1. Subject matter
The private international law (PIL) of representation determines under which national law the prerequisites and legal effects of an agency situation (see 2. below) should be assessed (conflit de lois en matière de représentation; internationales Stellvertretungsrecht). With a view to the various possible origins of the authority of agents, two main categories may be distinguished, ie voluntary or consensual agency (representation volontaire/conventionnelle; gewillkürte Stellvertretung) on the one hand and agency by operation of law (représentation légale; gesetzliche Stellvertretung) on the other. Whereas the former is based on a voluntary decision of the principal, the latter originates from objective law enabling such entities which lack capacity to act by themselves to participate in legal relations.
As regards conflict of laws, agency by operation of law follows a simple and uniform principle: it is subject to the same connecting factors (choice of law) applying to the legal relationship from which the respective authority originates in the first place. Thus, for example, agency relating to corporations is subject to the lex societatis (company law (international)), the authority of parents to the law governing the parent-child relationship (child law (international)), the mutual authority of married couples to the law applicable to the effects of marriage (family law (international)), that of an executor of a will to the conflicts rules for succession (succession law (international)) and that of a liquidator to the lex concursus (insolvency, cross-border).
In contrast, most legal systems have elaborated special regimes with respect to the law governing voluntary agency (see 3. below). In this respect, a tendency towards codification of the corresponding conflicts rules may be noted: some countries implemented the Hague Agency Convention (HAC, see 3. b) and 5. below) while many others adopted autonomous solutions (eg Bulgaria, Estonia, Italy, Liechtenstein, Lithuania, Austria, Portugal, Romania, Spain, Switzerland, Angola, Mozambique, China, Taiwan, Korea, Quebec). Some legal systems, however, still only have unwritten choice-of-law principles as to authorities granted by an act of will, with the precise scope and content not yet fully clarified in academia and practice (eg Germany, England).
2. Triangular setting of agency
The legal concept of agency is of fundamental economic interest and relates to a central issue of the general part of the substantive law of contracts, viz the formation of contract (contract (formation)). In principle, the idea of representation refers to the problem of whether and under what circumstances an acting person (agent) can effectively bind another person (principal) vis-à-vis a third person and what the exact legal consequences of these acts are. In this regard, the question of the existence and the precise scope of the agent’s (alleged) authority is of crucial importance. Due to differences in the various solutions laid down by the local substantive laws on representation, the legal effects of cross-border agency may differ considerably depending on which law actually governs these issues (authority of agents; representation). Therefore, the purpose of choice-of-law rules on agency is to mitigate the local discrepancies by guaranteeing a best possible ex ante legal certainty to the parties involved, and thus to meet the commercial need for a smooth employment of agents in international legal relations.
When dealing with problems of representation in private international law, one has to be aware of the fact that once an intermediary is involved in the conclusion of a contract, a triangular setting with three different relationships evolves. The main focus of interest is on the legal relationship between the principal and the contracting third party, ie the transaction to be mediated (main operation, external relationship). The latter’s effective conclusion is the proper purpose of the agency situation. Further legal relationships exist between the principal and the agent (basic or internal relationship) as well as between the agent and the third party negotiating with the former. Since the main operation as well as the internal relationship qualify as contractual obligations, they are both subject to the choice-of-law rules implemented by Reg 593/ 2008 (Rome I Regulation) on contractual obligations (PIL).
Representing an (alleged) power to disseminate declarations of intention vis-à-vis third parties with direct effect for and against the principal (Art 3:202 PECL; Art 2.2.3 UNIDROIT PICC), it is characteristic of the authority of an agent that it affects any of the three legal relations involved. As regards the external relationship, the effectiveness of the envisaged contract, ie the whole purpose of the agency situation, depends on the existence and scope of the agent’s authority. It thus also inversely influences the relationship between the agent and the third party, for instance with regard to potential liability as falsus procurator (Art 3:204 PECL; Art 2.2.6 UNIDROIT PICC). Ultimately, its existence and ambit may have, for example, an impact on a possible recourse within the internal relationship.
