Judicial Expropriation in Investment Arbitration: Opening a Can of Worms?

Print Friendly, PDF & Email

Authors: Harshvardhan Tripathi*



Judicial Expropriation
Investment Arbitration




While expropriation claims arising out of governmental abuse of legislative and executive powers are commonly known, expropriation claims arising from judicial abuse i.e. ‘judicial’ expropriation continues to be an obscure area of international investment law. It has received considerably less attention in scholarly writing and arbitral practice and its conceptual contours remain far from settled.

However, the can of worms that ‘judicial expropriation’ is extends far beyond merely theoretical challenges. In arbitral practice, there seems to exist a ‘phobia’ in the minds of the investment tribunals in holding acts of courts as being expropriatory. Although these tribunals do not discriminate between the three organs of the government while holding them responsible for breach of investment protection standards, in the majority of the cases, either the executive or the legislature is held liable for mistreating the foreign property. The tribunals have generally shirked away from holding acts of the judicial arm of the state as violative of treatment standards.

In light of this context, the current blog post seeks to highlight the vexatious issues surrounding judicial expropriation and investigate the possible reasons behind the tribunals’ hesitant approach. Subsequently, the blog post refutes the prevailing objections against tribunals holding judicial acts as expropriatory and makes a compelling case for why the tribunals should overcome their ‘phobia’ of judicial expropriation.


1. How to define ‘judicial expropriation’?

To put it simply, judicial interference amounts to expropriation only if it can be established that the judicial interference in question is:[1]

(a) discriminatory;

(b) arbitrary or ‘egregious’; and

(c) with the intent to prevent the investor from benefiting from the arbitral award.

In OAO Tatneft v. Ukraine,[2] the tribunal observed that judicial expropriation occurs in two circumstances: first, when the act of domestic court leads to the direct loss of a foreign asset, and second, when the domestic court’s decision is itself part of a series of events that collectively constitute creeping expropriation. However, determining whether an act constitutes judicial expropriation is itself a vexatious issue.

Whether a state act is classified as a compensable ‘expropriation’ or only a regulation (that does not warrant payment of compensation), will depend on which doctrine is chosen for the purpose. There are two main doctrines for determining whether a governmental action constitutes indirect expropriation: the ‘sole effects’ doctrine and the ‘police powers’ doctrine.[3]

While determining whether a particular act is expropriatory, the ‘sole effects’ doctrine takes into account only the effect of the governmental measure on the property, and disregards,[4] or in other cases, gives minor consideration[5] to the reasons behind such governmental actions. On the other hand, the ‘police powers’ doctrine places more emphasis on the nature and the purpose of a governmental measure. It posits that if the deprivation of property is a result of application of bona fide nondiscriminatory rules that fall within the ambit of the State’s police powers, then it would not amount to expropriation.

However, there are drawbacks to relying on either of these doctrines. The sole effect doctrine has been criticized for imposing an unusually low threshold for classifying an act as expropriatory. If strictly applied, the sole effect doctrine would result in a ridiculous situation in which every Court decision that thwarts the enforcement of a foreign award would be considered judicial expropriation. As a result, the scope of ‘judicial expropriation’ would be unreasonably broad.

The unusual result of relying on the police power doctrine, on the other hand, is that the ambit of judicial expropriation is unreasonably narrowed. Because judicial decisions are considered to be a legitimate exercise of the state’s regulatory power, they would never be considered tantamount to expropriation.[6] This conclusion is also difficult to accept. Thus, determining its scope is the very first conceptual challenge associated with judicial expropriation.

Another challenge associated with conceptualizing what constitutes judicial expropriation is that the existing understanding of expropriatory acts is not suited to describe expropriation resulting from judicial acts, since courts of a country operate in a different regulatory ecosystem as compared to other branches of the government.

So far, the legality of expropriation of foreign property has been measured by looking at factors such as:[7]

    1. The adequacy of the compensation
    2. The public purpose of the measure
    3. Compliance with Due Process
    4. Non-discrimination between parties.

