that the choice of a “floating” proper law, ie a proper law which was non-existent at the time when the contract was made but which was crystallised later on by the unilateral act of one of the parties, will be ineffective at the time when the contract is made. However, the Regulation does allow the parties to vary the applicable law during the subsequent life of the contract.234

(i)  The law is chosen by the parties

Any reference to choice of the applicable law raises a number of points which will be examined under the following headings: freedom of choice; limitations on choice; express choice; inferred choice; consent to choice. Before doing so, it must be pointed out that there are no formalities to be satisfied in relation to the parties’ choice.235

(a)  Freedom of choice236

(i)  The basic principle

Article 3 Rome I Regulation237 is entitled “Freedom of choice”, and paragraph (1) sets out the basic principle238 that “a contract shall be governed by the law chosen by the parties”. The parties’ freedom to choose the governing law had been accepted in all the Member States of the European Union already before the Convention for many years.239 In the United Kingdom (p. 707) the philosophical origin of this freedom is to be found in the fidelity of the Victorian judges to the Benthamite dogma of laissez-faire,240 although authority for allowing the parties expressly to select the governing law pre-dates this.241 In more modern policy terms, party autonomy provides the certainty and predictability which are essential in commercial matters.242 The philosophy of freedom of choice underlies not only the basic principle of allowing the parties to choose the law governing the contract but also some of the more detailed provisions relating to choice. Parties are given the freedom to pick and choose the applicable law so that it governs the whole or merely part of the contract. The parties are free to exercise their choice at any time and to vary their choice. These freedoms will now be examined.

(ii)  Chosen law may be unconnected with the contract

As can be seen from Article 3(3), the Regulation permits the choice of the law of another country even where all elements relevant to the situation are located in a country other than the country whose law has been chosen. Therefore, the parties are free to choose any law they desire, even if it may be wholly unconnected with their contract.

(iii)  Dépeçage243

The last sentence of Article 3(1) provides that: “By their choice the parties can select the law applicable to the whole or to part only of the contract.”244 The parties are given the freedom to pick and choose the applicable law and thereby sever the contract (dépeçage).245 The parties can choose different laws for different parts of the contract. Thus there could be an express choice of French law to govern one part, but an express choice of German law to govern the rest of the contract. The choice can be expressed by the parties or inferred by the court. If the parties choose different laws for different parts of the contract the choices must be logically consistent, ie they “must relate to elements in the contract which can be governed by different laws without giving rise to contradiction”.246 The Giuliano and Lagarde Report gives two contrasting examples.247 An index linking clause may be made subject to a law different from the rest of the contract. On the other hand, it was thought unlikely that repudiation of the contract for non-performance could be subjected to two different laws, one for the vendor and the other for the purchaser. If the chosen laws cannot be reconciled, both choices fail and the rules on the applicable law in the absence of choice248 have to be used. The “general obligation” under a contract (ie the contractual obligations governing the core of the parties’ relationship, such issues as whether the contract is discharged by frustration or the innocent (p. 708) party can terminate the contract on account of the other party’s breach) can only be governed by one law.249 But it is possible for the general obligation under an agreement to be governed by New York law whilst payment instruments (ie cheques) and their validity, etc are governed by the drawee’s law (with drawees from many different jurisdictions).250 A breach of an insurance policy condition precluding the insured from making recovery from the insurer under a particular part of the policy can be severed from the insured’s rights under other parts of the policy.251 But the words of the policy defining the insured cannot be severed so as to be interpreted by different laws and given possibly different meanings, depending on the part of the globe in which events may occur giving rise to the claim.252 The parties can choose a law to govern part of the contract but exercise no choice in respect of the remainder of the contract. In this situation the applicable law for the remainder of the contract must be ascertained, again, by the rules on the applicable law in the absence of choice. The Working Group on the Rome Convention rejected the notion of a presumption that the law chosen for one part of the contract should govern the entirety.253

What is meant by part of the contract? Obviously, this covers parts of the contract that may be regarded as independent of each other. The Giuliano and Lagarde Report seemed to go further and permitted the parties to choose one law to govern a particular clause,254 and a different law to govern other clauses. From the example given above relating to repudiation of the contract, it can also be deduced that part can include a particular issue255 relating to the contract. Accordingly, the parties could choose one law to govern the interpretation of the contract and a different law to govern its discharge. On the other hand, it seems from the same example that the parties are not free to take a single issue, such as repudiation of the contract, and to split this so that one law governs one party’s rights and a different law governs the other party’s rights. This is regarded as involving two choices which are logically inconsistent. Moreover, the Court of Justice has held for dépeçage in the absence of choice under Article 4 of the Convention that a part of a contract may be made subject to a different law only where the object of that part is independent in relation to the rest of the contract.256 In particular, the Court has found that the rules relating to the prescription of a right must fall under the same legal system as that applied to the corresponding obligation.257 While dépeçage as a consequence of the parties’ choice of law under Article 3 could be permitted more liberally than in the absence of choice under Article 4,258 it cannot be overlooked that both provisions use the words “part of the contract” which have been interpreted narrowly by the Court of Justice. Therefore, it can be accepted that the parties choose different laws to govern particular issues, provided that the issues are independent from each other.259 Some contracts are, by their very nature, severable, for example a contract which turns out to consist of several independent (p. 709) contracts. Different laws can clearly be applied to these different contracts, without having to resort to the dépeçage provision.

(iv)  Timing of choice

The first sentence of Article 3(2) provides that: “The parties may at any time agree to subject the contract to a law other than that which previously governed it, whether as a result of an earlier choice under this Article or of other provisions of this Regulation.” The policy underlying this provision260 is that of providing maximum freedom as to when the parties can make their choice.261 It can be made before the contract is concluded, at the time of or even after the conclusion of the contract. If the parties’ choice is made for the first time after the conclusion of the contract, then the applicable law at the time of the conclusion of the contract will have to be determined by reference to the rules determining the applicable law in the absence of choice (Article 4). This law will apply until the parties subsequently exercise their choice, which may involve a variation in the applicable law.

(v)  Variation of choice262

The parties’ freedom to vary the applicable law follows on logically from their right to choose the applicable law at any time.263 For example, the parties may have agreed at the time of contracting that Californian law shall govern the contract. They have the freedom under Article 3(2) to agree subsequently264 that, instead, Japanese law shall govern the contract.265 It is irrelevant that Californian or Japanese law might not allow variation.266 Equally, at the time the contract is made Luxembourg law may be applicable by virtue of the rules on the applicable law in the absence of choice (Article 4). The parties may subsequently agree that New York law shall govern. This subsequent agreement involves an exercise of the parties’ freedom of choice under Article 3(1) and so can be expressed or inferred.267 In The Aeolian268 the Court of Appeal held that a supply contract, which was governed by Japanese law, had not been varied by a subsequent undertaking (governed by English law) in relation to a claim for payment under a separate spares contract, even though both parties must be taken to have had in mind that the defence would rely, as a set-off to the claim for payment, on a cross-claim for breach of the supply contract. The agreement in the undertaking was confined to the claim for payment and did not extend to the cross-claim arising out of the supply contract.269 The undertaking made no reference to the supply contract and no inference could be (p. 710) drawn from the circumstances of the case that English law should govern the cross-claim.270 If the variation is made during the course of legal proceedings, it is for the forum’s law of procedure to decide the extent to which this is effective.271

The Working Group on the Rome Convention recognised that there were certain dangers in allowing a variation of the applicable law by the parties. The second sentence of Article 3(2) Rome I Regulation272 provides a safeguard in the following terms: “Any change in the law to be applied that is made after the conclusion of the contract shall not prejudice its formal validity under Article 11 or adversely affect the rights of third parties.”

As regards formal validity, the concern was that the new law chosen by the parties might contain formal requirements which were not known under the law originally applicable. This could create doubts as to the validity of the contract during the period preceding the new agreement between the parties;273 hence the rule that any variation by the parties is not to prejudice the formal validity of the contract under Article 11. The other danger recognised by the Working Group is in relation to third parties, who may have already acquired rights at the time of the conclusion of the contract between the original contracting parties. These rights cannot be affected by a subsequent change in the choice of the applicable law.

There are two other potential dangers that can arise from the parties’ variation of the applicable law. First, the parties might thereby evade the mandatory rules (eg controls on exemption clauses) of the country whose law was originally applicable.274 However, the normal limitations on the right to choose the applicable law will doubtless apply to a subsequent choice of the governing law in the same way that they apply to an initial choice. As will shortly be seen,275 there are limitations on choice which deal to some extent with this problem of evasion. Secondly, the parties might choose a new law which invalidates the contract. Logically, the contract appears to be rendered invalid. This presupposes, however, that the new choice of the applicable law is itself valid. This is a matter for the new law that has been chosen.276 Thus the validity of a New York choice of law clause (which operates as a subsequent choice) is a matter for New York law.

(vi)  Choice and the English rules on pleading and proof of foreign law277

What happens if the parties choose Utopian law to apply but subsequently neither party pleads Utopian law? The Regulation does not provide an answer to this. On the one hand, the English procedural rule preserved by the Regulation278 indicates that English law must be applied automatically.279 This would suggest that you can have a procedural variation of the applicable law.280 On the other hand, Article 3(1) is phrased in strong terms: the “contract (p. 711) shall be governed by the law chosen by the parties”.281 But if foreign law has to be applied, this leads on to a practical problem of what an English judge is to do if the parties fail to plead and prove foreign law.282 In view of this difficulty, English courts are likely to take a pragmatic line and simply apply English law under the English procedural rule.283 There would be no problem if it could be said that the failure to plead foreign law operates as a new agreement as to the applicable law by the parties replacing the original choice.284 However, the parties’ choice of the applicable law (whether an original or a later choice) must be made expressly or clearly demonstrated by the terms of the contract or the circumstances of the case.285 A mere omission to plead and prove foreign law would not appear to satisfy this requirement.286

(b)  Limitations on choice

Any discussion of freedom of choice inevitably leads on to the question of whether there is any restriction on the parties’ freedom to choose the governing law.287 The Regulation also lays down restrictions on the parties’ right to choose the governing law,288 which will now be examined.

(i) Article 3(3)289

A first limitation is contained in Article 3(3) which provides that:

Where all other elements relevant to the situation at the time of the choice are located in a country other than the country whose law has been chosen, the choice of the parties shall not prejudice the application of provisions of the law of that other country which cannot be derogated from by agreement.

The Rome I Regulation retained Article 3(3) Rome Convention in substance, but adopted a slightly different wording so as to align the provision as far as possible with the Rome II Regulation on non-contractual obligations.290 The most obvious change in wording is that the definition of “mandatory rules” is no longer used.291

Article 3(3) is concerned with the situation where there is an essentially domestic contract which is turned into a conflict of laws case by virtue simply of the parties’ choice of a foreign applicable law. The provision establishes a limitation on the right to choose in this situation, but only to the extent of preserving the rules which cannot be derogated from by agreement of the country where all the other relevant connections are situated. Such provisions “which cannot be derogated from by agreement” are sometimes referred to as (simple) mandatory rules.292 They have to be distinguished from overriding mandatory provisions in the sense of Article 9 which not only cannot be derogated from by agreement, but—in (p. 712) addition—“respect for which is regarded as crucial by a country for safeguarding its public interests”.293 As an example of English rules which cannot be derogated from by agreement in the sense of Article 3(3), there are the rules providing controls on exemption clauses or other standard terms such as contained in the Unfair Contract Terms Act 1977 and comparable laws in other countries.294 This Act makes it clear that these controls will, in certain circumstances, apply despite the parties’ choice of a foreign law to govern the contract.295 The effect of Article 3(3) is that if the parties to an entirely German contract, which contains an exemption clause, choose, for example, French law to govern, the court of any Member State which tries the case will have to apply any controls on exemption clauses contained in a German equivalent of the 1977 Act. The parties’ choice of French law would appear to have been made with a view to evading the German controls on exemption clauses. Article 3(3) will stop many cases of evasion of the law,296 although it goes wider than this and it will ensure that any German controls on exemption clauses apply even if the parties have chosen French law for some perfectly legitimate reason, such as the fact that this is the applicable law under some related contract between the parties.

There are a number of points that can be made in relation to Article 3(3). First, it requires the parties to have chosen a “foreign” law. This raises a problem in the following type of case:

Two Californian residents enter into an essentially Californian contract but choose English law to govern the contract. Trial of a subsequent dispute takes place in England. English law is “foreign” to the parties and the contract, but not “foreign” to the forum. Article 3(3) is concerned with the choice of a foreign law and with the situation at the time of the choice. At that time, English law was a foreign law, ie was foreign to the parties and the contract.

The result is that Article 3(3) operates and Californian mandatory rules are applicable.297

Secondly, Article 3(3) requires “all the other elements relevant to the situation”298 to be connected with a country in order that its provisions which cannot be derogated from by agreement are to be applied. But when is an element relevant to the situation? Take the facts of the well-known case of Golden Acres Ltd v Queensland Estates Pty Ltd.299

The case concerned the rate of commission to be paid to an estate agent. Many of the connections were with Queensland, but the plaintiff company was incorporated in Hong Kong. Was the place of incorporation a relevant element, or was the only relevant thing about the company the fact that it acted as an estate agent in Queensland?

Under the common law this case was decided on the basis that the choice of Hong Kong law to govern the contract was not made in good faith. Under the Regulation the parties’ motives (p. 713) are immaterial, but whilst one problem (ascertaining motives) has now disappeared, another problem (ascertaining whether all the relevant elements are with one country) has sprung up in its place.300 In Caterpillar Financial Services Corp v SNC Passion:301

A loan transaction had a number of connections with France: the borrower was a French company; the loan transaction was originally set up by French companies; it was argued that the loan agreement was made in France; the loan was for building a ship which would fly the French flag with a charter-party subject to French law. But the loan agreement was expressly governed by English law.

It was held that Article 3(3) had no application302 because significant elements of the loan transaction involved a connection with a jurisdiction other than France. The most important of these was that the lender bank was a Delaware bank which acted through its office in Tennessee. Moreover, the loan agreement was intimately connected with a shipbuilding contract with advances being made directly to the shipbuilder in Singapore. This was a clear-cut case and it would have been astonishing if the opposite conclusion had been reached.

Thirdly, the structure of the Regulation suggests that the country whose rules which cannot be derogated from by agreement have to be applied will be a foreign country and not the forum. However, there is nothing to say that mandatory rules of the forum are excluded under Article 3(3). Indeed, it is important that Article 3(3) should encompass the mandatory rules of the forum, since Article 9(2), although specifically designed to cover the overriding mandatory rules of the forum, is concerned with a different and narrower type of mandatory rule.303

Fourthly, you have to look to the law of the country with which there are all the other relevant connections to see whether under that country’s law the domestic rule is one which cannot be derogated from by contract.

