In Accordance with Which Host State Laws? Restoring the ‘Defence’ of Investor Illegality in Investment Arbitration

Investment treaties are often criticised for being too ‘investor-friendly,’ containing wide substantive protections and broad definitions of their scope, and only rarely containing exceptions clauses. With this in mind, it becomes important to clarify the mechanisms available to host states to defend against investment treaty claims. One such mechanism is found in the provisions included in many investment treaties, to the effect that investors must comply with host state law in order for their investment to enjoy treaty protection. Where states have proven that investors engaged in illegal conduct when making their investment, tribunals have in some cases rejected jurisdiction over the claim—thus providing a weapon for host states to defeat investment treaty claims in appropriate circumstances.

However, despite much recent attention to these ‘investor legality’ clauses, there are several persisting areas of uncertainty in the scope of this ‘defence’ for host states. In particular, certain tribunals have previously proposed limitations on the range of host state laws with which an investor is expected to comply. These limitations could pose obstacles for states seeking to rely on investor illegality to have a claim dismissed.

This article briefly outlines these proposed limitations, and argues that they are unjustified both in terms of their consistency with statements in other cases and their normative desirability.

All host state law, or only the ‘fundamental principles’?

The first proposed limitation stems from suggestions by certain tribunals that investors are not required to comply with all of the host state’s law. In the 2006 award in LESI v Algeria, the tribunal took the investor legality provisions in the Italy-Algeria bilateral investment treaty (BIT) to mean that investments would lose treaty protection only when made “in violation of fundamental principles in force.”[1] Two awards from 2008, Desert Line v Yemen and Rumeli v Kazakhstan, followed this view, holding that investor legality only related to breaches of “fundamental principles of the host State’s law.”[2] Thus, according to these tribunals, it does not matter whether investors have complied with all of host state law, as long as they have complied with the fundamental legal principles of that jurisdiction.

But there are three reasons to doubt whether this position is (or still remains) a justifiable interpretation of investor legality provisions. Firstly, these provisions, as typically worded in investment treaties, make no reference to ‘fundamental principles.’ Instead, they simply require that investments be made ‘in accordance with host state law’ or similar wording.

Secondly, the LESI, Desert Line or Rumeli tribunals were not required to make a finding on whether the investor had complied with host state law.[3] This means that their off-hand, obiter references to ‘fundamental principles’ carry far less force than if, say, jurisdiction had been accepted despite a clear breach of a regular, non-fundamental host state law.

Thirdly, no other tribunals have appeared to support the ‘fundamental principles’ limitation. Although in most cases the limitation has not been explicitly rejected, where questions of investor legality have arisen, the tribunal has simply not discussed whether or not the relevant law forms part of the host state’s fundamental principles.[4] Furthermore, in one 2012 jurisdictional decision, Quiborax v Bolivia, the tribunal expressly rejected the claimant’s support for a ‘fundamental principles’ limitation, considering that it went “beyond the terms of the BIT, in an attempt to further the investor’s protection without due regard for the State’s interests.”[5] While the lack of binding precedent in investment treaty arbitration means that the Quiborax rejection does not necessarily ‘overturn’ the LESI/Desert Line/Rumeli view, it certainly appears that the prevailing doctrinal position is one of scepticism towards the limitation.

Given this, is it possible to re-interpret the comments of the LESI/Desert Line/Rumeli tribunals in an effort to make them more consistent with other case-law? In fact, a careful reading of many cases[6] suggests that investor legality provisions have two elements to them. The first element—which might be termed ‘narrow investor legality’ —requires compliance with the positive terms of host state law, in its entirety. The second element, by contrast— ‘broad investor legality’—additionally requires the investor to comply with fundamental, general principles of law in the abstract, such as good faith, international public policy and prohibitions of fraud, corruption and deceit. This reading can explain why, for instance, the Inceysa v El Salvador tribunal declined jurisdiction after finding that the investor had violated principles of good faith, unjust enrichment and international public policy. In fact, the tribunal’s finding of breach of ‘broad investor legality’ was sufficient to decline jurisdiction without examining Inceysa’s compliance with positive Salvadoran law.[7]

In this sense, Desert Line, Rumeli and LESI were not wrong to suggest that investments must comply with fundamental legal principles. They were merely under-inclusive, in that investments must also comply with the rest of host state law as well.

One caveat must be applied to this analysis. It is well-accepted that tribunals will not decline jurisdiction merely because the investor has committed a trivial breach of local law. The minor defect in company paperwork at issue in Alpha Projektholding v Ukraine, for instance, did not exclude the tribunal’s jurisdiction.[8] Although, like the ‘fundamental principles’ limitation, this caveat is not found in investment treaty text either, it has much greater support from tribunals, and also carries a stronger normative justification. States must not be permitted to abuse the technicalities of their own laws to evade investment treaty claims. At the same time, though, a trivial breach is not simply the flipside of breach of a fundamental law: there are, of course, regular laws in the middle of the spectrum. As argued above, there is no reason to exclude the investor’s compliance with these laws.