Although the language of the investment treaties has not discriminated between branches of state when establishing these criteria, the direct application of these standards is especially tricky when the act of expropriation in question is by a national court.

Is the legality of a judicial expropriation, for example, to be determined by the amount of compensation offered? If the traditional model of understanding expropriation is extended to judicial expropriation, it will lead to an absurd conclusion on two counts: first, not all court judgments award compensation to the judgment debtor, and second, the state will be required to pay compensation for a judgment by its national courts that was attempting to correct a legal imbalance in its own right. Since different branches of government operate within different structural and regulatory frameworks, universally formulated protection standards may be inadequate to cover expropriation by all branches of government, especially judicial expropriation.


2. Vexatious Issues pertaining to judicial expropriation

This blog post will explore some of the most glaring vexatious issues pertaining to judicial expropriation. First, it will highlight how the tribunals have confused the threshold for finding judicial expropriation with the issue of ‘unlawfulness’ of judicial expropriation. Second, it will draw attention to how tribunals have often muddled the boundaries of judicial expropriation with the separate notion of ‘denial of justice’. Third, it will describe the divergent approach of the tribunals with regards to whether the exhaustion of local remedies is an essential component of finding judicial expropriation. Lastly, it will make a persuasive case for why the tribunal’s phobia of finding judicial expropriation is unfounded.


a. The ambiguous threshold of judicial expropriation vis-à-vis unlawfulness of judicial expropriation

When faced with the task of determining whether a certain act is expropriatory, a three-pronged method is generally adopted:[8]

  1. What are the interests that are protected under the treaty?
  2. Has an act of expropriation occurred?
  3. Can expropriation be justified in light of any of the conditions of the treaty?

However, when tribunals are faced with deciding if a particular act qualifies as being judicial expropriation, they not only seem to require the presence of substantial deprivation of the protected rights but also an additional elevated requirement of impropriety. These awards seem to emphasize the distinct nature of the expropriation that result from a judicial act. Hence, the approach emanating could be summarized as thus: generally, the acts of the national court are not held to be expropriation unless it is also accompanied by serious factors that affect the measure.

An instance of such an approach is found in Middle East Cement v. Egypt[9] where the tribunal was dealing with the issue of seizure and sale of investor’s ship. This seizure and sale through auction had been ordered by the court; however, it was not properly notified to the investor. The tribunal held that in the normal course of things both measures of seizure and sale would not be considered expropriation, but if they are not accompanied by due process of law, such act will be judicial expropriation.

However, it must be noted that the tribunal in Middle East Cement required an additional element of impropriety for finding an expropriation itself, and not for determining its unlawfulness. This reflects how in cases of judicial expropriation, the line between determining if an expropriation has occurred and determining if the expropriation is unlawful gets blurred.


b. Confusing the boundaries between judicial expropriation and ‘denial of justice’

As a matter of introduction, there exists both a broad and narrow definition of ‘denial of justice’. The early legal discourse on this issue adopted a broader definition, under which any breach of international law would be tantamount to a denial of justice. The relevant criterion to be looked into when deciding if denial of justice has occurred is to examine the legality of the action “solely upon the quality of lawfulness or unlawfulness which international law attaches to the act and not upon the means of redress afforded the individual against whom it was directed.”[10]

Those advocating for a narrower definition maintain that denial of justice is restricted to only certain international wrongs. The reason for carving out a species of particular international wrongs that fall within the ambit of denial of justice is that otherwise every international wrong can be brought under a denial of justice claim and eventually the term would lose its technical distinction.[11]

Denial of justice usually concerns maladministration by the local courts, such as lack of access to justice, discrimination and slow/nonexistent execution of court decisions favorable to the foreign investor.[12] A prominent instance of this is the Azinian v. Mexico arbitration, which identified four types of denial of justice:

A denial of justice could be pleaded if the relevant courts refuse to entertain a suit, if they subject it to undue delay, or if they administer justice in a seriously inadequate way […] There is a fourth type of denial of justice, namely the clear and malicious misapplication of the law.[13]

