Symposium on International Investment Law & Contemporary Crises: The Police Powers Doctrine and the Right to Regulate in Banking and Finance Investment Disputes – Drawing the Line in International Investment Law

[Saïda El Boudouhi is a professor at Université Paris 8. She has held academic positions across leading French and European institutions and contributes actively to contemporary debates on the evolution of public international law]

Drawing from the Empirical Study on International Investment Law Protections in Global Banking and Finance (‘the Study’), the present post focuses on the reliance by investment arbitral tribunals on the police powers doctrine as mobilized in that specific case law. The authors of the Study have dedicated to the topic some attention in the main discussion, as well as one of the four Annexes entitled ‘Findings of Tribunals regarding police powers of States involving emergency measures’.  The police powers doctrine thus stands out as playing a ‘critical role’ in investment arbitration in that sector (p. 36), even though it involves only a handful of cases out of the 149 covered by the Study.

This ‘critical role’ must, however, be qualified, as it seems that it is limited to a certain type of disputes arising in the banking and finance sector. Indeed, the authors of the Study have devised a section entitled ‘Who has prevailed on the merits?’ which comprises four main sub-sections and three rather miscellaneous ones, suggesting that the answer to the question asked in the section depends largely on the type of cases or measures involved. The four main sub-sections are:

• (1) sovereign debt obligations,
• (2) emergency interventions,
• (3) currency and exchange interventions,
• (4) general regulatory and sectoral reform.

Among these, only the emergency interventions leave space, according to the outline of the Study, for the police powers doctrine—or at least that is where it has been most evidently identified and dealt with. 

However, the State’s right to regulate, more generally, is invoked by the Study in sections dedicated to other types of cases such as currency and exchange interventions, which will also be included in this Note. This provides an opportunity to distinguish between two notions of international investment law that are too often conflated: the police powers doctrine (also known as the regulatory powers doctrine) on the one hand, and the State’s right to regulate on the other. While challenging, such an encompassing scope allows one to cover what could be considered as general exceptions in international investment law. 

It must be noted that the Study, due to its empirical nature, does not distinguish between the two notions, the awards under examination themselves often conflating them (para. 475). Nor does the Study circumscribe the subjects being studied to specific international investment law principles, the authors choosing to draw a rather broad picture of the field without going into the different types of investment rules that are involved. This post thus aims to add an analytical—and somewhat more formal—perspective to the excellent empirical material that is introduced in the Study regarding the police powers doctrine and the right to regulate. The following sections therefore clarify the distinction between these two notions: relying on the Study, Section 1 examines the police powers doctrine as a component of the indirect expropriation rule, while Section 2 considers the State’s right to regulate as a broader, largely declaratory concept that can nonetheless acquire operative significance, particularly through its interaction with the fair and equitable treatment (FET) principle.

The police powers doctrine as a component of the indirect expropriation rule

The police powers doctrine, also known as the State’s regulatory powers doctrine, appears to have been the first to become prominent in the debate about the need to balance investor protection with respect for State sovereignty. It has appeared in case law in relation to allegations of indirect expropriation and is therefore an operative legal component—i.e. a legal condition—of the rule regarding indirect expropriation. Let us recall that expropriation, whether direct or indirect, is not prohibited under international investment law provided it is in the public interest, in accordance with the law, non-discriminatory, and accompanied by adequate compensation. The characterization of a measure as an indirect expropriation will therefore entail an obligation of compensation even if the measure was adopted in the public interest.

It can thus become crucial for States to prove that their measures do not amount to indirect expropriation, even if it were to be legal. This is where the police powers doctrine intervenes: a measure can be recognized not only as being in the public interest but also as one adopted in the exercise of the State’s police powers. The doctrine allows the measure to escape characterization as ‘indirect expropriation’, thereby relieving the State from paying compensation. Accordingly, a measure affecting the value of an investment will be analyzed as expropriatory unless it results from the exercise of police powers.