Fifthly, the effect of applying a rule which cannot be derogated from by agreement is to override the parties’ choice of law, rather than to destroy it. Reverting to the earlier example, the German controls on exemption clauses will apply despite a French choice of law clause. Nonetheless, the choice of French law will still operate to govern other issues, such as interpretation of the contract, provided that this is an area where German law does not have mandatory rules.

The sixth and final point about Article 3(3) is that it tells us something, by implication, about freedom of choice. Article 3(3) is only concerned with provisions which cannot be derogated from by agreement. It follows that, even if all the other relevant connections are with Country X, the choice of law of Country Y will still apply as far as non-mandatory rules are concerned. This means that the parties can choose, as the applicable law, the law of a country with which there is no relevant connection.304

(ii)  Article 3(4): Mandatory rules of EU law

Article 3(4) of the Rome I Regulation introduces an additional limitation on the parties’ freedom to choose the applicable law. This provides that:

Where all other elements relevant to the situation at the time of the choice are located in one or more Member States, the parties’ choice of applicable law other than that of a Member (p. 714) State shall not prejudice the application of provisions of Community law, where appropriate as implemented in the Member State of the forum, which cannot be derogated from by agreement.305

This follows an almost identically worded limitation in the Rome II Regulation on non-contractual obligations.306 The provision treats the European Union as effectively one country and highlights the importance of EU rules which cannot be derogated from by agreement. Given that national law includes EU law, prima facie it may seem to be an open question whether Article 3(4) will in fact add anything to what is already provided for by Article 3(3). However, the difference is not the “quality” of law since both Article 3(3) and Article 3(4) address provisions “which cannot be derogated from by agreement” but the extension of the area from one state to the European Union being treated as one state. In other words: While Article 3(3) does not capture cases where the relevant elements are located in two EU Member States, Article 3(4) preserves mandatory EU law in such circumstances.

(iii)  Other provisions on mandatory rules,307 public policy308

These limitations are more appropriately dealt with later on, where they will be looked at in some detail.

We have dealt with the limitations on choice specifically set out in the Regulation itself. Nonetheless, there are a number of other possible limitations which have to be considered.

(iv)  Logically consistent choices

As has already been seen,309 if the parties are choosing two different laws for different parts of the contract, these choices must be logically consistent.

(v)  A meaningless choice of law

In keeping with the above limitation, a meaningless choice of law will be ignored. This was established by the Court of Appeal in Shamil Bank of Bahrain v Beximco Pharmaceuticals Ltd:310

A choice of law clause provided that the contract was governed by English law “subject to the principles of the Glorious Sharia’a”. It was common ground that only the law of a country311 can be chosen and therefore it was argued that incorporated into English law were specific principles of Sharia law. But there was no reference to or identification of those aspects of Sharia law which were intended to be incorporated into the contract. The principles of Sharia law were therefore repugnant to the choice of English law and rendered the clause self-contradictory and meaningless.

The construction adopted of the clause by the Court was that English law governed and the additional wording merely reflected the Islamic religious principles according to which the bank held itself out as doing business rather than a system of law intended to trump the application of English law. In contrast, at common law mere difficulty in ascertaining the governing law did not render the choice ineffective.312

(p. 715) (vi)  Choice of the law of a country

The original proposal for a Rome I Regulation intended to give a boost to the principle of freedom of choice by allowing the parties to choose as the applicable law the law of a non-state.313 Under this provision the parties were able to choose “the principles and rules of the substantive law of contract recognised internationally or in the Community”.314 This would include UNIDROIT principles, the UN Convention on the International Sale of Goods (the Vienna Convention) and the Principles of European Contract Law.315 It was not intended to cover the lex mercatoria, which was regarded as not being precise enough, or private codifications not adequately recognised by the international community.316 It has been questioned whether such a change in the law would do more than assist in interpretation of the contract since remedies, if they are to be effective, would have to flow from the law of a country.317 The provision allowing a choice of non-state law was subsequently deleted.318

As a consequence, under Article 3(1) of the final text of the Regulation, as under the Rome Convention, the parties’ right to choose the applicable law refers to the law of a country.319 The Regulation does not sanction the choice of a non-national system of law such as the lex mercatoria.320 Such a choice refers not to the national law of any country but rather to a kind of transnational law consisting of internationally accepted principles of trade law, to be ascertained by arbitrators. Other examples of the choice of a non-national system of law are “general principles of law,” the rules of law in the UN Convention on the International Sale of Goods 1980 (where they do not apply as part of a Contracting State’s law, but by abstract choice by the parties),321 the law of Sharia322 and Jewish law.323 Such a reference is outside the parties’ freedom to choose the applicable law.324

However, the parties are free to incorporate by reference into their contract a non-state body of law or an international Convention,325 such as the UN Convention on the International (p. 716) Sale of Goods 1980, or even provisions of religious law such as the Sharia, provided that the parties in their contract have sufficiently identified the specific provisions.326 It is also possible to incorporate “Jewish law”, without any such proviso, at least where the dispute is between Orthodox Jews who are agreed there is such a distinct body of law.327 Such an incorporation by reference is no choice of law, but only an agreement of the parties about the terms of their contract. As a consequence, the provisions incorporated by reference must comply with the law of the country which is applicable to the contract under the Regulation. Where this law includes provisions which cannot be derogated from by agreement, the incorporation by reference, as a simple contractual term, is ineffective. Finally, should the EU adopt in an appropriate legal instrument rules of substantive contract law, including standard terms and conditions, that instrument may provide that the parties may choose to apply those rules.328

(vii)  A “floating” applicable law

The parties cannot choose a “floating” applicable law to govern the contract. The applicable law must exist and be identifiable at the time when the contract is made.329 It follows that a clause which, for example, gives one party the option to determine the applicable law in the future by selecting the law to govern from a list of possible alternatives, will be ineffective at the time when the contract is made, and at this stage the applicable law will be determined objectively under Article 4 of the Regulation.330 But what happens if one party makes the selection at some future date?331 This subsequent choice may have been ineffective under the common law rules,332 but the choice at this future date will be given retrospective effect both under the Convention333 and the Regulation,334 since variation of the applicable law is permissible.335 The form of the variation is just rather unusual in this situation, in that the agreement is from the outset for a change of the applicable law in the future.

(p. 717) (c)  An express choice; mere incorporation

The parties can express a choice simply by including a choice of law clause in the contract stating that, for example, all disputes shall be governed by English law.336 Any question as to the validity or existence of this choice is governed by the rules on consent to choice, which will be examined later.337 Parties may choose a particular law for a variety of reasons.338 It is usually convenient for a party to have the familiar law of their home state apply. One country’s law may be more developed than another’s in technical commercial areas such as banking and insurance, when English law is commonly chosen by the parties. It may have become standard practice for a particular country’s law to apply to certain transactions. The content of one country’s law may be more favourable to one of the parties than that of another. Whatever the law chosen, it is important that the parties should make an express choice, for without this there is considerable uncertainty as to the applicable law.

The express selection of the proper law (which is a choice of law issue) must be distinguished from the quite different process of the incorporation in the contract of certain domestic provisions of a foreign law, which thereupon became terms of the contract (and which is an issue of the applicable substantive law).339 Article 3 is only concerned with selection of the applicable law. While contractual reference to specific articles of a foreign law may be interpreted as an inferred choice of law,340 this is not necessarily the case. It may also be understood as a mere incorporation in the contract of certain provisions of a foreign law as terms of the contract. Such incorporation may be effected either by a verbatim transcription of the relevant provisions or by a general statement that the rights and liabilities shall in certain respects be subject to these provisions. The latter is only a shorthand method of expressing the agreed terms. Thus the parties to an English contract may expressly provide that their duties with regard to performance shall be regulated by certain specific rules contained in the French Civil Code. English law is then applied as the governing law to a contract into which the foreign rules have been incorporated.341 It is possible to incorporate by reference the terms of the Sharia or of an international code. However, the doctrine of incorporation can only sensibly operate where the parties have, by the terms of their contract, sufficiently identified specific “black letter” provisions of the foreign law or international code.342 It is common in cases involving carriage of goods by sea to incorporate the Hague Rules. It is possible to incorporate all of the Hague Rules or merely specific provisions. Whether a particular term incorporated in this manner is valid and effective is a matter for determination by the law applicable under the Regulation.343 At common law, once a foreign law was incorporated into the contract as a term it remained constant in the sense that it was static, ie unaffected by any change in the relevant foreign law occurring after the date of the contract.344 Under the Regulation, the question of a static or dynamic reference to foreign law incorporated as (p. 718) a contractual term will be decided by the applicable law, as it is a matter of interpretation of the contract.345

(d)  An inferred choice

Article 3(1) provides, as an alternative to an express choice, that there can be a choice “clearly demonstrated by the terms of the contract or the circumstances of the case.” This provision is concerned with a real choice by the parties, the court inferring what the parties’ actual intentions346 were from the terms of the contract or the circumstances of the case. This concept of an inferred or implied choice is well known in English law, and is to be found in civil law countries as well.347 Inferred choice under the Regulation appears at first sight to be very close to the English law in this area. Nonetheless, there are differences, and cases decided at common law should be treated with caution: “It would be a mistake to attempt to apply article 3 through the prism of the preceding common law.”348 Different from the wording of Article 3(1) of the Rome Convention, which required such a choice to be demonstrated with “reasonable certainty”, the Regulation provides now that an inferred choice has to be “clearly demonstrated”.349 It is not entirely clear whether this represents a change in substance, requiring a higher standard to be demonstrated. Against a change from Convention to Regulation, it has been argued that the wording in the Regulation merely brings the English and German texts in line with the French text of the Convention.350 On the other hand, the fact that the more stringent French wording prevailed could, together with a remark in the Explanatory Memorandum,351 hint towards a stricter test for inferred choice of law.352 Even if one finds the latter position more convincing, the effect of this change is likely to be negligible, because a court prepared to find an inferred choice “demonstrated with reasonable certainty” is likely to arrive at the same result also under the “clearly demonstrated”—test. Therefore, it is submitted that the examples discussed under the Rome Convention are still equally relevant under the Regulation.

(i)  Drawing the inference

An inference as to the parties’ intentions can be drawn from either the terms of the contract or the circumstances of the case. The test whether an implied choice of law has been established is objective:353 A party asserting an implied choice of law has to establish that, on an objective view, (p. 719) the parties must have taken it without saying that their contract should be governed by that law. The party does not have to prove that there was in fact a subjective conscious choice, but it has to satisfy the court that the only reasonable conclusion to be drawn from the circumstances is that the parties should be taken to have intended the putative law to apply.354

The terms of the contract

The Giuliano and Lagarde Report355 on the Rome Convention provides a number of examples of situations where a court may draw an inference as to the parties’ intentions. In most of these examples the inference is being drawn from the terms of the contract. Thus an inference can be drawn in cases where the contract is in a standard form known to be governed by a particular system of law, such as a Lloyd’s policy of Marine Insurance; the contract contains an exclusive choice of forum clause356 or an arbitration clause naming the place of arbitration (at least in circumstances indicating that the arbitrator should apply the law of that place); or there is a reference to specific articles of the French Civil Code.357 The reasoning in such cases is that if, for example, the parties intended that trial should take place in England they must also have intended that English law should apply, it being inconvenient and expensive for a foreign law to be applied.

The interesting thing about these examples, from an English point of view, is that the first three are the standard examples of an inferred choice under the common law rules. Thus in Amin Rasheed Shipping Corpn v Kuwait Insurance Co358 Lord Diplock359 said that the terms of a standard Lloyd’s SG form of policy showed by necessary implication that the parties (a Liberian company and a Kuwaiti insurance company) intended that the English law of marine insurance should apply. Kuwait had no law of marine insurance at that time, and the parties could not have intended Kuwaiti law to apply. The House of Lords unanimously held that the proper law of the contract was English law. This sort of case will be decided in exactly the same way under the Convention360 and under the Regulation. Similarly, where a reinsurance contract was placed in London on the London market, it was held that the terms of the slip (referring to standard form clauses used in England where they were developed) pointed to an implied choice of English law under the Convention.361 If, on the other hand, a contractual clause is universal in international trade, such as “cif UK port”, this cannot support an inferred choice of law.362

(p. 720) Similarly, as at common law, a choice can be inferred from an arbitration clause indicating the place of arbitration or a choice of (exclusive) jurisdiction clause.363 Thus in Egon Oldendorff v Liberia Corpn,364 a clause providing for arbitration in London365 was treated as an inferred choice of English law to govern the contract by virtue of Article 3(1). It was also relevant in this case when inferring intention that the parties used a well-known English language form of charter-party containing standard clauses with well-known meanings in English law. The case for an inferred choice is even stronger if the arbitration clause not only provides for arbitration in England but also expressly refers to the English Arbitration Act.366 Likewise an (exclusive) jurisdiction clause providing for trial in England has been treated as an inferred choice of English law under the Convention.367 For exclusive jurisdiction clauses, Recital (12) of the Regulation now provides that: “An agreement between the parties to confer on one or more courts or tribunals of a Member State exclusive jurisdiction to determine disputes under the contract should be one of the factors to be taken into account in determining whether a choice of law has been clearly demonstrated.” The background to this Recital is that a choice of jurisdiction agreement is an example of an inferred choice under the Rome Convention, but courts in some Contracting States have been more reluctant than the English courts have been to draw the inference in such a case.368 The original proposal for a Regulation therefore provided that if the parties have agreed to confer jurisdiction on one or more courts or tribunals of a Member State to hear and determine disputes that have arisen or may arise out of the contract, they shall be presumed to have chosen the law of that Member State.369 But this provision was subsequently deleted. Recital (12) in its final wording does not go as far as the presumption would have done in terms of the impact of a choice of jurisdiction agreement. As far as England is concerned, it will make no difference to drawing an inference. But it is possible that it might prompt the courts in other Member States to draw the inference when they would not have done so otherwise.

These are only examples of situations where it is possible to infer a choice from the terms of the contract, albeit particularly good ones. Could such an inference be drawn, for example, from the fact that the currency in which payment is to be made is that of a particular country? At common law the English courts drew inferences as to parties’ intentions from a wide variety of factors relating to the terms of the contract, including this very factor. However, a note of caution should be struck when it comes to inferring a choice under the Regulation. First, the inferred choice must be “clearly demonstrated”. If it is not, you move on to the provisions on the applicable law in the absence of choice (Article 4) to decide the case. Secondly, you are looking, on the basis of an objective test, for the actual or real intentions of the parties.370 It is just about credible to say in a case like Amin Rasheed (or in cases involving arbitration or (p. 721) choice of jurisdiction clauses) that the parties had real intentions, but the same cannot be said in a case where all that can be shown is that, for example, there is a clause in the contract relating to the currency in which payment is to be made, or where a specific language is used.371 In the common law cases, although the language of inferred intent was often used it was by no means clear that the courts were looking for a real or actual intention on the parties’ part.372 The upshot is that it may be harder to draw an inference of intention under the Regulation than it was under the common law rules.373

The circumstances of the case374

The inference can be drawn not only from the terms of the contract, but also from the circumstances of the case.375 The Giuliano and Lagarde Report376 gives two examples of inferred choice which would seem to fit within this category. The first is the situation where there is an express choice in a related transaction. The second is the situation where there is a previous course of dealing under contracts containing an express choice of the applicable law and this choice of law clause has been omitted in circumstances which do not indicate a deliberate change of policy by the parties.