A much broader definition is found in the Loewen v. USA arbitration, where the tribunal held that for denial of justice to be found a “[m]anifest injustice in the sense of a lack of due process leading to an outcome which offends a sense of judicial propriety” was required.[14]

The arbitral awards have indeed treated denial of justice and judicial expropriation as separate concepts. However, the difference between these concepts has not been adequately demarcated with clarity.[15] Investment arbitral awards like that of Robert Azinian Kenneth v. United Mexican States have further complicated the interplay between these two separate legal concepts. The observation that caused much confusion was that even though the tribunal conceded that a State could be held liable for the judgments passed by its national courts, it also noted that this did not mean that an automatic right accrued to the claimant by virtue of which they could seek review of a national court’s decision before the tribunal as if the tribunal had plenary appellate jurisdiction.[16]

In some awards, prominently in Ares International v. Georgia[17]and MNS v. Montenegro[18] the Tribunal held that: first, denial of justice is an essential component of judicial expropriation and second, that the finding of judicial expropriation has to necessarily stem from a denial of justice finding.

This is the classic case of conflating the two notions and placing an additional element in the finding of judicial expropriation. As per international law standards, denial of justice is primarily concerned with procedural propriety, non-discrimination and independent adjudication before national courts. Therefore, cases of refusal to access to justice, inordinate delay, state interference in the proceedings, due process violations, prejudice etc. will be covered under the denial of justice claims.

In a number of arbitral awards that have found judicial expropriation, the ground for illegality often stems from a denial of justice. However, this is not always the case. For instance, the following table lists certain investment arbitral awards that have found judicial expropriation and the grounds of illegality therein:


Award Name Grounds of illegality
 Saipem v. Bangladesh Abuse of rights
 Saipem v. Bangladesh and ATA v. Jordan  Violations of the New York Convention
 Rumeli v. Kazakhstan Arbitrariness
 Dissenting opinion in Ickale v. Turkmenistan Excessiveness/Lack of Proportionality
 Sistem, Tatneft, Middle East Cement The four “classic” conditions of unlawfulness:

·      Not for legitimate public purpose

·      Conducted in discriminatory manner

·      Undue process of law

·      Lack of prompt and adequate compensation

 Tatneft v. Ukraine Denial of justice
 Middle East Cement v. Egypt


Improper notice


It can be seen from the table that a number of the grounds of illegality mentioned are a consequence of procedural irregularity before national courts. Hence, it is true that the majority of judicial expropriations stem from a denial of justice. However, there are also instances where tribunals find judicial expropriation in cases where the merits of the national court’s decisions run contrary to international law, or more specifically treaty law. Two such cases are: (1) when the commercial awards are nullified unlawfully and, (2) when the issue of compensation is involved. In such cases, it becomes apparent that tribunals are willing to find judicial expropriation independent of the denial of justice.

The distinctiveness of judicial expropriation vis à vis denial of justice established early on by the Saipem award, has again been reinforced by a recent award in Eli Lily v. Canada.[19] In this award, the tribunal did not find any breach of NAFTA by the Canadian courts while revoking Eli Lily’s patents. However, the tribunal went on to discuss the correlation between Minimum Standard Treatment found in Article 1105 of NAFTA vis-à-vis Expropriation in Article 1110 of NAFTA. Article 1110(1)(c) defines an expropriation to be unlawful if it is not “in accordance with due process of law and Article 1105(1).” This in turn confers the minimum protection standards including Fair and Equal Treatment to the foreign investment.

In this context, the Tribunal drew a distinction between denial of justice and other conduct that is ”sufficiently egregious and shocking”, such as manifest arbitrariness and blatant unfairness.[20] Even though this distinction is drawn in the obiter dicta, the tribunal in Eli Lily opened the gateway for investment tribunals to consider cases of judicial expropriation independent in the absence of denial of justice claims. For instance, in Rumeli Telekom AS and Telsim Mobil Telekomunikasyon Hizmetleri AS v. Republic of Kazakhstan,[21] while considering the judicial valuation of the compensation, the tribunal held that it was “manifestly and grossly inadequate”[22] and therefore violative of effective, immediate market value compensation. Therefore, a claim for judicial expropriation was allowed by the Tribunal independent of denial of justice. Thus, the threshold for unlawfulness of a judicial expropriation is not merely limited to a denial of justice, but may also extend to other violations of international law.