All host state law, or only laws related to investment?

Alongside the ‘fundamental principles’ limitation, one tribunal has supported a second limitation on the laws with which an investor must comply. In Saba Fakes v Turkey, the tribunal commented that Turkey’s claim of the investor’s non-compliance with domestic competition law and telecommunications regulatory law was irrelevant, because these were not laws “related to the very nature of investment regulation.” For the Fakes tribunal, the investor legality provision only required compliance with laws “governing the admission of investments in the host State.”[9]

Again, though, there is no indication of such a limitation in typical investment treaty text. It is hard to see why an investor should not be required to comply with competition and telecoms laws if these laws affect the entry of new players to a state’s telecommunications market. In any case, almost by definition, any law that an investment might potentially breach is surely a law “related to the very nature of investment regulation” —if the law in question does not regulate investment, it seems unlikely that an investment could breach it. The Fakes tribunal’s comment is therefore puzzling on its face.

Moreover, no other tribunal has considered the subject-matter of the law allegedly breached by an investor to be relevant. In Anderson v Costa Rica and Hamester v Ghana, the state alleged breaches of domestic criminal laws on fraud—which are, arguably, laws that are not “related to the very nature of investment regulation,” since they naturally address a wide range of actors and situations. However, the tribunals in those cases did not suggest that compliance with those laws could be ignored. In Vannessa Ventures v Venezuela, where the claimant explicitly relied on the Fakes view in its submissions, the tribunal side-stepped the question, finding simply that no Venezuelan laws had been breached.[10] Even in Fakes itself, the tribunal was not required to apply its own limitation, since it ultimately found that there was no investment at all.[11] Restricting the investor legality requirement only to investment-related laws, then, seems quite unjustifiable.


Compliance with host state law is a crucial part of the investor-state bargain. The investor legality requirement represents one of the few instances where the traditional one-sided nature of investment treaties—imposing many obligations on host states, and few on investors—is broken. Treaty negotiators are starting to expand on this, with the latest leaked CETA text, for instance, including an explicit provision barring arbitration claims by investors which made their investments through “fraudulent misrepresentation, concealment, corruption, or conduct amounting to an abuse of process”,[12] alongside the more usual requirement to invest “in accordance with the applicable [host state] law.

This article contends that a proper interpretation of the legality requirement, as traditionally expressed, would not limit the range of laws with which investors must comply. Rather, the default position must be that investors are obliged to comply with all of host state law, with trivial violations as the only exception. In a context of backlash by states against the investment treaty regime, it is important not to restrict the available defence mechanisms beyond what is permitted by the text, object and purpose of investment treaties.

Author: Jarrod Hepburn is a Lecturer in Law at the University of Exeter, UK.

[1] LESI SpA and Astaldi SpA v Algeria (ICSID Case No ARB/05/3), Decision on Jurisdiction, 12 July 2006 [83].

[2] Desert Line Projects LLC v Yemen (ICSID Case No ARB/05/17), Award, 6 February 2008 [104], also citing

LESI; Rumeli Telekom AS v Kazakhstan (ICSID Case No ARB/05/16), Award, 29 July 2008 [319].

[3] In Rumeli it was not clear that the state had alleged any specific breaches of Kazakh law: [169]. In Desert Line ([93(c)], [104], [118]) and LESI ([83(iii)]), the only relevant issue was whether the investment had met certain formal requirements for recognition in domestic law.

[4] See, eg, Saluka Investments BV v Czech Republic (UNCITRAL), Partial Award, 17 March 2006 [213]–[214]; Teinver SA v Argentina (ICSID Case No ARB/09/1), Decision on Jurisdiction, 21 December 2012 [324].

[5] Quiborax SA v Bolivia (ICSID Case No ARB/06/2), Decision on Jurisdiction, 27 September 2012 [263].

[6] Phoenix Action Ltd v Czech Republic (ICSID Case No ARB/06/5), Award, 15 April 2009 [100]–[105]; Gustav F W Hamester GmbH & Co KG v Ghana (ICSID Case No ARB/07/24), Award, 18 June

2010 [123]; Teinver [317]; Metal-Tech Ltd v Uzbekistan (ICSID Case No ARB/10/3), Award, 4 October 2013 [127]; Plama Consortium Ltd v Bulgaria (ICSID Case No ARB/03/24), Award, 27 August 2008 [140].

[7] Inceysa Vallisoletana SL v El Salvador (ICSID Case No ARB/03/26), Award, 2 August 2006 [264].

[8] Alpha Projektholding GmbH v Ukraine (ICSID Case No ARB/07/16), Award, 8 November 2010 [297].

[9] Saba Fakes v Turkey (ICSID Case No ARB/07/20), Award, 14 July 2010 [119]–[120].

[10] Vannessa Ventures Ltd v Venezuela (ICSID Case No ARB(AF)/04/6), Award, 16 January 2013 [132], [135], [154], [160], [164], [167].

[11] Fakes [148].

[12] Article X.17.3, Consolidated CETA Text of 1 August 2014.