However, whether the inference can be drawn will depend very much on the facts of the case. It will be recalled that in The Aeolian377 the Court of Appeal held that an inference as to the law governing a supply contract, out of which a counterclaim arose, could not be drawn from the fact that the parties agreed that English law governed an undertaking in relation to a claim for payment under a separate spares contract. On the other hand, an inference could be drawn from the fact that an excess insurance policy followed the choice of law in the primary reinsurance policy.378 An inference is likely to be drawn where the parties made a choice of the governing law for a main contract and later conclude a contract ancillary to that main contract, such as a guarantee,379 or a collateral contract such as a contractual indemnity,380 or an agency contract,381 or an implied contract of warranty of authority:382 If the rights and obligations under the main contract are governed by a chosen system of law it would be incongruous for the ancillary contract to be governed by a different system of law because (p. 722) this could potentially result in a mismatch between the obligations of the party to the main contract and the obligations of the ancillary contract to fulfil those obligations.

The English courts, under the common law rules, also inferred an intent from purely objective factors, such as the residence of the parties383 or the nature and location of the subject of the contract.384 This approach has continued to be used in relation to the Rome Convention. In American Motorists Insurance Co (Amico) v Cellstar Corp385 the Court of Appeal inferred a choice of Texas law from the fact that a company from its base in Texas chose on behalf of its whole group of companies to negotiate world-wide cover in Texas with insurers also based in Texas. Whatever the position was at common law, it is submitted that it is inappropriate to infer an actual intention as required under the Convention and Regulation from purely objective factors such as residence of the parties or location of the subject of the contract. The same—no inference—holds true for the mere fact that a contract was written in a particular language or for ritual conduct out of mere politeness, such as a handshake.386

The circumstances of the case can also be taken into account to show that no inference can be drawn as to the governing law. This was the situation in Samcrete Egypt Engineers and Contractors SAE v Land Rover Exports Ltd387 where the fact that one party put forward a choice of law clause and the other party deleted it from the draft contract was a positive indication that the parties made no choice as to the applicable law under Article 3(1).388 The result was that no choice of English law could be inferred from the fact that this was a guarantee contract which was dependant on a principal contract which was governed by English law. Without taking into account the circumstances of the case, there would have been a strong argument for drawing such an inference.389

(ii)  Conflicting inferences

The Giuliano and Lagarde Report on the Rome Convention states390 that any inference which arises from a choice of jurisdiction clause “must always be subject to the other terms of the contract and all the circumstances of the case” (ie the very matters from which an inference can be drawn). For example, a choice of exclusive jurisdiction clause may point to an intention that the law of Country A shall apply, whereas a previous course of dealing may point to an intention that the law of Country B shall apply. However, no conflicting inference could be drawn from the fact that an earlier draft of the contract contained an express choice of law clause (providing that English law governed) and this clause was deleted.391 This notion of conflicting inferences is not confined to the situation where an inferred choice is being drawn from the presence of a choice of jurisdiction clause in the contract. Thus with an excess reinsurance contract, the inference to be drawn from the fact that the primary reinsurance was governed by Mauritian law conflicted with the inference to be drawn from the fact that the excess reinsurance was written on a Lloyd’s slip in the London market on London market (p. 723) terms.392 There was a stalemate.393 If there are conflicting inferences it cannot be said that the choice has been “clearly” demonstrated, and Article 3(1) does not permit the court to infer a choice of law that the parties might have made if they had no clear intention of making a choice.394 The result is that you have to turn to the rules on the applicable law in the absence of choice395 to determine the governing law.396

What is less clear is how the inference to be drawn from the terms of the contract and the circumstances of the case stands in relation to the objective connections that the contract has with different countries. For example, a previous course of dealing may raise the inference that English law governs, but many of the objective connections, such as the residence of the parties and the place of performance, could be with France. Under the Regulation there is, quite properly, a rigid separation of intention (dealt with under Article 3) from objective connecting factors (dealt with under Article 4). The inference that the parties intended English law to govern can only be challenged by a conflicting inference, ie by evidence showing a real intention that French law should govern. It has been considered as possible, though, to take a robust view that such an inference can be drawn also from the factual connections with France. These could be considered on the basis that they constitute the surrounding circumstances.397

This robust view was discussed by Mance J in Egon Oldendorff v Liberia Corpn.398

In this case the plaintiff company was German, the defendant Japanese, and the contract provided for arbitration in London. It was argued that the arbitration clause was a minor factor and that other factors pointed objectively to Japanese law. Not only was the defendant corporation Japanese but also a Japanese shipbroker acted as intermediary between the parties, and the ships chartered were to be delivered and redelivered in Japan.

Mance J rejected this argument, saying that these matters were inadequate to lead him to conclude that the parties intended London arbitration under Japanese law,399 and held that there was a good arguable case for the purposes of service out of the jurisdiction that English law governed. Thus, although it may not be entirely excluded to draw a conflicting inference from objective factors, it seems that the circumstances where it will be right to do so may be relatively rare. Once it had been decided that there was jurisdiction, the issue arose of whether English law did in fact govern. Clarke J held that it did.400 Having agreed a “neutral forum”, the reasonable inference is that the parties intended that forum to apply a “neutral” law, namely English law. In the circumstances, the arbitration clause was a strong indication of the parties’ intention to choose English law. Clarke J thought that the approach towards arbitration clauses under Article 3 was little or any different from that at common law.401

(e)  Consent to choice

There can be a dispute as to whether one of the parties has consented to the choice. Article 3(5) provides that issues in relation to the validity and existence of consent are determined in (p. 724) accordance with the special rules in the Regulation relating to material validity (Article 10), formal validity (Article 11) and incapacity (Article 13).402 These provisions will be discussed later in this chapter.403 If there is no valid consent to the choice, the applicable law must be determined under the rules on the applicable law in the absence of choice.404

(ii)  The applicable law in the absence of choice

In a surprising number of cases the parties fail to choose the applicable law. This may be because they have contracted without first consulting lawyers, or they cannot agree on the applicable law, or because the parties’ choice has been ineffective.405 Today, the determination of the applicable law in the absence of choice is governed by Article 4 of the Regulation which applies a purely objective test. Therefore, it is inappropriate to talk in this context about the intentions of the parties.406

(a)  The applicable law in the absence of choice under the common law

Under the traditional proper law of the contract approach, in the absence of an express or inferred choice, the court looked for the system of law with which the transaction was most closely connected. This took into account such factors as the place of residence407 or business408 of the parties, the place where the relationship between the parties was centred,409 the place where the contract was made410 or was to be performed,411 or the nature and subject matter412 of the contract. Nonetheless, the common law rule was not as open ended and flexible as might at first appear. In order to promote certainty in the law the courts identified specific factors as having great weight in identifying the closest connection in relation to certain contracts. For example, for insurance contracts the objective proper law would normally be the law of the state where the insurer carried on business.413 There can be little doubt that when the English courts set out to ascertain the objective proper law of the contract they also sought to achieve certain underlying policy objectives, such as giving business efficacy to the contract.414

(b)  The applicable law in the absence of choice under the Rome Convention

After the implementation of the Rome Convention415 into English law, the determination of the applicable law in the absence of choice was dealt with under Article 4 of this Convention, (p. 725) which consisted of three main parts. First, there was the basic rule that the contract shall be governed by the law of the country with which it is most closely connected.416 Secondly, there was a general presumption, based on the concept of characteristic performance, designed to identify the country with which the contract is most closely connected,417 together with special presumptions418 for two particular types of contract.419 Thirdly, there was a provision which, inter alia, stated that the presumptions shall be disregarded if the contract appeared to be more closely connected with another country.420 These provisions sought to combine certainty, provided by the presumptions, with flexibility, provided by the more closely connected test and the power to rebut the presumptions. Nonetheless, the scheme of Article 4 of the Convention raised an initial dilemma: It was not clear whether what was intended is a three-, two- or even one-stage process.421 The sequence in which the provisions were set out in Article 4 pointed to a three-stage process.422 However, the Giuliano and Lagarde Report423 contemplated that the applicable law can be determined solely by applying the presumptions, without searching for the country with the closest connection: it thus envisaged, at least in many cases, a one-stage process which starts and finishes with the presumptions. This view was supported by Continental case law.424 Against this, it was hard to see how a court can decide whether it is appropriate to rebut a presumption unless it has first applied the more closely connected test. This would suggest a two-stage process, which starts with a presumption, but then moves on425 to consider the more closely connected test in order to see whether this presumption can be rebutted. This view was supported by English case law426 and, arguably, by a decision of the Court of Justice which was rendered after the adoption of the Rome I Regulation.427 The question of whether it was normally a one-stage or two-stage process became particularly important when it came to the question of how easy it is to displace the general presumption.

After the adoption of the Convention, some of the factors relevant under the proper law of the contract approach remained relevant under Article 4,428 while the relevance of others, such as the place of performance, declined.429 Also, in contrast to the proper law approach, (p. 726) it was possible to take account of factors which have supervened after the conclusion of the contract.430 Moreover, the presence of the presumptions and the fact that the law was codified reduced the flexibility, and thus the room for manoeuvre, for a court that wants to achieve certain policy objectives under the Convention. The system of presumptions under the Convention was also criticised for its complexity, which compromised the desired certainty, and for the fact that it elevated one connection (combination of habitual residence and characteristic performance) to a position of importance above all others.431

(c)  The applicable law in the absence of choice under the Rome I Regulation432

Article 4 of the Rome I Regulation sets out one of the most radical departures from the position under the Convention.

(i)  Article 4(1): Fixed rules for specific contracts

In order to promote legal certainty and foreseeability of the applicable law, it is provided in Article 4(1) that, where the law applicable to the contract has not been chosen by the parties, there should be a list of eight fixed rules dealing with many of the most commonly encountered contracts.433 Together with the special provisions for contracts of carriage, insurance contracts, consumer contracts and employment contracts in Articles 5 to 8, the Regulation is much more specific than the Convention, where only two special presumptions434 and two specific rules for consumer and employment contracts existed.435 Some of the fixed rules of Article 4(1) of the Regulation, in particular (a) and (b) on sale of goods and provision of services, can be understood as expressions of the principle of characteristic performance. Moreover, the principle of characteristic performance is preserved, as a general principle, (only) for those cases where the contract cannot be categorised as being one of the specified types of Article 4(1) or where its elements fall within more than one of the specified types covered by Article 4(1).436 Other of the rules in Article 4(1) deviate from the concept of characteristic performance, either because the place of performance prevails over the habitual residence of the party effecting such performance (as in the case of immovable property, (c) and (d)),437 or in order to favour a party to a complex agreement (as in the case of distribution and franchising, (e) and (f)),438 or in order to let the law at the market prevail (as in the case of auctions and multilateral settlement systems, (g) and (h)). Still, under many of the fixed (p. 727) rules of Article 4(1) of the Regulation439 the result will be the same as that obtained applying the characteristic performance presumptions under Article 4(2) and (3) of the Rome Convention. But under Article 4(1) of the Regulation the route to the result is more direct in that it is no longer necessary to identify what the characteristic performance of a particular contract is.

Sale of goods

Under the Regulation, a contract for the sale of goods440 shall be governed by the law of the country where the seller has his habitual residence.441 The same result was reached by English and Scottish courts under Article 4(2) of the Rome Convention, holding that the characteristic performance of a contract for the sale of goods is that of the seller delivering the goods.442 Article 4(1)(a) covers also contracts where the purchaser of the goods had specified certain requirements for the goods which have yet to be produced, provided that the purchaser did not provide the raw materials.443 It does not cover the sale of land,444 the hire of goods,445 the barter of goods,446 or a pledge.447 It is noticeable that there is no attempt made to co-ordinate choice of law and jurisdiction.448 This is understandable. With a contract for the sale of goods, jurisdiction in some cases will be allocated to the buyer’s home state (as the place of delivery),449 the courts of which will apply the law of the seller’s home state. But in other cases it will be allocated to the seller’s home state (as the place of delivery). In other words, the jurisdiction is not geared to the buyer’s or seller’s home state but to the more complex concept of place of delivery (place of performance), which in some cases will be the seller’s home state and in others the buyer’s.

Provision of services

A contract for the provision of services—a concept to be defined in the same way as in Article 7(1)(b) Brussels I Recast,450 except distribution and franchising for which there are special rules451—shall be governed, under the Regulation, by the law of the country where the service provider has his habitual residence.452 Here again, courts arrived at the same result under Article 4(2) the Rome Convention by identifying the characteristic performance of a service contract as being the performance of the service provider, eg the (p. 728) insurance453 or reinsurance454 provided by the insurer, the services of a pop group,455 the provision of testing and consultancy services,456 the provision of architectural457 or diving458 services, the service of a commercial agent,459 the development of a website,460 or the service of a broker.461 All of these examples would today be considered as contracts for the provision of services under Article 4(1)(b), leading to the application of the law where the service provider has his habitual residence. Under the Regulation, the concept of service implies, at the least, that the party who provides the service carries out a particular activity in return for remuneration.462 Therefore, a license contract under which the owner of an intellectual property right grants its contractual partner the right to use that right in return for remuneration is not a contract for the provision of services, even if the licensee is obliged to use the intellectual property right.463 Under the Convention, the courts encountered some difficulty in identifying the characteristic performance of a loan contract. In most cases this will be the provision of funds by the lender, not the repayment (normally with interest) by the borrower; thus the provision of funds can be considered as the equivalent of a provision of services under Article 4(1)(b) of the Regulation.464

Immovable property

While Article 4(1)(a) and (b) of the Regulation opt for the law of the country where the seller or service provider is habitually resident, a contract relating to a right in rem in immovable property or to a tenancy of immovable property shall be governed by the law of the country where the property, ie the object of the contract, is situated, not where one party, ie the seller or landlord resides.465 Examples of contracts governed by the law of the country where the property is situated are contracts for the sale of property466 or for the rental of a holiday home.467 On the other hand, a contract for construction or repair would not fall under Article 4(1)(c) because the main subject matter of the contract is not the immovable property itself.468 The provision in Article 4(1)(c) of the Regulation on immovable property is similar to the presumption of Article (p. 729) 4(3) Rome Convention,469 with some changes. Under the Convention, it was considered unfortunate that short term leases fell under the presumption, a problem which could only be solved by applying the exception clause.470 This weakness is now remedied by Article 4(1)(d), which states that a tenancy of immovable property concluded for temporary private use for a period of no more than six consecutive months shall be governed by the law of the country where the landlord has his habitual residence, provided that the tenant is a natural person and has his habitual residence in the same country.471 Moreover, Art 4(1)(c) has been aligned with Article 24(1) of the Brussels I Recast, thus replacing the words “right to use property” by “tenancy”. This has led to doubts whether rights to use property on a different basis than a tenancy, eg usufructuary leases or agreements giving rise to equitable rights in immovable property, fall under Article 4(1)(c).472 While it may be right that such rights are no “tenancy”,473 they relate to a right in rem in the property, as both an usufruct or an equitable right to use property are effective not only against the other party to the contract, but also against third parties. They thus fall under Article 4(1)(c) of the Regulation.