Although the tribunal in Eli Lily correctly revived the separateness of the concepts, it introduced a new complication by requiring that the judicial decision should contain aggravated elements that are “egregious and shocking”. It is submitted that the tribunal should have refrained from introducing additional intricacy into the threshold requirements of judicial expropriation. As mentioned before, tribunals have always sought a higher threshold for holding judicial act to be judicial expropriation. The introduction of an “egregious and shocking” requirement by the Eli Lily Tribunal is indicative of the same trend where the tribunal requires something more in order to hold a judicial act to be judicial expropriation


c. Is exhaustion of local remedy an essential component of finding judicial expropriation?

As discussed in the preceding section, often the unlawfulness of the judicial expropriation might be a result of a denial of justice. As a matter of consensus, it is generally believed that exhaustion of local remedies is a substantive requirement for denial of justice claims.[23] The rationale behind this rule is that justice cannot be said to have been denied until the national courts of the host state get an opportunity to deliver a final judgment. Thus, in order to succeed in a denial of justice claim, the claimant must demonstrate to the tribunal that the judgment of the lower court was taken in appeal to the highest judicial authority of the host state or that such appeals were futile as part of the ‘judicial finality’ requirement. However, the pertinent question that arises now is: for claims of judicial expropriation that are independent of denial of justice, will the exhaustion of local remedy still be a necessary requirement?

There exist divergent answers to this question, both supported by legitimate considerations, and the debate is far from being settled. Those supporting the exhaustion of local remedies requirement argue that it gives the state an opportunity to address the injustice caused within its own legal system and that international remedies should be pursued only as a last resort. Contrariwise, it can also be argued that when tribunal dispenses justice on an international level without interference it is also derived from the state’s own sovereign power, which inter alia includes the right to expressly waiver certain aspects of sovereignty, and therefore the exhaustion of local remedies requirement is not mandatory for a finding of judicial expropriation.

However, the answer to this question lies in the extent to which judicial expropriation is connected to the denial of justice. If one is to follow Saipem, the tribunal therein explicitly mentioned that while dealing with judicial expropriation, denial of justice should not automatically be presupposed. Therefore, exhaustion of local remedies is not a necessary requirement for a finding of judicial expropriation. Ex contrario, does this mean that in cases where judicial expropriation does indeed stem from the denial of justice the exhaustion of local remedies requirement will have to be followed? It will be interesting to see the approach of tribunals on this issue—whether they apply the exhaustion of local remedies rule even where the denial of justice forms part of the greater wrong of judicial expropriation.


3. Fighting the ‘phobia’ of finding judicial expropriation

Arbitral tribunals seem reluctant to hold judicial acts as constituting ‘judicial expropriation’. The ease with which expropriation of the legislative or the executive organ of the State is identified, is not visible in cases where the alleged expropriatory act is by a national court of the host state. In fact, tribunals seem to treat expropriatory acts of the judiciary differently from the legislature and executive—not only should the court take the investment, but the process of taking should also involve serious procedural illegality.

Symptomatic of this phobia is the recent arbitral award in Krederi v. Ukraine.[24] The purchase of land blocks by the investor from the Kiev City Council was unwound by the National Court of Ukraine. The Court found deficiencies in the selling process by the City Council and therefore ruled for unwinding the transaction. The unwinding was, however, partial. The blocks of land were returned to the city council but the purchase price was not refunded to the investor. In what was a clear case of expropriation caused by the courts of Ukraine, the arbitral tribunal shirked away from holding it as judicial expropriation because merely taking the investment was not sufficient to constitute judicial expropriation. The tribunal required that the process of taking away should have suffered some serious procedural irregularity.[25] Since this case did not have such aggravated elements, there was no judicial expropriation.

