Extraterritorial liability for corrupt practices

On 13 June 2019, I was invited to speak in The Hague at the book launch of the new Commentary on the UN Convention against Corruption (Oxford University Press 2019, edited by Cecily Rose et al.). For this book, Friederycke Haijer and I (both Ucall) wrote a commentary to the jurisdictional article of the Convention (Article 42 UNCAC). At the launch, I addressed the question whether a state could hold a person liable for ‘extraterritorial’ corrupt practices – practices that largely take place outside the regulating state. I argue that, in principle, it can do so, but that extraterritorial liability is not limitless. Especially expansive US enforcement practices, which are based on only a tenuous US connection, may amount to jurisdictional overreach.  

Article 42 UNCAC allows state parties to exercise jurisdiction over corruption on the basis of the classic principles of jurisdiction: territoriality, nationality/personality, protection/security, and aut dedere aut judicare (i.e., jurisdiction based on the presence of the presumed offender, provided that the state does not extradite him). Article 42 also contains a savings clause, which allows states to exercise other types of jurisdiction, provided that they are in keeping with (customary) international law. It is doubtful, however, whether such other types, particularly universal jurisdiction (i.e., jurisdiction that is only based on the gravity of the crime of corruption, regardless of connections with the state exercising the jurisdiction), are currently authorized by international law.

Under Article 42 UNCAC, the exercise of jurisdiction is largely discretionary. This means that states are entitled to exercise jurisdiction on the basis of one of the listed principles, but they are not obliged to. Only in two situations jurisdiction is mandatory, namely when the offence is committed in the territory of the state, and when the offence is committed by a national of the state, and that state does not extradite him. The latter scenario is relatively peculiar, and only applies to natural persons. Hence, let us zoom in on territorial jurisdiction, as it is on the basis of territoriality that states often exercise jurisdiction over corrupt practices involving foreign persons.

The text of UNCAC does not clarify when exactly a corrupt practice takes place on the territory of a state party. Article 4 of the OECD Anti-Bribery Convention (1997), UNCAC’s sister convention, is more explicit where it states that ‘[e]ach Party shall take such measures as may be necessary to establish its jurisdiction … when the offence is committed in whole or in part in its territory.’ This article thus refers to the ‘constituent elements’ approach, pursuant to which it suffices for the exercise of territorial jurisdiction that one element of the offense has been committed in the territory. Ultimately, however, the determination of whether an offence has been committed in part in the territory is a matter of construction of domestic anti-corruption statutes.

So far it has been mainly the United States which has vigorously clamped down on foreign corrupt practices. Also, US law and practice have informed the adoption of the OECD Convention and later UNCAC. Accordingly, a closer look at how US law-enforcers have construed territoriality under US law is warranted.  I will refrain from giving a general overview of the enforcement of the US Foreign Corrupt Practices Act, and limit myself to the threshold question of jurisdiction. The jurisdictional question is a crucial one, however, as without jurisdiction, US law-enforcement agencies cannot start an investigation, and thus cannot find persons liable for foreign corrupt practices.

US law-enforcement agencies have had no qualms going after non-US corporations in relation to foreign corrupt practices. This obviously begs the question whether the nexus of these persons and their activities with the US is strong enough to legitimately ground US jurisdiction. Let us look at the relevant connections which can ground jurisdiction over non-US persons under the FCPA. First, since 1977, the FCPA applies to ‘issuers’, i.e., corporations that trade on a US stock exchange, including non-US corporations (15 U.S.C. Section 78dd-1). Admittedly, a US listing does not turn a foreign corporation into a US one, but at least such a corporation can be considered as ‘resident’ in the US, and it can anticipate the liability consequences of trading stock in the US. Much more problematic, however, is 15 U.S.C. Section 78dd-3 FCPA, as amended in 1998. This section prohibits any person, including a non-US person, from ‘using mail or interstate commerce in the US in furtherance of any act that constitutes bribery of a foreign official’. In practice, this provision, which is based on territoriality, has been interpreted very broadly by US law-enforcement agencies, with tenuous connections to the US sometimes sufficing for the application of US law. Thus, the agencies have considered as sufficient such ‘incidental’ connections as using a correspondent bank account in the US to clear transactions in US dollars, and sending an email through a US server. On this basis, and bearing in mind the reality of US financial and technological dominance, many international transactions become subject to US law.

One may argue that such a broad construction of territoriality is desirable from the perspective of stamping out foreign corrupt practices as a global ill –  which is moreover an obligation under UNCAC and the OECD Convention. Arguably, the US is simply implementing the conventions which mandate the exercise of territorial jurisdiction as soon as a constituent element of the offence takes place in its territory. Nevertheless, while it is true indeed that the international community has considered corruption as globally reprehensible, there is no consensus on what acts precisely constitute corruption and how they should be penalized. US interpretations and especially US sanctions may well be very different from practices elsewhere. This becomes particularly problematic if a situation has almost no link with the US, but is much more closely tied to another state – in which case the US may even impinge on that other state’s sovereignty. It remains that so far, there has been very little foreign governmental protest against FCPA assertions. This is probably the case because foreign states, especially those parties to the conventions, do not want to be seen as condoning corruption.

Foreign persons at the receiving end of the FCPA could obviously contest the agencies’ interpretation of the scope of the FCPA before a US court. However, in reality, this almost never happens, as the agencies ordinarily enter into out-of-court settlements, and for reasons of costs and legal uncertainty, foreign firms do not seek judicial review. As a result, there is little to no case-law on the permissible reach of the FCPA. Accordingly, the enforcement agencies decide for themselves, without guidance from the courts.

Still, there is one case where a legal challenge occurred, and a successful one at that: the Hoskins case (2018). This case may perhaps be a harbinger of more to come. Hoskins was a British national and a former executive of French corporation Alstom, based in Paris. US agencies charged him with conspiracy to violate, and with aiding and abetting FCPA violations by Alstom’s US subsidiary (a ‘domestic concern’ subject to FCPA jurisdiction). However, Hoskins never worked for Alstom US and never travelled to the US during the bribery scheme. Faced with this reality, the US Court of Appeals for the 2^nd Circuit ruled that the US Congress wanted to exclude such foreign nationals from FCPA liability as long as they do not act as agents of a domestic concern of US issuer, and when they act outside the US. The Court held that the US Department of Justice had overreached, and that in this case, the US had no jurisdiction and could not proceed to hold Hoskins liable for his extraterritorial acts. Hoskins goes to show that it may make sense to challenge the US agencies’ interpretation of the jurisdictional reach of the FCPA.

Concluding: while corruption has been internationally criminalized, notably by UNCAC, and UNCAC gives broad jurisdictional powers to states, this is not a blank check for states to exercise quasi-universal jurisdiction over foreign acts of corruption. Under the international law of jurisdiction, a substantial connection with the state is required. And under the presumption against extraterritoriality (e.g., in the US), statutes are only intended to apply in the territory, unless a contrary congressional intent appears. The combined effect of the law of jurisdiction and the presumption may narrow the geographic scope of states’ foreign anti-corruption legislation. This can prevent international discord and safeguard the legitimate interests of foreign persons.