Franchise and distribution contracts

Although franchise and distribution contracts are contracts for services,474 they are the subject of specific rules:475 A franchise contract shall be governed by the law of the country where the franchisee has his habitual residence,476 a distribution contract477 by the law of the country where the distributor has his habitual residence.478 These rules are based on the consideration that EU law seeks to protect the franchisee and the distributor as the weaker parties.479 Thus, these rules do not reflect the principle of characteristic performance, as can be seen from the case Print Concept GmbH v GEW (EC) Ltd480 which involved an oral distributorship agreement:

Print Concept (the distributor) was under an implied obligation to maximise sales in the German-speaking marketplace and GEW was under an implied obligation to supply its products to Print Concept which the latter would purchase.

Applying the Rome Convention, the Court of Appeal held that it was the supply of the products that was characteristic of the distribution agreement, for the penetration of the German (p. 730) market could not take place without the supply and purchase of the products.481 Therefore, according to Article 4(2) of the Rome Convention, the law of the supplier governed the contract. Under Article 4(1)(f) of the Regulation, the opposite result would be reached, as the contract would be governed by the law of the distributor.482 Likewise, for franchise contracts the characteristic performance could be regarded as that of the person granting the franchise for which the franchised person pays, leading to the application of the law of the franchisor under the characteristic performance test of Article 4(2) Rome Convention. Under Article 4(1)(e) of the Regulation, however, the law of the franchisee will apply.

Auction sales and multilateral trading systems

A contract for the sale of goods by auction shall be governed by the law of the country where the auction takes place, if such a place can be determined.483 Finally, a contract concluded within a multilateral system which brings together or facilitates the bringing together of multiple third-party buying and selling interests in financial instruments,484 in accordance with non-discretionary rules and governed by a single law, shall be governed by that law.485 Both provisions provide for the application of the law which governs the marketplace in which the contracts are concluded.

Intellectual property rights

The original proposal for a Regulation contained a rule for contracts relating to intellectual or industrial property rights, but the compromise package of the Presidency of April 2007, agreed by the Council, deleted this rule and it is not present in the final version of the Regulation. The reason for this was probably that the original proposal—general application of the law of habitual residence of the party granting the license—was considered as inappropriate in light of the diversity of license contracts, some of which might rather justify the application of the law of the licensee, or even of the country in which the intellectual property right is protected. After deletion of the special rule, license contracts—which cannot be regarded as service contracts486—will now fall under the characteristic performance test of Article 4(2), or, if no characteristic performance can be identified, under the closest connection rule of Article 4(4).

Contracts of carriage

Under the Rome Convention, contracts for the carriage of goods487 were regarded by the Working Group as having peculiarities which merited a separate special rebuttable presumption, which referred to the principal place of business of the carrier at the time the contract was concluded, provided that one of a number of alternative connections with that country must also be satisfied, eg that it is also the place of loading or discharge. If not, no presumption operated and the applicable law had to be ascertained using the closest connection test under Article 4(1) of the Convention.488 The original proposal for a (p. 731) Regulation also contained a rule under Article 4(1) for contracts of carriage,489 but this rule did not form part of the compromise package of the Presidency of April 2007 agreed by the Council, and there is no such rule under Article 4 of the final version of the Regulation. However, there is a separate Article 5 setting out a special regime for contracts of carriage. This provision distinguishes between contracts for the carriage of goods490 and contracts for the carriage of passengers.491 Article 5(1) on the carriage of goods provides a special rule only for the law applicable in the absence of choice and is thus comparable to the rules of Article 4(1), whereas Article 5(2) on the carriage of passengers also limits the choice of law.492 According to Article 5(1), the law applicable to a contract of carriage of goods in the absence of choice shall be the law of the country of habitual residence of the carrier, provided that the place of receipt or the place of delivery or the habitual residence of the consignor is also situated in that country. If those requirements are not met, the law of the country where the place of delivery as agreed by the parties is situated shall apply.493 As regards the interpretation of contracts for the carriage of goods, no change in substance is intended with respect to Article 4(4), third sentence, of the Rome Convention.494 Therefore, single-voyage charter parties and other contracts the main purpose of which is the carriage of goods should be treated as contracts for the carriage of goods. For the purposes of the Regulation, the term ‘consignor’ should refer to any person who enters into a contract of carriage with the carrier and the term ‘the carrier’ should refer to the party to the contract who undertakes to carry the goods, whether or not he performs the carriage himself.495 With contracts of carriage international conventions may apply, and these take precedence over the Regulation.496

(ii)  Article 4(2): The characteristic performance test

Article 4(2) of the Regulation retains the characteristic performance criterion of the Rome Convention497 for contracts for which Article 4(1) lays down no special rule, such as complex contracts that are not easy to categorise or contracts involving mutual performance by the parties in terms that can be regarded as characteristic on both sides.498 Article 4(2) deals thus with two situations: first, where the contract is not covered by Article 4(1); and, secondly, where the elements of the contract would be covered by more than one of the fixed rules in Article 4(1). In both situations, Article 4(2) provides that the contract shall be governed by the law of the country where the party required to effect the characteristic performance of the contract has his habitual residence.

(p. 732) The concept of characteristic performance499

The origins of the concept of characteristic performance, which is defined neither in the Convention nor in the Regulation, are to be found in Swiss private international law.500 As far as English law is concerned, the place of performance is well known as a connecting factor, but it suffers from the obvious defect that in a typical contract both parties have to perform and may have to do so in different states. The concept of characteristic performance seeks to avoid this difficulty by concentrating on just one performance, the one which is characteristic of the contract as a whole—ie the one which constitutes the essence of the contract. It is this feature of characteristic performance which was used by the Working Group on the Rome Convention to justify its elevation, above all other connections, to the position of becoming a presumption.501 More grandiosely, it was said that “the concept of characteristic performance essentially links the contract to the social and economic environment of which it will form a part”.502

Identifying the characteristic performance

This leaves the problem of identifying the characteristic performance. There is no difficulty if only one party has to perform as, for example, in the case of a contract of gift, or a unilateral contract involving payment of a termination fee on discontinuance of a project.503 But more typically, one party will perform services or provide goods, and the other will pay money for these. It is not immediately obvious which of these performances constitutes the essence of the contract. As far as each party is concerned what is important to them is the counter-performance by the other party. However, the Giuliano and Lagarde Report states504 that the characteristic performance is “of course” the performance for which the payment is due; eg the delivery of goods, the granting of a right to make use of property, the provision of a service, transport, insurance, banking operations, security, “which usually constitutes the centre of gravity and the socio-economic function of the contractual transaction”.505 Thus, for example, in an agency contract concluded in France between a Belgian commercial agent and a French company, the characteristic performance being that of the agent, the contract will be governed by Belgian law if the agent has his place of business in Belgium.506

Examples

Many of the examples discussed in the Giuliano and Lagarde Report and in the case law on the Rome Convention (sale, provision of services) fall today under Article 4(1) of the Regulation which directly determines the applicable law. The criterion of characteristic performance remains relevant only for contracts which are either not covered by Article 4(1), or where the elements of the contract would be covered by more than one of the points of Article 4(1). In both cases, in ascertaining the characteristic performance, one is not confined (p. 733) to the terms of the contract.507 The global picture must be assessed508 and the background to the contract may be of particular importance.509

An example for a contract not covered by Article 4(1) is a contract of guarantee.510

With a contract of guarantee, the characteristic performance is that of payment of money by the guarantor, whether in relation to the principal debtor or the creditor.511 Likewise, in an indemnification agreement the characteristic performance is the promise of indemnity.512 In Bank of Baroda v Vysya Bank513 Mance J regarded the position of a bank confirming a letter of credit as being analogous to that of a guarantor. The characteristic performance of the contract between the confirming bank and the issuing bank was the honouring by the confirming bank of its confirmation of credit in favour of the beneficiary.514 The characteristic performance of the contract between the issuing bank and the beneficiary was the issue of the letter of credit515 and that between the confirming bank and the beneficiary was that of the bank providing the banking service of payment under the letter of credit.516

This leaves the second category of contracts, contracts that would be covered by more than one of the points of Article 4(1). As a first observation, it should be noted that Article 4(1) is not excluded only because the contract includes a minor part which falls under another point of paragraph 1.517 If, for example, the parties agree on a lease of immovable property, and in the same contract agree that the tenant buys furniture from the landlord the value of which is minor in relation to the value of the lease, this should not lead to a disapplication of Article 4(1)(c) in favour of Article 4(2).

If the second element of the contract is not minor and entirely subordinate in relation to the first, the applicable law is determined by Article 4(2). A rather easy example is a contract where one party provides both goods and services, ie a contract where the fixed rules of Article 4(1), even if more than one applies, point in the same direction. Even if such a contract is considered to fall under Article 4(2) and not under a combination of Article 4(1)(a) and (b), the characteristic performance is effected by the party providing the goods and services, leading to the same result as a combined application of Article 4(1)(a) and (b) would.518 More difficult is the case where the different points of Article 4(1) point in different directions. In the case of such a contract consisting of a bundle of rights and obligations capable of being characterised as falling within more than one of the specified types of (p. 734) contract, the characteristic performance of the contract should be determined having regard to its centre of gravity.519

Criticism

A number of criticisms can be levelled at the concept of characteristic performance and at the way it has been defined in the Giuliano and Lagarde Report. This criticism is also relevant for the Regulation’s combination of fixed rules for specific contracts and the general characteristic performance rule, as the rules of the Regulation are less flexible than the presumptions found in the Convention, which amplifies any deficiencies in the design of these provisions.

First, there are some contracts which cannot be fitted easily within the concept. Letters of credit are an example. Article 4(2) assumes the ability to identify a single party charged with the single performance characteristic of the contract.520 But with a letter of credit there are a number of autonomous bilateral contracts and it is desirable that each contractual relationship arising in the course of the transaction has the same governing law.521 Even worse, there are some contracts that cannot be fitted at all within the concept. With a contract of barter it is difficult, if not impossible, to say that one party’s performance is more characteristic of the contract than the other’s.522 The same is true with complex contracts for the commercial exploitation of intellectual property rights.523 It was not possible to identify the characteristic performance of a trade mark agreement under which each party had to do (or refrain from doing) the same acts vis à vis the other (ie using their marks in their respective defined fields of use).524 However, the Regulation allows for this by providing in Article 4(4) that, if neither a fixed rule applies nor the characteristic performance can be determined, the law of the country with which the contract is most closely connected shall apply.

Secondly, the definition of characteristic performance in terms of the performance for which payment is due does not stand up well to close scrutiny. The payment of money was presumably rejected as the characteristic performance because this is a common feature of many contracts and therefore fails to distinguish between different types of contract. Nonetheless, there are some contracts where the payment of money is arguably the essence of the obligation, for example contracts of pledge or hire-purchase.525 The Working Group qualified their statement about the payment of money by saying that this is not usually526 the essence of the obligation. The payment of money can thus, in unusual cases, constitute the characteristic performance. For example, the characteristic performance of a reinsurance contract has been described as being the making of payment in the event of a claim.527

Thirdly, the effect of generally denying that the payment of money constitutes the essence of the contract is to favour the seller of goods over the buyer. This has been justified on the basis (p. 735) that the seller’s performance is generally more complicated and to a greater extent regulated by rules of law than that of the buyer.528 However, it is questionable whether this is sufficient to justify completely ignoring the buyer’s performance in most cases.

Fourthly, in terms of economic strength, the large enterprise, the manufacturer of goods, the provider of services (such as banks and insurance companies) and the professional is favoured against the other party who may well be in a weaker economic position.529 It is curious to find a pro-manufacturer stance being taken in a Regulation which is sufficiently concerned about protecting weaker parties to have special rules for passengers, consumers, employees and policyholders of insurance.