The police powers doctrine thus maintains the other legality criteria but makes it possible for a measure not to be subject to compensation if it is adopted for regulatory purposes in pursuit of a legitimate public aim. In investment arbitral practice overall, the doctrine was first relied on in the NAFTA Methanex v. USA (2005) case as stemming from ‘general international law’ (para. 7). In the banking and finance sector, it was widely discussed in Saluka v. Czech Republic (2006), in which the Tribunal considered that the police powers doctrine ‘forms part of customary international law today’ (para. 262). That case—an emergency intervention not referenced in the Annex to the Study, perhaps because the State did not prevail overall—used the doctrine to discard the characterization of ‘indirect expropriation’.  However, this did not preclude a finding of violation under the FET principle. This underscores that the doctrine could only be used to characterize whether a measure constitutes an indirect expropriation and was of no avail for other principles, particularly the FET. In that regard, the Study notes that ‘in every published case of emergency measures where the tribunal has decided in favour of the State on the merits, the Tribunal has emphasized the State’s police powers in banking and finance matters’ (at para. 36). It may be underlined that the reverse is not necessarily true: emphasis on police powers does not guarantee success on the merits, as shown in Saluka.

Beyond the Saluka line of reasoning, the Study also illustrates how tribunals have progressively refined the doctrinal vocabulary through which they recognize legitimate regulatory measures. The Study interestingly relies on the Marfin Investment Group award to stress that:

‘the concept of public welfare in international law extends to “the health and optimal operation of the banking system…, the protection of depositors and clients, and ultimately, the protection of taxpayers”’.

(para. 905), (Study, at p. 36)

The Marfin award itself refers to previous awards which had used the concept of public welfare, starting with Saluka (para. 255) or Invesmart (para. 497) , which used the close expression of ‘general welfare’. Annex D of the Study puts forward another occurrence of ‘public welfare’ in an emergency intervention case, the case of Renée Rose Levy v. Republic of Peru (paras. 475-476). This shift towards ‘public welfare objectives’ appears in recent trade and investment agreements as is shown in the Philip Morris v. Uruguay award regarding a more traditional component, i.e., public health (para. 300). 

This reliance on public welfare, rather than on the more traditional notion of ‘public interest’ or ‘public purpose’—which was used in the Methanex award for instance (para. 7)—signals a shift in arbitral reasoning—from a narrow assessment of a measure’s legality to a broader appraisal of its economic and systemic justifications. This potential expansion of acceptable regulatory measures under the heading of what contributes to the public welfare is however accompanied by a condition of proportionality that adds to the traditional due process and non-discrimination requirements (Philip Morris, at para. 409; Marfin Investment Group, paras. 826-829). This line of reasoning, and especially the introduction of the proportionality test, allows for a comparison, if not for an alignment, with general exceptions that can be found in trade and investment agreements and even beyond with restrictions of rights in human rights systems. The concept of public welfare as a legitimacy criterion is thus distinguishable from public interest as a legality criterion for expropriation. While the public interest refers to the formal justification of the measure, the public welfare captures the substantive range of legitimate objectives pursued by the State in regulating key sectors such as banking and finance. It denotes the set of legitimate interests that allow a measure which might otherwise be characterized as an indirect expropriation (and therefore compensable) to be treated instead as a legitimate regulatory measure not requiring compensation.

The right to regulate: from a theoretical principle to an operative standard in the context of fair and equitable treatment

The right to regulate is distinct from police powers even though both stem from the same concern to balance investor protection and State sovereignty. The right to regulate is broader and primarily theoretical or declaratory. It is not operative in itself, since investment treaties or chapters are designed to grant rights to investors, not to States. A tribunal may recall a State’s general ‘right to regulate’ in its opening considerations, but unless that right is translated into an operative legal element linked to a specific investment-law principle, it produces no independent legal effect. 

Accordingly, the right to regulate is often invoked alongside the police powers doctrine even though it adds little, since the latter concretely manifests the former. Most of the cases listed in Annex D show tribunals using both concepts together, as if the police powers doctrine were the operational expression of the right to regulate. This can be illustrated by the Invesmart award, referenced in the Study in the section on emergency intervention cases invoking the right to regulate but not among those relying on police powers in Annex D. Yet, while the ‘right to regulate’ appears only in the opening reasoning (‘investment treaties were never intended to do away with their signatories’ “right to regulate”’), the police or regulatory powers doctrine is used to distinguish between compensable and non-compensable expropriation and ultimately leads the Tribunal to discard the characterization of indirect expropriation (para. 520). Invesmart could thus well be added to the list of cases where the police powers doctrine was relied on and the State prevailed.