When drafting an appropriate connecting factor for the power of attorney, the following conflict of interests occurs. From the third party’s perspective it is crucial to know whether and how the main operation can be concluded validly with the principal according to the relevant substantive law. Hence, he wants to be able to assess, at reasonable costs and no later than the formation of contract, which law governs the authority of the agent as well as the precise content of such law. If, under the substantive law thus identifiable to the third party, an (apparent) power of attorney of the (supposed) agent exists, the former should be protected in his reliance on this (apparent) authorization and thus on the applicability of the identified law as a matter of safeguarding the functioning of and the confidence in international commercial intercourse. In contrast, the (alleged) principal is interested in not being bound by such contracts, which may be formed in his name, but to which he has given no consent by way of an actual authorization. Finally, the agent also has an interest in a clear and foreseeable connecting factor for the power of attorney in order to assess his risks of liability vis-à-vis the principal and the third party.
3. Principles of national law
Two basic types of connecting factors (with several subcategories) for cross-border agency may be identified among the European legal systems. On the one hand, an accessory solution which subjects the authority either to the proper law of the internal relationship or to that of the main operation, and, on the other, an autonomous solution referring to a separate conflicts rule which is independent from the law governing either of these two relationships.
In France, a power of attorney was traditionally subjected to the law applicable to the internal relationship until the entry into force of the HAC in 1992 (just as in Belgium, Luxembourg, Italy (in part), England and Switzerland). Inter alia, this is probably due to the fact that, as a matter of substantive law, the French Code civil does not distinguish between the internal and external relationship, ie the agent’s authorization is not independent of (abstract from) the former. This correlation, however, does not seem to be compelling: in Austria, for example, the authority has been treated separately (abstractly) in private international law for a long time already, even though it is not construed abstractly under the substantive law of the Allgemeines Bürgerliches Gesetzbuch (ABGB). As regards the conflict of laws, a synchronization with the internal relationship achieves the protection of the principal and the agent since they negotiated the contract constituting the internal relationship. However, powers of attorney are naturally designed to have external effects on third parties. The latter usually have no insight into the agent’s appointment and, in case of a choice of law by the parties in particular, can scarcely determine the law applicable to the internal relationship. In other words, such an accessory connecting factor does not provide third parties with an adequate opportunity to check whether the agent can bind the principal effectively, which might hamper the smooth functioning of cross-border market transactions. Therefore, the power of attorney is today almost unanimously separated from the basic relationship as regards the conflict of laws.
Primarily in the interests of the third party, a second version of an accessory conflicts rule has been proposed, ie the application of the law governing the main operation determined according to the general conflicts rules on contractual obligations (PIL). This position is (probably) adopted by English common law although the relevant decisions partly contradict each other and may be cited for several solutions simultaneously. Despite this uncertainty, however, the prevailing opinion today qualifies issues related to an agent’s authority as being an inseparable part of the business transaction as a whole, and thus applies the proper law of the main operation. That way, the third party is freed from an obligation to ascertain another law. Primarily, however, this model does not entail the problem of separating the ambit of the law applicable to the authority from that governing the main contract. Such delimitation may be difficult to implement in practice and constitutes an inevitable consequence of a separate conflicts rule. Thus, some continental authors also approve of the English solution. The majority, however, submit that a synchronization with the main contract does not properly cater for the peculiarities of the trilateral relationship of agency. For the agent could expand his powers together with the third party by choosing an extraordinarily liberal law for the main contract. Conversely, a retroactive limitation at the agent’s expense via choice of law between principal and third party could occur as well. The fact identified as the core problem of the accessory conflicts rule is that, until the main transaction is concluded, the power of attorney is in a vacuum as regards the conflict of laws. For as long as the contract does not exist as a reference point for general conflicts rules on contractual obligations, neither the subjective nor the objective connecting factors can point to any particular applicable law. Hence, the parties involved would have no possibility to check the existence and scope of the agent’s powers before concluding the contract. This would lead to uncertainties right from the beginning of the negotiation of mediated cross-border contracts.