(iii)  Article 4(4): Closest connection as a residual rule

The criticism of Articles 4(1) and (2) reveals a possible problem in those cases where the law applicable cannot be determined under Article 4(1) or (2). This problem arises where the contract is none of those for which there is a fixed rule under Article 4(1) or where the elements of the contract would be covered by more than one of points (a) to (h) of Article 4(1)—so that Article 4(1) does not apply—and the performance of the contract which is characteristic of the contract cannot be identified under Article 4(2), so that Article 4(2) does not apply.530 The same problem arose under Article 4 of the Convention and the solution adopted under the Regulation531 is the same one, namely that the contract shall be governed by the law of the country with which it is most closely connected. In such cases there will be the usual problems that arise under an objective approach which seeks to localise the contract. The court will have to ascertain all the connections with the different countries, give weight to these, and, if they are evenly balanced between two countries, find one especially important connection which tips the balance in favour of one country.532 In order to determine that country, account should be taken, inter alia, of whether the contract in question has a very close relationship with another contract or contracts.533 Under Article 4(4) the court has to ascertain the law of the country,534 rather than the system of law with which the contract is most closely connected,535 but in most cases the difference between these two formulations will not be significant.536

(iv)  Article 4(3): Escape clause
The escape clause under the Rome Convention

Under the Rome Convention, there were different views concerning the relationship of the general presumption in Article 4(2) and (p. 736) the escape clause in Article 4(5).537 These differences related to the weight to be given to the presumption and to the significance of the place of performance by the characteristic performer when it came to evidence in favour of displacement. On one end of the scale, the attitude in some English cases was that the escape clause of Article 4(5) “formally makes the presumption very weak”.538 The presumption was displaced already if the court considered that it was not appropriate to apply it in the circumstances of the case, in particular where the place of performance differed from the place of business of the party whose performance is characteristic of the contract.539 In this situation, displacement in practice meant applying the law of the place of performance by the characteristic performer, rather than the law of the place of business of the characteristic performer.540 This struck at the heart of Article 4(2) because it favoured the place of performance over the place of business. While other decisions have given more weight to the presumption and less to the place of performance,541 English courts have been regarded as those most ready to apply the escape clause of Article 4(5). On the other end of the scale, the Dutch Supreme Court in Société Nouvelle des Papeteries de l’Aa v Machinefabriek BOA542 held that the presumption is of great weight and should only be rebutted in exceptional cases, that Article 4(2) was the main rule not Article 4(1), and that the law identified by the presumption prevailed unless it had no real significance as a connecting factor.543 The Dutch approach has been rejected by the Court of Appeal in England (p. 737) as being too rigid.544 However, it has been approved by a majority (Lords Cameron and Marnoch) in the First Division of the Court of Session in Scotland in Caledonia Subsea Ltd v Micoperi SRL.545 The presumption in Article 4(2) was said to be a strong one,546 which could only be displaced in exceptional cases.547 The place of performance was regarded as only one of a number of factors to be assessed in the exercise of Article 4(5),548 playing a subordinate role as a circumstance in fixing the applicable law.549

After the adoption of the Rome I Regulation, the Court of Justice held that the presumptions may be disregarded not only where they do not have any genuine connecting value. In view of the primary objective of Article 4 of the Convention to apply the law of the country with which it is most closely connected,550 the Court of Justice advocated a two-step approach: In the interest of foreseeability and legal certainty, the court must always start to determine the applicable law on the basis of the presumptions.551 In a second step, the court may refrain from application of the presumptions where it is clear from the circumstances as a whole that the contract is more closely connected with a country other than that identified on the basis of the presumptions.552 In this second step, the court must compare the connections existing between the contract and, on the one hand, the country in which the party who effects the characteristic performance has his or its habitual residence at the time of conclusion of the contract and, on the other, another country with which the contract is closely connected.553 In other words, the court must conduct an overall assessment of all the objective factors characterising the contractual relationship and determine which of those factors are, in its view, most significant. Significant connecting factors to be taken into account include the presence of a close connection between the contract in question with another contract or contracts which are, as the case may be, part of the same chain of contracts, and the place of delivery of the goods, being the place of performance of the obligation characteristic for the transport contract in question.554

(p. 738) The escape clause under the Rome Regulation

The diverging views under the Rome Convention, in particular the little weight given to the presumption of Article 4(2) were a concern also for the Commission in the preparation of the Rome I Regulation.555 They prompted the pendulum to swing in the opposite direction: Under the original proposal for a Regulation the rules in Article 4(1) and (2) were fixed ones with no exception which would allow the rules to be disregarded if it appears from the circumstances as a whole that the contract was more closely connected with another country. The desirability of achieving certainty in the law was given overriding importance, at the price of total inflexibility. This was subject to criticism,556 and it is welcome to see that Article 4(3) of the final Regulation contains a flexible let-out557 stating that “Where it is clear from all the circumstances of the case that the contract is manifestly more closely connected with a country other than that indicated in paragraphs 1 or 2, the law of that other country shall apply.” When applying this escape clause, account should be taken, inter alia, of whether the contract in question has a very close relationship with another contract or contracts.558 The onus is on the party relying on Article 4(3) to establish that it is a proper case for disregarding Article 4(1) or (2).559 It is not enough to show that the contract is more closely connected, it has to be “manifestly” more closely connected. The addition of the word “manifestly” (and the elevation of the criteria in Articles 4(1) and (2) to tests from mere presumptions) is doubtless designed to underscore the exceptional nature of this escape clause,560 which is intended to be a narrower one than that under Article 4(5) of the Rome Convention. This can be regarded as an acceptable compromise which reconciles the needs of certainty and flexibility.561 The English courts have thus rightly concluded that the “new language and structure suggests a higher threshold, which requires that the cumulative weight of the factors connecting the contract to another country must clearly and decisively outweigh the desideratum of certainty in applying the relevant test in Article 4.1 or 4.2”.562 In view of the difference between Convention and Regulation, the case law on the escape clause of the Rome Convention must be used with caution. What may be applicable from the earlier case law is the requirement of an overall assessment of all the objective factors characterising the contractual relationship.563 However, (p. 739) the threshold to overcome the rules of Article 4(1) and (2) of the Regulation is significantly higher than it was under the Convention.

(v)  Habitual residence

Despite the emphasis placed on the nature of the contract or the characteristic performance of it, Article 4(1) and (2) do not apply the law of the place of performance. Instead, these Articles, as other provisions in the Regulation,564 refer to the law of the country where the party who is to effect a certain performance of the contract has his habitual residence, or, in the case of a company, its central administration. Article 19(1) of the Regulation introduces a definition of habitual residence for a company. This provides that, for the purposes of the Regulation, “the habitual residence of companies and other bodies, corporate or unincorporated, shall be the place of central administration”. Provision is then made for the situations where the contract is concluded in the course of the operations of a branch, agency or any other establishment, or if, under the contract, performance is the responsibility of such a branch, agency or establishment. In these situations, the place where the branch, agency or any other establishment is located shall be treated as the place of habitual residence.565 Thus in a contract in relation to a bank account, performance by way of repayment of the sum deposited is effected through the branch where the account is kept. If the account is kept in England, English law will govern.566 In a contract between a bank issuing a letter of credit and another bank confirming this, if the performance characteristic of the contract (honouring the confirmation) is effected through the confirming bank’s London office, English law will govern the contract.567 In determining whether the principal place of business is to be displaced by some other place of business, it is necessary to ascertain whether “under the contract” the characteristic performance is to be effected through that other place of business. An express or implied term must require performance in this other place so that that there would be a breach of contract if it were performed through another place.568 It is not enough that the parties anticipated that performance would be effected through a place of business other than the principal place of business.569 The question whether the “performance is the responsibility” of that other place of business involves looking at the type of contract involved and if the obligation of the performer is, for example, to ship the goods then the question will be whether this is arranged by that place of business.570

There is no definition of the habitual residence of a natural person, except for the situation where such a person is acting in the course of his business activity.571 Article 19(1) provides (p. 740) that, in this situation, the habitual residence of a natural person shall be his principal place of business. The idea of giving also an autonomous definition to the concept of the habitual residence of a company echoes the defining of the domicile of a company under the Brussels I Recast. The intention is to produce certainty in the law. It was not possible to adopt the same definition for contract choice of law as has been adopted for jurisdiction. It will be recalled that the latter definition is in terms of three alternatives.572 Whilst it is fine for jurisdictional purposes for a company to be located in three different states, for choice of law purposes the company must be located in just one state; there cannot be three different laws applying.573 The definition of the habitual residence of a company in terms of its place of central administration follows that contained in the Rome II Regulation on non-contractual obligations which again builds on the concept of central administration found in the Brussels I Recast.574 When determining habitual residence, the relevant point in time is that of the conclusion of the contract,575 because of the possibility of changes in this connecting factor. It is noticeable that, whilst provision is made for the situation where characteristic performance cannot be ascertained, there is no corresponding provision dealing with the situation where the habitual residence, etc cannot be ascertained.576

(vi)  Severing the contract577

As has been seen,578 the parties have the freedom to sever the contract when exercising their choice as to the applicable law. Under the Convention, Article 4(1) last sentence stated that the courts have the same power when determining the applicable law in the absence of choice. However, the power of severance was to be exercised by the courts “as seldom as possible”.579 The Giuliano and Lagarde Report gave, as examples, joint ventures and complex contracts.580 Under the Regulation, Article 4—in contrast to Article 3—no longer contemplates that different parts of the contract may be governed by different laws, which seems to suggest that severance is no longer possible in the absence of choice of law.

(iii)  Special contracts

In general, the Regulation applies the same rules regardless of the type of contract involved. Nonetheless, it acknowledges that particular contracts can produce special problems. More particularly, there are two groups of special rules in the Regulation. As has been seen,581 the special rules in Article 4(1) for sale of goods, provision of services, immovable property,582 franchise and distribution contracts, sale of goods by auction, and contracts concluded in a multilateral trading (p. 741) system and in Article 5(1) for carriage of goods are only concerned with ascertaining the applicable law in the absence of choice. When it comes to carriage of passengers,583 consumer contracts,584 insurance contracts,585 and individual employment contracts,586 the Regulation goes further and, in Articles 5(2)–8, introduces special choice of law rules which are concerned with both choice of law and the applicable law in the absence of choice.

(a)  Consumer and employment contracts

The aim of the special rules for consumer587 and employment contracts588 is to provide protection for the consumer and the employee.589 The way this is done is by, first, ensuring that the protection given to the consumer or employee by provisions in domestic law that cannot be derogated from by agreement590 is not thwarted by the parties’ choice of the applicable law.591 Secondly, in the absence of choice pre-eminence is given to the law of the consumer’s habitual residence and the employee’s habitual place for carrying out work in performance of the contract.592 Finally, the rules in the Regulation dealing with the issue of formal validity make special provision for consumer contracts593 and contracts for immovable property.594 In general, the Rome I Regulation retained the means for providing protection (p. 742) for consumers and employees which were found in the Rome Convention.595 Some of the terminology in relation to consumer contracts has been altered to bring this into line with the consumer jurisdiction agreement rules under the Brussels I Recast, the idea being that this is better suited to e-commerce.596 The Regulation also excludes a number of additional contracts from the consumer protection provisions.597 As regards individual employment contracts, the law has been amplified to deal with personnel working on-board aircraft598 and to provide guidance as to whether an employee posted abroad is temporarily employed there.599

(b)  Contracts of carriage and insurance contracts

More radically, Article 5 of the Rome I Regulation introduces special rules for contracts of carriage. The provision has separate rules for carriage of goods and for carriage of passengers. The choice of law rule for contracts for the carriage of goods (Article 5(1)) is in substance very similar to that contained in Article 4(4) of the Rome Convention;600 as its predecessor, it applies only in the absence of choice of law by the parties.601 What is new is the introduction of a special choice of law rule for carriage of passengers (Article 5(2)).602 This restricts the parties’ freedom to choose the applicable law to five designated countries, one of which is the country where the passenger has his habitual residence. In the absence of choice, the applicable law will normally be the law of the country where the passenger has his habitual residence. Where it is clear from all the circumstances of the case that the contract, in the absence of a choice of law, is manifestly more closely connected with a country other than that indicated in paragraphs 1 or 2, the law of that other country will apply.603

The most radical feature of the Regulation,604 at least as regards special rules, is the introduction in Article 7605 of special rules for insurance contracts. These special rules are intended to replace those in various insurance directives606 and essentially reflect the position under those directives. This has the advantage of the relevant choice of law rules being contained in one instrument. Article 7 applies to insurance contracts covering large risks, whether or not the risk covered is situated in a Member State, and to all other insurance contracts covering risks situated inside the territory of the Member States; it does not apply to reinsurance (p. 743) contracts.607 Insurance contracts falling outside the special rules of Article 7 which are not excluded from the scope of the Regulation608 will be governed by the general choice of law rules under Articles 3 and 4 of the Regulation. The general choice of law rules will also apply to reinsurance. The choice of law rules in Article 7 are complex and fall outside the scope of this book, which has previously left insurance to more specialised works.

(d)  Limitations on the dominance of the applicable law

(i)  Mandatory rules609

The concept of mandatory rules was one of the key concepts under the Convention, with no fewer than six different provisions using it.610 Under the Regulation, the terminology has been changed to distinguish between “provisions that cannot be derogated from by agreement” in Articles 3(3), 3(4), 6(2) and 8(1) of the Regulation and “overriding mandatory provisions” under Article 9 of the Regulation,611 a distinction which existed in substance already under the Convention. The concept of mandatory rules is a particularly difficult concept for English lawyers to apply because it is not known under English law, at least not under that name. Before turning to look at the definition of the concept and at the provisions in which the concept is used, it is important to look at the background to its introduction into the Convention.

(a)  The background to the concept

In domestic contract law there are now two very different sorts of rules.612 There are the traditional rules which are concerned with settling disputes between parties, such as the rules on consideration. There are then the more modern rules which are concerned with protecting a group of persons or the national economic system—rules that arise as the result of state interference with contracts. The concept of mandatory rules only deals with this second class of rules. Consumers and employees provide good examples of groups of persons who are given special protection under the law. There are rules controlling exemption clauses and laying down requirements in relation to hire purchase and consumer credit transactions which are designed to protect consumers. There are also rules on industrial safety and hygiene and in relation to periods of notice for dismissal which are designed to protect employees. When it comes to protecting the national economic system, rules on monopolies, anti-trust, import and export prohibitions, price controls, exchange control legislation,613 and the regulation of estate agents614 are all designed to serve this purpose.

(p. 744) A state’s interest in upholding protectionist laws may be so strong that it prohibits the parties from contracting out of such rules in a domestic situation. Going beyond this and into the realms of private international law, the state’s interest in upholding certain laws may dictate that those laws must apply even though the issue is, in principle, governed by a different law selected by contract choice of law rules. An exception to the normal choice of law rules is thereby created. This sort of exception was well known in a number of Member States before the Rome Convention. The Giuliano and Lagarde Report615 referred to the Dutch decision in the Alnati case616 which is a predecessor of Article 7(1) of the Convention (the general provision dealing with mandatory rules of a foreign country).617 In Alnati, the Dutch Supreme Court said that there could be cases when the interest of a foreign state in having its law applied outside its territory was so great that the Dutch courts should take this into account and give priority to the application of such provisions in preference to the law of another state chosen by the parties. In England, the Unfair Contract Terms Act 1977 with its provisions restricting the parties’ freedom to choose the applicable law618 was considered as an example of what in English law is called an overriding statute,619 ie a statute that overrides normal choice of law rules so as to apply the rules in the statute, even though the contract is not governed by English law. Thus the parties may have chosen French law to govern the contract, but in certain circumstances the controls on exemption clauses contained in the 1977 Act will still apply. The Working Group was concerned to retain this sort of exception in the Rome Convention.

(b)  The definition of (overriding) mandatory rules

(i)  The definition of mandatory rules under the Rome Convention620

Under the Rome Convention, mandatory rules were defined in Article 3(3) as rules “which cannot be derogated from by contract”. At first sight, this definition seemed to apply universally to all of the Convention’s six provisions on mandatory rules,621 an impression that was in fact untrue, because in substance there were two different forms of mandatory rules found in the Convention.622

First, there was the basic wide type of mandatory rule, where all that has to be shown is that the definition of a mandatory rule is satisfied, ie under the law of the country with whose rule one is concerned, the rule cannot be derogated from by contract.623 Articles 3(3),624 5(2),625 and 6(1)626 of the Convention were all concerned with this basic wide type of mandatory rule. These mandatory rules operated merely as a limitation on the freedom to choose (p. 745) the applicable law, namely in purely internal (domestic) cases, as well as in consumer and employment contracts. Here, the effect given to the mandatory rule under the Convention was merely to override the parties’ freedom to choose the applicable law.