The limited justiciability of the right to regulate as a stand-alone concept and its contrast with the police powers doctrine, is illustrated further in Daimler Financial Services AG v. Argentina. While Argentina had argued, as an objection to the Tribunal’s jurisdiction, that the matter fell within its ‘exclusive sovereignty under international law’ (para. 94), the Tribunal distinguished between the ‘Respondent’s …right to regulate its economy as it sees fit’ and ‘[its] general sovereignty…’ on the one hand and its ‘obligation to observe its treaty commitments’ on the other (para. 100). This makes it clear that the State’s general right to regulate enshrined in customary international law cannot in and of itself set aside a State’s treaty obligations. The position differs when the right to regulate is expressly enshrined in the treaty itself, as in Article 8.9 of the Comprehensive Economic and Trade Agreement (CETA). In such cases, other provisions must be interpreted in light of that clause. In CETA, Article 8.9 aims at restricting the scope of the legitimate expectations doctrine included in the FET treatment principle by Article 8.10(4). The right to regulate thus becomes an operative component of the FET principle.

But this inclusion of the right to regulate into the FET principle is not limited to cases where the former is provided explicitly in the treaty. It has also appeared in the most recent case law, including in the banking and finance sector. This analysis is confirmed by the Study and the cases it cites. Where a treaty does not include a right-to-regulate clause, tribunals may still invoke it as one contextual element when assessing alleged FET breaches. The Study appears to adopt this view, explicitly referring (albeit indirectly) to legitimate expectations while omitting any link to the indirect-expropriation rule when referring to the police powers doctrine. The right to regulate thus appears to be tightly linked with the legitimate expectations doctrine, perhaps more than the police powers doctrine is identified with the indirect expropriation rule. In FET analyses, the right to regulate functions as one criterion for assessing the legitimacy of an investor’s expectations. This is evident in Addiko Bank v. Montenegro (2021) and Metalpar v. Argentina (2008), where tribunals invoked the right to regulate not in relation to police powers but as an interpretative element amongst others that must be taken into account in assessing investors’ expectations. Hence the right to regulate becomes an operative legal element for evaluating their reasonableness.

According to Addiko Bank, ‘when balancing a State’s right to regulate against an investor’s expectations, the Tribunal must afford significant latitude to the State to decide what is appropriate for its own internal needs’ (para. 560). Similarly, in Cyprus Popular Bank:

‘the legitimacy or reasonableness of the investor’s expectations must be assessed in conjunction with the political, socioeconomic, cultural and historical conditions in the host State, and in particular, balancing the right of the State under international law to regulate within its borders’.

(para. 1129)

Interestingly, the Study identifies reliance on the State’s ‘general right to regulate and its freedom of monetary policy’ even when these phrases are absent from the awards. In Metalpar S.A. and Buen Aire S.A. v. Argentina, the paragraphs cited do not mention ‘right to regulate’ or ‘freedom of monetary policy’, but simply reject the reasonableness of investors’ expectations based on the fact that ‘there was no bid, license, permit or contract of any kind’ (para. 186). By treating this as an implicit invocation of the right to regulate, the Study broadens the concept’s scope to the point of diluting its operative content. Such an expansive interpretation risks blurring the contours of the State’s right to regulate and may undermine its conceptual precision.

The extensive discussion by the Study of States’ right to regulate nevertheless shows that international investment rules, in their conventional or customary law components, nowadays offer a significant margin to States for the adoption of policies aiming at legitimate public welfare interests. While the Study does not mention cases in the banking and finance sector that have been decided on that basis, it is worth adding that the right to regulate can also be applied to the non-discrimination principles, i.e., national treatment and most-favored-nation clause. Thus, at least in other subject areas, the right to regulate has also been applied beyond the indirect expropriation and FET principles. While the arbitral awards are not always explicit about it, it is not disputed that the divide between prohibited discrimination and regulatory distinctions—‘like circumstances’—is also based on the right to regulate as an operative and interpretative element enshrined in most investment protection principles.