Therefore, most of the legal systems have implemented a separate connecting factor for the issues related to cross-border agency. This holds true for all mentioned codifications of private international law containing explicit rules on powers of attorney (see 1. above) and is in line with the unwritten principles under German and probably Greek law. From a comparative perspective, two objective criteria are generally agreed to be connecting factors: the place of acting and the principal place of business of professional agents. In order to protect the principal, the question of whether an authorization exists in the first place was sometimes assessed according to the principal’s domestic law until the 1960s (eg in Germany and Switzerland). This approach, however, has been almost completely given up to facilitate the smooth functioning of commercial intercourse (though Russian law and partly English law may still differ).
Apparently, all countries allow a choice of the law by the parties governing the authorization. The solutions differ, however, with regard to the technical question which persons have to be involved in this choice. While some legal regimes decide in favour of a bilateral choice between the principal and third party (HAC, Bulgaria, Romania, Switzerland, Quebec, Taiwan), others enable the principal to designate the applicable law unilaterally, provided this determination was recognizable to the other involved parties before the conclusion of the contract (Germany, Liechtenstein, Austria, Spain, Korea).
In the absence of a choice of law, all of the cited legal systems refer to the lex loci actus as the decisive connecting factor at some point. This solution is explained by the need to safeguard the smooth functioning of and confidence in cross-border market transactions. Accordingly, third parties should be protected in their trust in the application of those rules of representation which apply at the place at which the agent performs the relevant legal acts. The place of performance is the only point of reference based on which third parties can determine the law applicable to the agent’s powers as long as no other appropriate criteria become evident to the third party. In addition, the place of performance is usually situated in the same country in which the third party conducts his business activities. Against this background the connecting factor of the place of acting is supposed to foster the market participant’s trust in the smooth settlement of mediated transactions on a specific market and thus to facilitate commercial trade. In the age of maximum mobility and borderless electronic communication, however, the tangible place of acting can often be accidental and/or difficult to localize.
Therefore, most legal regimes do not exclusively link the power of attorney to the lex loci actus (see, however, Estonia, Lithuania, Spain, China) but, subject to certain prerequisites, to the law applicable at the place of business of a professional agent (eg HAC, Belgium, Bulgaria, Germany, Italy, Liechtenstein, Austria, Romania, Switzerland, Korea, Angola, Mozambique). This approach accounts for the fact that it is common practice in international trade to employ permanent professional agents. Their diverse services are thus of pivotal significance to the European internal market (see commercial agents; insurance intermediaries; franchising; legal profession, financial intermediary). With regard to the interests and expectations of the parties involved, the agent’s place of business is situated ‘in the middle’. This connecting factor is usually recognizable just as easily and just as early for both the principal and the third party (eg through business correspondence and business cards). Moreover, being the starting point of the agent’s mediating activity, it constitutes a fixed criterion which is less open to manipulation and can only rarely be viewed as accidental. Primarily, though, this connecting factor allows for uniform treatment of a permanent power of attorney of professional agents. The national solutions differ, however, as to the basic order of priority of the two connecting factors ‘place of establishment’ and ‘place of acting’. While in many countries the latter applies only subsidiarily in the event that the agent’s business establishment was actually not recognizable to the third party (Germany, Italy, Liechtenstein, Austria, Romania, Switzerland, Korea, Angola, Mozambique, similarly Belgium), other legal regimes de facto invoke the lex loci actus primarily as it applies in any case in which either the third party or the principal is established in the country of the performance of the agent’s acts (HAC, Romania, Quebec).
The main difficulty linked to the continental solution is the necessity to determine the ambit of the law governing the issues of agency from that applicable to the main operation. In this respect, German law in particular assigns a relatively narrow scope to the law governing the authority (Vollmachtsstatut) which is essentially limited to core issues of the power of representation (existence, scope, apparent authority). The majority of legal systems, however, construe a more comprehensive ‘statute of representation’ (statut de la représentation) which comprises not only the mere scope of the power of attorney but also all the legal consequences of the agent’s activities as to the envisaged contract, including a possible liability as falsus procurator and, in many countries, the effects of an indirect (undisclosed) agency (representation).