Secondly, there were the narrower overriding mandatory rules. With these, it must not only be shown that the rule is a mandatory one, within the above definition of being non-derogable, but also that under the law of the country with whose rules you are concerned, the mandatory rule overrides the applicable law. Articles 7(1), 7(2)627 and 9(6)628 all required this additional element.629 Naturally, a rule falling within this narrower type of mandatory rule will automatically come within the wider type as well. If the parties’ right expressly to choose the governing law is taken away, it follows that the domestic rule is one that cannot be contracted out of. With mandatory rules under Articles 7(1), 7(2) and 9(6) the effect given to the rule under the Convention was much greater: the mandatory rule was able to override all of the rules on the applicable law under the Convention (including the rules on the applicable law in the absence of choice).

(ii)  The definition under the Rome I Regulation

Using the same expression for two very different concepts caused confusion. In order to avoid such confusion, the Rome I Regulation abandoned the term “mandatory rules” altogether. Instead, different terms were introduced for the two different types of “mandatory rules”. For the first category, Article 3(3) of the Regulation (as do Articles 3(4), 6(2), 8(1), 11(5) of the Regulation)630 refers now merely “to provisions which cannot be derogated from by agreement”.631 In order to determine whether rules of a particular country are “provisions which cannot be derogated from by agreement”, reference must be made to the law of that country. For example, if an English court is concerned to ascertain whether a French domestic rule is a mandatory one, it has to ask whether under French law that particular rule cannot be derogated from by contract. This definition is asking simply whether the parties can agree to depart from the rule in domestic cases.

For the second category, which is relevant only for the purposes of Article 9, Article 9(1) of the Regulation introduces the term “overriding mandatory provisions”. This term has to be distinguished from the expression “provisions which cannot be derogated from by agreement” in that it should be construed more restrictively.632 Thus, all “overriding mandatory (p. 746) provisions” will be “provisions which cannot be derogated from by agreement”, but it requires something more to qualify as “overriding mandatory”: the provision must not only be non-derogable under a country’s law, but also be regarded as crucial by that country for safeguarding its public interests. Article 9(1) of the Regulation defines overriding mandatory provisions as “provisions the respect for which is regarded as crucial by a country for safeguarding its public interests, such as its political, social or economic organisation, to such an extent that they are applicable to any situation falling within their scope, irrespective of the law otherwise applicable to the contract under this Regulation”. This definition is inspired by the decision of the Court of Justice in Arblade,633 which was concerned with the relationship between national mandatory provisions of Member States and the EC Treaty. In order to respect the choice of law made by the parties under Article 3, the concept of “overriding mandatory provision” in Article 9 must be interpreted strictly.634 It is thus the task of the national court, in the course of its assessment of whether the national law in question is an “overriding mandatory provision”, to take account not only of the exact terms of that law, but also of its general structure and of all the circumstances in which that law was adopted in order to determine whether it is mandatory in nature in so far as it appears that the legislature adopted it in order to protect an interest judged to be essential by the Member State concerned.635 The definition in Article 9(1) brings out the protectionist nature of overriding mandatory provisions. The danger of defining what these provisions are protecting is that this may restrict the use of the concept. In particular, the definition will exclude provisions which aim to protect purely private interests. The other feature that is brought out in Article 9(1) is the overriding nature of such provisions. They are applicable “irrespective of the law otherwise applicable to the contract under this Regulation” because the country regarding them as crucial requires them to be applied. This underlines the private international question of whether the rule is intended, according to the law of its country of origin, to apply regardless of the otherwise governing law.

(c)  The mandatory rules of the forum

Article 9(2) states that: “Nothing in this Regulation shall restrict the application of the overriding mandatory rules of the law of the forum”.636 The opening words of Article 9(2) make it clear that this provision was inserted so that the forum could continue to apply its own mandatory rules to override contract choice of law rules even after the new regime under the Rome Convention (now Rome I Regulation) entered into force.637 The Giuliano and Lagarde Report638 gives some examples of the sort of protectionist domestic rules which Contracting States were anxious to preserve as overriding rules: rules on cartels, competition and restrictive practices, consumer protection and certain rules concerning carriage. Article 9(2) leads to no difficulty in identifying the country whose mandatory rules are in issue: they are solely the mandatory rules of the forum. However, it is not enough merely to show that the forum has a mandatory rule (ie a rule which cannot be derogated from by contract). It has to be shown that what is involved is an overriding mandatory provision in the narrow meaning of Article 9(1).639

(p. 747) (i)  Mandatory statutory rules

Assuming that England is the forum, English rules expressing a strong socio-economic policy are more likely to be contained in statutes than in common law rules. Moreover, when it comes to identifying such rules this may be much easier with statutory rules than with common law rules. Unlike a common law rule, a statute may state whether it is intended to have overriding effect.640 The statute could expressly provide that in certain situations it is to have complete overriding effect,641 it is to have limited overriding effect, or it is to have no overriding effect. An example for a statute with limited overriding effect is—arguably642—the Consumer Rights Act 2015.643 Pursuant to sections 32(1) and 74(1), most provisions of Chapter 2 of Part 1 (on contracts to supply goods to a consumer) and those of Part 2 (on unfair terms in consumer contracts) of this Act must be applied by an English court, irrespective of the parties’ choice of law, if the law of a country or territory other than an EEA State is chosen by the parties to be applicable to a consumer contract, but the consumer contract has a close connection with the United Kingdom.644 Thus, the overriding effect is limited to scenarios where the parties have made a choice of law; it does not apply where the applicable law is determined in the absence of choice. Alternatively, the statute may say nothing about its overriding effect, in which case it is a matter of statutory construction as to whether, in particular situations, it is intended to have overriding effect. This latter, most difficult possibility will now be examined.

The statute has no express provision on its overriding effect but has a provision on its territorial scope

In cases where a statute says nothing about its overriding effect it is a matter of construction of the statute to ascertain whether it is intended to have overriding effect in a particular situation.645 Drawing this inference is much easier if the statute has a provision dealing with its territorial scope, albeit whilst not spelling out explicitly whether it is intended to have overriding effect. This can be illustrated by Boissevain v Weil.646 The case concerned a Defence Regulation made under the powers conferred by the Emergency Powers (Defence) Act 1939. This regulation made it an offence, subject to severe penalties, for a British subject647 to carry out certain currency transactions. In the House of Lords, Lord Radcliffe declared that whether such an offence was committed could not depend on whether the law governing a loan contract was English or foreign. In other words, it was a regulation which, in the situation in the case, was construed as being of complete overriding effect. Article (p. 748) 9(2) would now apply, and the result would be the same under the Regulation. Similarly, in Chiron Corpn v Organon Teknika (No 2),648 the Court of Appeal held that (former) section 44 of the Patents Act 1977, sub-section (1) of which stroke down certain provisions (constituting an abuse of patent rights) in contracts relating to patents, applied regardless of whether English law governed the contract.649 Section 44 was territorially limited to United Kingdom patents. The court was not prepared, in the light of this important restriction, to add on a further requirement that the contract had to be governed by English law.650 Again, the result would be the same under the Regulation by virtue of Article 9(2).

The statute has no express provision as to its overriding effect and no provision on its territorial scope

Many statutes say nothing about their extra-territorial scope or their overriding effect. The construction of such a statute to ascertain whether it is intended to have an overriding effect in a particular situation is especially difficult. In English v Donnelly651 a Scottish court gave at least limited overriding effect to Scottish mandatory hire purchase requirements, with the result that the parties were not allowed to contract out of the statute in question by an express choice of law clause. In Irish Shipping Ltd v Commercial Union Assurance Co plc652 Staughton LJ showed some concern that “The intention of Parliament could be frustrated if it were open to the parties to a contract of insurance to exclude the operation of section 1 [of the Third Parties (Rights against Insurers) Act 1930] by choosing a foreign proper law”. On the other hand, in Sayers v International Drilling Co NV653 the Court of Appeal did not give overriding effect to the Law Reform (Personal Injuries) Act 1948, which is concerned to protect injured employees from clauses exempting the employer from liability.654 A majority of the Court of Appeal held that the objectively determined proper law of the contract—Dutch law—applied, according to which the exemption clause was valid.

(ii)  Mandatory common law rules

Whilst in principle it is possible to have mandatory common law rules, examples are likely to be very rare because of the difficulty in identifying such rules.655 It has been suggested656 that in some of the cases decided at common law where the English courts have refused to apply a foreign law on grounds of public policy and have instead applied English law, eg in those where the fundamental concepts of English justice were disregarded,657 public policy operates in a positive way, the concern being to apply English law, and, accordingly such cases should be regarded as involving an overriding mandatory provision of the forum. In other cases, it is admitted that public policy operates in a negative way, the concern being not to apply an objectionable foreign law (the effect is then that English law will apply), and it is rightly said that in such cases Article 21 of the Regulation658 will apply rather than Article 9(2). However, it is submitted that public policy primarily operates in a negative way659 and (p. 749) it is extremely difficult to identify with any precision cases where public policy operates in a positive way,660 given that the consequence is that the foreign proper law was not applied. In truth, each case of public policy has a positive and negative aspect but the latter dominates. The better approach is to treat all of these cases of public policy at common law, where the foreign proper law has not been applied and instead English law has been applied, as operating in a negative way and therefore coming not within Article 9(2) but within Article 21.661

(iii)  Mandatory rules in EU legislation and international conventions

It is not only national rules that can be mandatory. So can provisions of EU legislation.662 This was established in Ingmar GB Ltd v Eaton Leonard Technologies Inc,663 where the Court of Justice held that Articles 17 and 18 of the self-employed commercial agents Directive, which guarantee certain rights to commercial agents after termination of agency contracts,664 were overriding mandatory provisions665 in the sense used in Article 9(2).666 These Articles applied even though the expressly chosen applicable law was that of California. The Court reached this conclusion in the light of the purpose of the Directive and of these Articles and in the light of Article 19, which provides that the parties may not derogate from Articles 17 and 18 to the detriment of the commercial agent. There was no express provision in the Directive on the territorial scope of these two provisions. However, looking at their purpose the Court held that Articles 17 and 18 applied where the situation was closely connected with the EU, in particular where the commercial agent carries on his activity in the territory of a Member State. These Articles would therefore apply even though the principal was established in a non-Member State country, California. On the facts, the Rome Convention did not come into play because the contract was concluded in 1989. However, it seems that the result would have been the same if the contract had been made after the Convention (and now the Regulation) came into force. The mandatory rules in the Directive667 would then presumably have applied by virtue of Article 9(2) of the Regulation.668 Another example of overriding mandatory provisions are the provisions of EU competition law.669

(p. 750) It should be noted that Member States may also go beyond an EU directive and enact overriding mandatory provisions by extending the scope of a directive or by choosing to make wider use of the discretion afforded by a directive, eg by offering greater protection to commercial agents than the commercial agents Directive provides for.670 However, in deciding whether such a national extension of a directive qualifies as an overriding mandatory provision in the sense of Article 9(1) of the Regulation, the court must take into account that the law which is to be rejected in favour of the law of the forum is the law of a Member State which had transposed the underlying directive, so that a certain level of protection is already guaranteed.671

Finally, international conventions may also include overriding mandatory provisions. For example, the United Kingdom has entered into a number of conventions in relation to carriage. It is believed that the uniform rules which these conventions contain will, in general, be regarded as overriding mandatory provisions of the law of the forum and take effect under Article 9(2) of the Regulation.672 However, the precedence of the uniform rules contained in these conventions over the law applicable under the Regulation could better be justified by applying Article 25 of the Regulation. While it is true that Article 25 refers only to international conventions “which lay down conflict-of-law rules relating to contractual obligations”, the rules of the conventions which define their substantive, personal and territorial scope could be regarded as conflict-of-law rules in a wide sense. Such an approach would avoid applying the narrow definition of Article 9(1) to all mandatory rules in international conventions. The true reason for giving precedence to such rules is to safeguard the uniform application of the convention in all Contracting States, and this would be reflected by applying Article 25, not Article 9 of the Regulation.673

(iv)  The effect given to mandatory rules under Article 9(2)674

Article 9(2) operates as an exception to the normal choice of law rules under the Regulation by giving overriding effect to mandatory rules. Obviously it overrides the rules on the applicable law under Article 3 (choice) and Article 4 (the applicable law in the absence of choice). But Article 9(2) probably has an even wider effect than this. The opening wording of Article 9(2) is that: “Nothing in this Regulation shall restrict the application of the rules of the law of the forum.” This suggests that Article 9(2) should be regarded as a general exception to all the choice of law rules contained in the Regulation, in the same way that public policy, which is more clearly worded in this respect, provides such an exception.675 This means that the special rules for particular issues, such as formal validity, and for special contracts will also be overriden.676 However, there is a particular problem in relation to consumer contracts and individual employment contracts, in that there are provisions on mandatory rules contained in the special regimes set out in Articles 6 and 8. Nevertheless, Article 9(2) should be regarded as overriding the other provisions in the Regulation on mandatory rules. The result (p. 751) is that English mandatory rules would take priority over foreign mandatory rules. This has wide implications. For example, in a case where the applicable law under Article 4 of the Regulation may be French and English law has mandatory rules, then, in principle, Article 9(2) should operate so that the English mandatory rules override the French ones. However, it must be for English law to decide whether its own mandatory rules are of such importance that they should exceptionally apply in such a case. In all cases, the overriding effect is limited to the extent required by the mandatory provision. Thus, those aspects of the contract which have nothing to do with the mandatory provision, or areas of the law in which the mandatory provision does not purport to intrude, remain to be governed by the law as determined under Articles 3 to 8 of the Regulation.677

(d)  The mandatory rules of other countries678

Article 9(3) of the Regulation679 provides that:

Effect may be given to the overriding mandatory provisions of the law of the country where the obligations arising out of the contract have to be or have been performed, in so far as those overriding mandatory provisions render the performance of the contract unlawful. In considering whether to give effect to those provisions, regard shall be had to their nature and purpose and to the consequences of their application or non-application.

The background to this rule is that reservations by Member States in relation to particular Articles, which existed in relation to the corresponding Article 7(1) of the Convention, are in principle incompatible with a Regulation.680 This means that if the United Kingdom were to opt in to the Regulation, its courts would for the first time have to consider giving effect to the mandatory provisions of other countries in an Article 9(3) situation. This was the most important single objection that the United Kingdom had to opting in to the Regulation. It was objected that the introduction in the United Kingdom of an Article 7(1) of the Rome Convention type of rule would create uncertainty,681 in particular in the financial markets.682 There has been an attempt to meet such concern by limiting the new rule in the final version of the Regulation in two major respects. First, Article 9(3) of the Regulation is only concerned with the overriding mandatory provisions of the law of “the country where the obligations arising out of the contract have to be or have been performed”. This was perceived as being much more certain than the wording of Article 7(1) of the Rome Convention, with its reference to the mandatory rules of law of another country with which the situation has a close connection. And indeed, the Court of Justice has confirmed that the list, in Article 9(2) and (3) of the Rome I Regulation, of the overriding mandatory provisions to which the court of the forum may give effect is exhaustive.683 As a consequence, overriding mandatory provisions of another country than the forum state (Article 9(2)) or the state where the obligations have to be or have been performed (Article 9(3)) cannot be applied under the Regulation. (p. 752) What remains possible is to take into account overriding mandatory provisions of a state (other than the forum state or the state of performance) as a matter of fact, in so far as this is provided for by a substantive rule of the law that is applicable to the contract pursuant to the Regulation.684 For example, it may be taken into account on the level of substantive contract law, eg under the doctrine of illegality or the doctrine of impossibility, that the contract violates an overriding mandatory provision of the law of a country other than the forum state or the state where performance is to take place. This, as a matter of the applicable substantive law, is not precluded by the Regulation. Secondly, Article 9(3) is only concerned with that country’s overriding mandatory provisions in so far as they render the performance of the contract unlawful. Article 9(3) thus enables the English courts to deal with cases of illegality by the foreign place of performance. Such cases were a concern at common law and, because of the absence in the United Kingdom of a provision on foreign mandatory rules, were not effectively dealt with by the Rome Convention.685

(ii)  Public policy686

Article 21 of the Regulation687 is entitled “Public policy of the forum” and provides that the “application of a provision of the law of any country specified by this Regulation may be refused only if such application is manifestly incompatible with the public policy (ordre public) of the forum”. In civil law countries ordre public operates as a well-established exception to normal choice of law rules, as does public policy in common law jurisdictions. Any clash between the civil and common law concepts of public policy is resolved by the reference to the application of a rule of law being “manifestly” incompatible with the public policy of the forum. This word has been regularly used in Hague Conventions on private international law in an attempt to restrain the use of the doctrine. This adds nothing as far as English law is concerned since there has long been a reluctance to invoke the public policy doctrine in this country.688 However, for civil lawyers it makes clear that what is in issue is the narrow concept of international ordre public as opposed to the wide concept of domestic ordre public. It has to be shown that the application of a foreign rule of law is against the forum’s public policy. The circumstances of the case have to be considered. If, for example, a contract governed by French law restrains a party from competing in England then the application of a French rule allowing restraint of trade would appear to be contrary to the well-known English public policy against restraint of trade because of the involvement of England. This limitation is entirely consistent with common law decisions on public policy, which normally have required some relevant connection with England which justifies English courts in invoking the public policy exception.689 The intention then is that Article 21 will only be used in exceptional circumstances.690 The forum’s public policy includes also the fundamental rules (p. 753) of EU law, such as the European human rights.691 Where the alleged offense to public policy concerns rules of EU law, this does not alter the conditions for being able to rely on the clause on public policy: It is for the national court to ensure with equal diligence the protection of rights established in national law and rights conferred by EU law.692

When Article 21 does apply, it provides an exception to all of the preceding choice of law rules contained in the Regulation. Presumably, it can even operate to override the provisions on mandatory rules which are themselves an exception to normal choice of law principles.693 Thus, an English court could refuse to apply the mandatory rules of a foreign country on the basis that the application of that mandatory rule would be against English public policy. It is possible to envisage this happening in cases where a foreign mandatory rule is in conflict with an English mandatory rule.

Common law cases694 which involved the consideration of an objectionable foreign law which was held to be contrary to our distinctive English public policy will doubtless be decided in the same way under the Regulation, using Article 21.695 Thus the English court will continue not to enforce such contracts as those in restraint of trade, assigning a cause of action, involving a certain type of duress,696 or contracts for prostitution, provided normally697 that the circumstances involve a sufficient connection with England to justify this.698 However, there was a very different category of public policy cases at common law based on the notion of the comity of nations. Thus, in one case699 the Court of Appeal held that it would be against the comity of nations to enforce a contract, the whole object of which was to import whisky into the USA contrary to the prohibition laws of that country.700 Such cases do not appear to fit within Article 21.701 This is a negative provision, being concerned with a refusal to apply (p. 754) some objectionable foreign rule.702 If the English forum’s concern, as appears to be the case in the above example, is to uphold a foreign law703 on a matter of great importance to that foreign country, this does not look to come within either the wording or purpose of Article 21. The appropriate provision to refer to under the Regulation is Article 9(3).704 Under this provision, effect may be given to the overriding mandatory provisions of the place of performance in the situation where performance is rendered unlawful in that country. If, however, the place of performance is not in the country of prohibition, Article 9(3) will not apply.705 What remains possible is to take into account overriding mandatory provisions of a state other than the state of the forum or the state of performance as a matter of fact, in so far as this is provided for by a substantive rule of the law that is applicable to the contract pursuant to the Regulation.706 Therefore, the foreign prohibition could be taken into account in the context of the English domestic rule on the invalidity of contracts made with the object of breaking the laws of a foreign country. This leaves a gap only for the scenario where the contract is governed not by English, but by a foreign law different from the law which provides for the overriding mandatory provision. Is it possible to use Article 9(2) then? It has been suggested that such cases involve a domestic rule of English law which can be regarded as an overriding mandatory rule of English law within this provision.707 However, to regard a common law rule of public policy based on the comity of nations, and which therefore only applies in international cases, as a provision the respect for which is regarded as crucial by England for safeguarding its public interests would appear to be a fiction. It looks more like the enforcement of a third country’s overriding mandatory provision, and, as such, is implicitly abolished by the Rome Regulation in all cases where the latter applies and does not allow such enforcement.708

(e)  Particular issues

The scheme of the Regulation is that, having set out the rules on the applicable law, there are then special rules dealing with the particular issues of consent and material validity,709 formal (p. 755) validity,710 incapacity711 and a number of other matters.712 By implication, all other issues must be governed by the rules on the applicable law set out earlier in the Regulation. The point is spelt out by a provision on the scope of the applicable law713 which gives particular instances of issues governed by the rules on the applicable law. With special rules for particular issues, a problem of classification inevitably arises. Under the common law rules, the English courts adopted the system of classification employed in the domestic law of contract. However, under the Regulation it is important that the classification of issues is made in accordance with the principle of uniform interpretation714 and in the light of the intentions of the drafters of the Convention (as predecessor of the Regulation) and Regulation.715 For example, the issue of whether a contract has to be in writing is intended to come within the category of formal validity716 under the Regulation, whereas this was classified as a procedural issue under the common law rules. The provisions in the Regulation relating to particular issues will now be examined,717 after which one issue that causes a special problem for the United Kingdom, that of illegality, will be considered.

(i)  Consent and material validity

Article 10 of the Regulation—which718 is virtually identical in wording to Article 8 of the Convention—is entitled “Material Validity” and contains two provisions.719 Before examining these provisions it is necessary to see what is encompassed within the concept of material validity.

(a)  What is meant by consent and material validity?

Material validity under the Regulation covers a wide variety of different issues. This is apparent from both Article 10(1), which is concerned with “the existence and validity of a contract”, and Article 10(2), which deals with the existence of consent. The intention is that not only are issues of material validity in the English sense covered (eg the issue of illegality), but also issues relating to formation of the contract720 (eg offer and acceptance, and consideration). The validity of consent to the contract (eg issues of mistake, misrepresentation and duress) is doubtless also covered under Article 10(1). Issues relating to the existence and validity of the contract itself are obviously covered, but so also are such issues in relation to the terms of the contract.721 As has already been mentioned,722 the existence and validity of consent to a choice of the applicable law are to be referred to Article 10.723 Nonetheless, not (p. 756) all issues of validity are dealt with under Article 10; formal validity has a special rule to itself under Article 11.724

(b)  The putative applicable law725

Article 10(1) states that: “The existence and validity of a contract, or of any term of a contract, shall be determined by the law which would govern it under this Regulation if the contract or term were valid.” This means that the normal rules on the applicable law under the Regulation are applied to the issue of material validity. The only gloss on this is that with material validity one has to assume that the contract or term is valid in the first place before ascertaining the applicable law. In the terminology of the common law rules, the “putative” governing law is applied.726 As the Giuliano and Lagarde Report explains: “This is to avoid the circular argument that where there is a choice of the applicable law no law can be said to be applicable until the contract is found to be valid.”727 The important point of substance is that the parties are free to choose the governing law under Article 3 of the Regulation, eg by putting a choice of law clause in the contract, even though the issue in the case is whether a valid contract exists between them. When it comes to validity of terms of the contract it means that the validity of a New York choice of law clause is governed by substantive New York law, excluding its conflict of laws rules.728 Similarly, the question whether a clause providing for arbitration in England is validly incorporated into a contract is governed by English law (the arbitration clause being an inferred choice of the governing law).729 In other words, the parties are able to pull themselves up by their own bootstraps. The principle has much to commend it.730 Businessmen use choice of law clauses in order to avoid the problems of ascertaining the objective governing law and their wishes should be respected whatever the issue.731 However, this can lead to unfairness to one of the parties. The Working Group acknowledged this particular problem and included a safeguard in relation to consent to the contract.

(p. 757) (c)  The safeguard in relation to consent732

This is contained in Article 10(2) which provides that: “Nevertheless, a party, in order to establish that he did not consent, may rely upon the law of the country in which he has his habitual residence if it appears from the circumstances that it would not be reasonable to determine the effect of his conduct in accordance with the law specified in paragraph 1.” This is designed733 to cater for the following sort of example:

A makes an offer to B, and inserts a choice of law clause in the contract stating that the law of Utopia will govern all disputes between the parties. B remains silent, neither expressly accepting nor rejecting the offer. Under the law of Utopia silence can constitute an acceptance. It would be manifestly unfair for B to be contractually bound. The effect of Article 10(2) is that B can assert that he did not consent to the contract according to the law of his habitual residence.

The major proviso is that it would have to appear from the circumstances that it would not be reasonable to determine the effect of his conduct under Utopian law. On the above facts, doubtless it would not be. The circumstances to be taken into account include the parties’ previous practices inter se and their business relationship734 and whether the transaction is a conventional one.735

The burden is on the party who wishes to displace Article 10(1) to show that the terms of Article 10(2) have been met.736 Article 10(2) allows a party to rely on the law of his habitual residence to deny the existence of a contract. It cannot be used in a positive way to create a contract which did not exist under the applicable law. It was specifically devised with the question of silence constituting acceptance of an offer in mind. Nonetheless, it is wide enough to cover any issue of offer and acceptance. It is, though, only concerned with the existence of consent, not with the validity of consent (eg with duress, mistake, misrepresentation).737 In cases raising these issues Article 10(1) will no doubt apply; but, on its own, without the safeguard contained in Article 10(2). As far as the issue of consent is concerned, the combined effect of Article 10(1) and (2) is that the contract can be invalidated either by reference to the applicable law or by reference to the law of the habitual residence of the party denying that he consented.

There was an unsuccessful attempt in Egon Oldendorff v Liberia Corpn738 to use Article 10(2) in a very different situation from that for which it was designed. The Japanese defendants sought to rely on Japanese law, which allegedly requires (in an agreement “subject to details”) recapitulation and confirmation of contractual details by both parties, to establish that they (p. 758) did not consent to the incorporation of an arbitration clause into the contract. Mance J held that the onus was on the party who sought to invoke Article 10(2) to negative consent, and that reliance could only be placed on Japanese law if it was not reasonable to determine the effect of the defendants’ conduct in accordance with English law. It was unreasonable to determine consent in accordance with Japanese law because this would mean ignoring the English arbitration clause and to do this would appear to be contrary to ordinary commercial expectations when everything suggested that the defendants must already have considered and accepted the clause. The issue of consent raised by Japanese law looks to have more to do with the validity of consent than with its existence.739 Accordingly, the case could have been decided on the basis that this issue fell outside the scope of Article 10(2), without going on to apply the “reasonableness” test.

(ii)  Formal validity740

Article 11 of the Regulation deals with this issue. It contains general rules relating to the formal validity of contracts, a rule for unilateral acts intended to have legal effect (such as notice of termination of a contract), and special rules for consumer contracts and contracts in respect of immovable property. Article 11 of the Regulation is by and large the same in substance as Article 9 of the Convention.741 There is, though, one significant difference. The rules in the Convention on formal validity were regarded as being too restrictive in the light of the growing frequency of contracts made at a distance.742 In order to facilitate the formal validation of contracts, a third alternative connecting factor has been introduced in Article 11(2) and (3), namely the law of the country where either of the parties had his habitual residence at the time of conclusion.

(a)  What is meant by formal validity?

It is not always easy to decide whether a matter is one of form or substance. English lawyers have particular difficulty with the classification of issues as ones of form because of the relative dearth of formal requirements under English law. This has meant that issues have sometimes been given a surprising classification. For example, the issue of whether a contract has to be in writing looks to be one of form, yet traditionally this has been classified under English law as one of procedure, and thus to be determined by the law of the forum.743 However, it now appears that under the Regulation this issue should be classified as one of form. The Giuliano and Lagarde Report gives welcome guidance as to what formal validity encompasses. It includes “every external manifestation required on the part of a person expressing the will to be legally bound, and in the absence of which such expression of will would not be regarded as fully effective”.744 The following were given as examples of formal requirements:745 the requirement that there must be two signatures to the contract;746 that the contract must be made in duplicate; and, of most interest to English lawyers, that a (p. 759) non-competition clause in a contract of employment must be in writing. On the other hand, it was said747 that it did not include the special requirements which have to be fulfilled where an act is to be valid against third parties, eg the need in English law for a notice of a statutory assignment of a chose in action. It would be best if the concept of formal validity were to be given an independent European meaning, under which the English courts will need to take a broader view of the concept than they have in the past.

(b)  The general rules

(i)  The contract is concluded between persons who are in the same country

The first of the general rules on formal validity is contained in Article 11(1) which provides that: “A contract concluded between persons who, or whose agents, are in the same country at the time of its conclusion is formally valid if it satisfies the formal requirements of the law which governs it in substance under this Regulation or of the law of the country where it is concluded.” The policy underlying this provision is clear: to avoid the invalidation of contracts on the basis of formal defects. It does this by a validating rule of alternative reference. The normal rules on the applicable law under the Regulation are applied, but if the contract is formally invalid under those rules, as an alternative, recourse can be had to the law of the country where the contract was concluded in order to validate it.