4. Principles of Union law
The prerequisites and legal effects of representation constitute central problems of the ‘existence and validity of a contract’ in terms of Art 10(1) Reg 593/2008 (Rome I Regulation) and thus exist among the core issues of international contract law (contractual obligations (PIL)). However, Art 1(1)(g) Rome I Regulation explicitly excludes ‘the question whether an agent is able to bind a principal […] in relation to a third party’ from the scope of the regulation. The identically-worded predecessor of this exclusion, ie Art 1(2)(f) of the Rome Convention on the Law Applicable to Contractual Obligations, was originally explained by the fact that party autonomy (which represents the core principle of the Convention) is not properly approved in the context of the law governing agency. Today, this view has actually been overruled by the described status of the various national laws. The first Commission proposal for a Rome I Regulation, also contained an independent conflicts rule for ‘contracts concluded by an agent’ (Art 7 COM(2005) 650 final). It was substantially modelled on the corresponding provisions of the HAC (see 3. b) above). The proposed wording, however, was generally viewed as inappropriate for the most part. In the further legislative process the new approach was completely withdrawn, which was probably due to the United Kingdom’s objection to a separate conflicts rule on agency. Ultimately, a provision on the law governing the effects of the agent’s activity was not included in the Commercial Agents Directive (Dir 86/653 of 18 December 1986 on the coordination of the laws of the Member States relating to self-employed commercial agents). The Ingmar decision (ECJ Case C-381/98 – Ingmar  ECR I 9305) merely relates to issues of the internal relationship.
5. Principles of uniform law
Under the auspices of the Hague Conference on PIL, the Convention on the Law Applicable to Agency of 14 March 1978 (HAC) was produced. However, it did not prove to be particularly successful as the HAC has only been in force since 1992 and only in a total of four countries (Argentina, France, Netherlands and Portugal; it served as a model for autonomous PIL in Bulgaria and Romania). The HAC applies both to direct as well as to indirect representation (Arts 1, 2). Its content may be divided into two main parts: the first contains conflicts rules for the internal relationships between principal and agent (Arts 5–9) excluding labour contracts (Art 10); the second relates to the law applicable to the legal effects of representation regarding the principal/third party (Arts 11–14) or agent/third party (Art 15) relationships, respectively. The latter two relationships adhere to an independent connecting factor in terms of a comprehensive statute of representation as described in 3. b) above. As to the internal relationship, a choice of law (implicitly) concluded between principal and agent is primarily significant (Art 5). If no law has been chosen, their relationship is subject to the applicable law at the agent’s place of business (Art 6(1)); unless, according to the contract, the agent is primarily to act in the principal’s country of establishment, in which case that country’s law shall be decisive (Art 6(2)). In sum, the HAC contains rules generally similar to those of Arts 3, 4 Rome I Regulation. Among its contracting states, the Convention’s solutions prevail over the European rules even in pure intra-Union settings (Art 25(1) Rome I Regulation).
François Rigaux, ‘Agency’ in IECL III/2 (1973) ch 29; Ulrich Spellenberg, Geschäftsstatut und Vollmacht im internationalen Privatrecht (1979); Jürgen Basedow, ‘Das Vertretungsrecht im Spiegel konkurrierender Harmonisierungsentwürfe’ (1981) 45 RabelsZ 195; HLE Verhagen, Agency in Private International Law (1995); Ricardo Rueda Valdivia, La representación voluntaria en la contratación internacional (1998); Max Planck Institute for Foreign Private and Private International Law, ‘Comments on the European Commission’s Green Paper on the Conversion of the Rome Convention of 1980 on the Law Applicable to Contractual Obligations into a Community Instrument and its Modernization’ (2004) 68 RabelsZ 1, 90 ff; Lord Collins and others (eds), Dicey, Morris and Collins on the Conflict of Laws, vol II (14th edn, 2006) paras 33R-404 ff, 33R-428 ff; Max Planck Institute for Comparative and International Private Law, ‘Comments on the European Commission’s Proposal for a Regulation of the European Parliament and the Council on the Law Applicable to Contractual Obligations (Rome I)’ (2007) 71 RabelsZ 225, 298; Ulrich Spellenberg, ‘Vertreterverträge’ in Franco Ferrari and Stefan Leible (eds), Ein neues internationales Vertragsrecht für Europa—Der Vorschlag für eine Rom I-Verordnung (2007) 152; Simon Schwarz, ‘Das internationale Stellvertretungsrecht im Spiegel nationaler und supranationaler Kodifikationen’ (2007) 71 RabelsZ 729.