The Working Group748 justified the reference to the law of the place where the contract was concluded on the basis of the historical importance of this law. A modern policy justification would be that disputes as to form arise at the time when and in the country where the contract is concluded. It is therefore convenient that that country’s law should be applied to resolve the dispute. There is no problem in determining where a contract was concluded because of the limitation of Article 11(1) to contracts concluded between parties who are in the same country.

When it comes to applying, as an alternative, the law that governs the contract, there are a number of difficulties. The first is that this law can be varied by the parties after the contract has been concluded. However, Article 3(2)749 provides that a subsequent variation will not prejudice the formal validity of the contract. A subsequent variation of the governing law will not be allowed to invalidate the contract. On the other hand, a subsequent variation of the governing law which has the effect of formally validating a contract, invalid at its inception, will presumably be allowed,750 validating the contract from that date. The second difficulty stems from the fact that different laws may govern different parts of the contract. Which of these governing laws is to determine its formal validity? According to the Giuliano and Lagarde Report “it would seem reasonable to apply the law applicable to the part of the contract most closely connected with the disputed condition on which its formal validity depends”.751 Thirdly, when Article 11 refers to the law that governs the contract, apparently this means the putative governing law, ie the law which would govern the contract if it were formally valid.752

(ii)  The contract is concluded between persons who are not in the same country

If the parties are not in the same country at the time of the conclusion of the contract, the second general rule, which is contained in Article 11(2), applies. According to this: “A contract concluded between persons who, or whose agents, are in different countries at the time (p. 760) of its conclusion is formally valid if it satisfies the formal requirements of the law which governs it in substance under this Regulation, or of the law of either of the countries where either of the parties or their agent is present at the time of conclusion, or of the law of the country where either of the parties had his habitual residence at that time.” Thus, under Article 11(2), a contract is formally valid if it satisfies the formal requirements of: (i) the law which governs it in substance under the Regulation, or (ii) the law of either of the countries where either of the parties or their agent is present at the time of conclusion, or (iii) the law of the country where either of the parties had his habitual residence at that time. This means that recourse may be had to the law of up to five753 different countries in order to validate the contract. The third alternative connecting factor, the law of the country where either of the parties had his habitual residence at the time of conclusion, has been introduced by the Rome I Regulation to facilitate the formal validation of contracts because the rules in the Convention on formal validity had been regarded as being too restrictive in the light of the growing frequency of contracts made at a distance.754 In the situation where offer and acceptance has been made by an exchange of e-mails, it is not easy to determine when the contract is concluded, and hence the place where each party is at that time.755 The rules in Article 11(1) and (2) set out alternatives for validating the contract, rightly without giving a priority to any one alternative.

(iii)  Acts intended to have legal effect

The third and last part of the general rules756 is concerned with formal requirements in respect of acts757 intended to have legal effect, such as an offer or notice of termination, and is analogous to Article 11(2) in that it refers, as alternatives, to the law applicable to the contract in substance under the Regulation, or the law of the country where the act was done, or the law of the country where the person by whom it was done had his habitual residence at that time. Here again the third alternative was introduced by the Rome I Regulation to facilitate the formal validation of unilateral acts.

(c)  The special rules for particular contracts

(i)  Consumer contracts

The first of the special rules is concerned with consumer contracts. The formal validity of a consumer contract is governed by the law of the country in which the consumer has his habitual residence.758 This means that, for consumer contracts,759 formal validity is governed by the law that governs the substance of the contract in the absence of choice, and the consumer is protected by having the law of his habitual residence applied.760

(ii)  Immovable property

The second special rule is concerned with contracts relating to immovable property. Article 11(5) states that:Notwithstanding paragraphs 1 to 4, a contract the subject matter of which is a right in rem in immovable property or a tenancy of immovable property shall be subject to the requirements of form of the law of the country where the property is situated if by that law

(p. 761)

  1. (a)  those requirements are imposed irrespective of the country where the contract is concluded and irrespective of the law governing the contract; and

  2. (b)  those requirements cannot be derogated from by agreement.

This provision shows a concern to give effect to the rules of form of the law of the situs in cases involving immovable property.761 It is not enough to show simply that these rules cannot be derogated from by contract. It must also be shown that the requirement of form is imposed “irrespective of the country where the contract is concluded and irrespective of the law governing the contract”. In other words, according to the law of the situs the mandatory rule has to have overriding effect.762 Different from Article 9(6) of the Convention, albeit not in substance, Article 11(5) of the Regulation no longer uses the confusing term “mandatory” requirements. Instead it spells out that in the present context we are concerned with requirements which (a) are imposed irrespective of the country where the contract is concluded and irrespective of the law governing the contracts and (b) cannot be derogated from by agreement.

(iii)  Capacity

As has already been seen, questions involving the capacity of corporations are excluded altogether from the scope of the Rome I Regulation.763 When it comes to natural persons the position is more complex. The status or legal capacity of natural persons is, in general, excluded from the scope of the Regulation.764 Member States are therefore left to apply their traditional private international law rules to the issue of capacity to contract. However, this is subject to Article 13 of the Regulation. This is a narrow rule concerned with protecting parties who have contracted with a natural person under an incapacity from being caught unawares by this. The traditional English common law rules on capacity will now be examined, and then Article 13 of the Regulation.

(a)  The traditional common law rules

What law governs capacity to make a commercial contract is a matter of speculation so far as the English common law authorities are concerned. There is no clear decision and the dicta are not very helpful. It is clear, though, that the choice lies between the law of the domicile,765 the law of the place where the contract was made766 and the proper law in the objective sense.767

It may be conceded that in modern conditions of trade domicile alone is not a satisfactory test. It is incompatible with justice and with the trust that lies at the basis of commercial dealing that, for instance, a person over eighteen years of age should be able to escape liability for the price of goods sold or delivered to him in a London shop on the ground that he is still a minor by the law of his domicile abroad.768 Indeed, under civil law systems the rule that (p. 762) capacity is governed by the personal law cannot be relied on by a person who, though lacking capacity by his personal law, has capacity according to the law of the place where the contract was made.769 Under English law, in many cases, contracts by persons under eighteen years of age are not enforceable against them,770 but the court might well restrict this rule (and it would be reasonable for it to do so) to contracts in respect of which the objective proper law is English law.771 It is also argued that, in the converse case, capacity conferred by the law of the domicile should not be invalidated by the proper law, ie a person should be regarded as capable if capable by the law of his domicile.772 So far, however, English courts have not been pressed to adopt such attitudes.

Not only has it been advocated frequently that the law of the place where the contract was made governs the question of capacity,773 but there is also one old English decision to this effect.774 This view, if it implies that the law of that place exclusively governs the matter, is clearly untenable, for it would enable a party to evade an incapacity imposed upon him by the law that governs the contract in other respects by the simple device of concluding the contract in a country where the law is more favourable. Moreover, the law of the place of contracting is ill adapted to govern the matter if, as may well happen, the parties conclude the contract in a place where they are only transiently present.

Such modern authority as there is would indicate that capacity to conclude a commercial contract is regulated by the proper law of the contract objectively ascertained. This is supported by the Canadian decision in Charron v Montreal Trust Co.775 It was held there that capacity to enter a separation agreement is to be determined by the law of the country with which the contract is most substantially connected,776 ie the proper law; though in the actual case this was also the law of the place where the contract was made.

More recently, the issue of capacity arose in Bodley Head Ltd v Flegon:777

The defendant argued, inter alia, that an agreement between the Russian author, Alexander Solzhenitsyn, and H, a Swiss lawyer, was invalid as Solzhenitsyn had no capacity under Russian law, which was both the law of the domicile and the law of the place where the contract was made, to enter a contract to appoint an agent to contract abroad on his behalf.

Whilst doubting the correctness of the allegation that Solzhenitsyn was incapable under Russian law, Brightman J had no doubt that the question of his capacity was to be decided by Swiss law as the proper law of the contract. Although the point was not discussed, the facts of this case did raise the issue that, in stating that capacity is governed by the proper law of the contract, this expression must be taken to mean the law of the country with which the contract is most substantially connected. Intention cannot here be allowed free play.778 A person cannot confer capacity upon himself by deliberately submitting himself to a law to which factually the contract is unrelated.

(p. 763) In light of the uncertain state of the common law, it has most recently been suggested to combine the law of the domicile and the proper law of the contract in an objective sense. Under this approach, “a contract will be valid if a party has capacity either under the law with which the contract is most closely connected779 or under his personal law”, ie the law of the country of his domicile.780

(b)  Article 13

Article 13 of the Rome I Regulation is entitled “Incapacity”781 and states that:

In a contract concluded between persons who are in the same country, a natural person who would have capacity under the law of that country may invoke his incapacity resulting from another law only if the other party to the contract was aware of this incapacity at the time of the conclusion of the contract or was not aware thereof as a result of negligence.

This is an unusual article in that it grafts a specific rule dealing with one aspect of capacity onto national rules of private international law on this topic. The aspect it is concerned with is the position of a party who contracts with a natural person who is under an incapacity but where the first party is unaware of this incapacity. In certain circumstances it protects such a party by imposing a limitation on the right of the natural person under the incapacity to invoke his own incapacity. This idea has its origin in the law of certain civil law countries. In order for this limitation to apply, certain stringent conditions have to be satisfied.782

First, there must be a contract concluded between persons who are in the same country. The Working Group did not want to prejudice the protection of, for example, minors when a contract was made at a distance. The person under an incapacity must be a natural person. However, there is no such requirement as regards the other party, and this could presumably be a corporation. In cases where the incapacity of a corporation is at issue the traditional common law rules will apply, and in such cases this is not subject to Article 13. Secondly, it must be a situation where, according to the traditional private international law rules applicable in the forum, a natural person has capacity under the law of the country where the contract was concluded, but lacks capacity under another law. For example, a Member State under its traditional private international law rules may apply the proper law (as—probably—in the case of England) or the law of the domicile or nationality (as in the case of some civil law countries) to the issue of capacity, and under that law a person lacks capacity. Under Article 13 it is then necessary to turn to the law of the place where the contract was concluded in order to see if there is capacity by that law. There is no difficulty in identifying the country where the contract was concluded in cases where (as will always be the case under Article 13) both parties are in the same country at the time of the conclusion of the contract. On the other hand, if a Member State applies the law of the place of contracting to the issue of capacity under its traditional private international law rules, Article 13 will not operate.

If these requirements are met, the limitation on the right of an incapacitated person to invoke his own incapacity applies. The person under the incapacity can only invoke his own incapacity if the other party was aware of this incapacity or was not aware thereof as a result of negligence. The burden of proof as to this lies on the incapacitated party.783 If satisfied, the (p. 764) incapacitated party lacks capacity to contract. On the other hand, if the incapacitated party does not satisfy the burden of proof he will have capacity to contract. The limitation is a narrow one. It only affects the rights of the person acting under the incapacity. The other party can raise an incapacity that exists according to the law applied by the traditional private international law rules of the forum even though he or she knew of the incapacity at the time of contracting. Furthermore, it only affects the rights of the person acting under the incapacity when that person is seeking to invoke his own incapacity. It does not prevent, for example, a minor from seeking to uphold a contract, and the other party cannot escape from a contract (valid by the applicable law) by saying that he was unaware that he was contracting with a minor.

(iv)  Scope of the applicable law784

Article 12 of the Rome I Regulation is entitled “Scope of the law applicable” and in paragraph (1) gives a number of examples of issues coming within the scope of the law applicable to the contract by virtue of Articles 3 to 8.785 This is not intended to be an exhaustive list,786 and it is implicit that all other issues are governed by the rules on the applicable law. The only exceptions are issues classified as ones of formal validity787 or incapacity788. Material validity is, of course, governed by the rules on the applicable law because of Article 10(1). Article 12 of the Rome I Regulation leaves the terms of its predecessor in Article 10 of the Convention largely unaltered.789 The examples provided by Article 12 are as follows:

(a)  Interpretation

Under the common law rules the province of interpretation was to discover the true intent and meaning of the parties as expressed by the language of the contract. This was a question of fact. Nevertheless, a question of choice of law could arise for, if an expression was ambiguous and if it bore different meanings in different legal systems, its interpretation had to be determined by reference to one only of those systems. This distinction between fact and law is still valid under the Regulation, since the Regulation is only concerned with choice of law. When a choice of law problem arises, the Regulation adopts the simple solution that the law applicable to the contract will govern the issue of interpretation.790 A problem which arises from this approach can be illustrated by the situation where the contract is governed by Utopian law, but the parties have expressly provided that the contract is to be interpreted according to the law of Ruritania.791 At common law it seems that Ruritanian law would govern the interpretation of the contract. Under the Regulation it appears, at first glance, that Utopian law has to be applied, since this is the law applicable to the contract. However, Article 12(1) refers to the law applicable to the contract “by virtue of this Regulation”. Under Article 3(1) the parties are able to choose a law for part of the contract which may include a specific issue such as interpretation. The law applicable to the issue of interpretation in the (p. 765) above example would accordingly be Ruritanian law, which has been expressly chosen to govern that part of the contract.

(b)  Performance

(i)  What is encompassed within the concept of performance?

Whilst interpretation of a contract is a fairly self-explanatory category, some explanation is needed of what is encompassed within the concept of performance under Article 12(1)(b). The Giuliano and Lagarde Report gives helpful examples of issues coming within Article 12:

The diligence with which the obligation must be performed; conditions relating to the place and time of performance; the extent to which the obligation can be performed by a person other than the party liable; the conditions as to performance of the obligation both in general and in relation to certain categories of obligation (joint and several obligations, alternative obligations, divisible and indivisible obligations, pecuniary obligations); where performance consists of the payment of a sum of money, the conditions relating to the discharge of the debtor who has made the payment, the appropriation of the payment, the receipt, etc.792

It is not clear whether all of these examples come within Article 12(1)(b) or whether some of them are intended to come within Article 12(1)(c) (failure to perform) or (d) (the various ways of extinguishing obligations). However, it is not necessary to decide this since the position is the same under each of the sub-paragraphs of Article 12(1): the law applicable by virtue of the Regulation governs. Nonetheless, it is important to distinguish all of these cases relating to the substance of performance from “the manner of performance and the steps to be taken in the event of defective performance” because of the special provision in Article 12(2), which deals with the latter.

(ii)  The manner of performance: a special rule

Article 12(2) provides that: “In relation to the manner of performance and the steps to be taken in the event of defective performance, regard shall be had to the law of the country in which performance takes place.” This provision deals with the situation where the law of the country of the place of performance is different from the law of the country whose law is applicable under the Regulation.793 Two questions arise in relation to Article 12(2). The first is a question of definition: what matters fall within the concept of manner of performance? The second is a question of substance: what is a court supposed to do if it is faced with a matter of the manner of performance?

The definitional question

The Working Group