However, there is no doctrinal backing for demanding a higher threshold in judicial expropriation vis-à-vis expropriation by the legislature or the executive. Usually, the demand for a higher threshold in judicial expropriation is justified on two grounds by its proponents: first, if such higher requirement did not exist, state responsibility for judicial conduct would be indeterminate and second, if such higher requirement did not exist it would appear akin to international appellate review of domestic decisions.

The first concern of indeterminate state responsibility envisages a situation where, for instance, a court orders that an investor turn over assets to another person. The investor could seek compensation citing judicial expropriation, which would be manifestly undesirable. However, this line of argument misunderstands the causal effect of taking over the assets. Although it might appear that the expropriation occurs due to causal conduct of the judiciary, the judicial conduct is not voluntary. Instead, all that the court is doing is applying the set of laws that have been enacted by the legislature to the legal dispute.

So does that mean that the legislature, and not the court, is responsible for the expropriation? In most situations, this is not the case because the causal conduct that leads to expropriation is the act of the investor itself. If an investor loses an investment as a consequence of the application of law, and the investor could have complied with the law in the first place, the causal responsibility for the expropriation does not rest upon the legislature or the executive but upon the investor itself. Hence, requiring a higher threshold for judicial expropriation in the fear of an indeterminate state responsibility is merely an illusory fear.

The second common reason for demanding a higher threshold for holding acts of the judiciary as expropriation is that the arbitral tribunal does not want to convey the impression that it is acting as an international appellate review authority over the domestic court’s decisions. Furthermore, there is an apprehension that instead of appealing the decision of a court of first instance, the investor might instead directly appeal to the arbitral tribunal claiming judicial expropriation. This apprehension was resonated in Saipem where the Claimant chose not to exercise its right to appeal two key decisions of the issue: first, by the Subordinate Judge of Dhaka and second, the Supreme Court of Bangladesh, and instead claimed judicial expropriation before the arbitral tribunal. This was interpreted as a creative way of circumventing the appellate structure of the host state and approaching the international tribunal.

In another prominent case, the English High Court set aside an SCC award which had ruled on the issue of whether the judgment of the Warsaw Court of Appeal, which was subsequently confirmed by the Polish Supreme Court, constituted an expropriation in violation of the BIT.[26] The High Court thereby rejected an attempt by the tribunal to exercise jurisdiction in examining the judgment of the Warsaw Court of Appeal.

However, this concern about tribunals sitting as international appellate review over domestic decisions is misplaced too. The core function of an appellate review is the correction of errors. The purpose of an appeal to an appellate court is to verify whether the lower court correctly interpreted and applied the substantive law. However, this is not the exercise that has to be undertaken with respect to judicial expropriation. The relevant question to be asked is: has the domestic court caused one of the consequences? Answering this question does not involve the tribunal correcting any errors and therefore cannot be equated to an appellate review. Hence, the fear of international appellate review is equally misplaced.



Judicial expropriations have wide-ranging implications for investment protection, but presently, the concept suffers from several ambiguities. As claims of judicial expropriation become more and more frequent post Saipem, an urgent need has emerged to define its conceptual contours clearly. It is now an established fact that expropriation can occur due to an act of the judiciary both as a standalone expropriation or part of a creeping expropriation. Often, such claims overlap with a denial of justice, and tribunals need to take a stand on whether the exhaustion of local remedies should be required before bringing a claim for judicial expropriation in such cases. Clarity with respect to the elements of judicial expropriation vis-à-vis the elements of an unlawful judicial expropriation would also go a long way in achieving uniformity of approach and doing away with the phobia.

*Harshvardhan Tripathi has recently completed his BA. LLB (Hons.) from NALSAR University of Law, Hyderabad and is joining India’s leading corporate law firm Cyril Amarchand Mangaldas this year.

[1] GEA Group Aktiengesellschaft v. Ukraine, ICSID Case No. ARB/08/16, Award, ¶ 236 (Mar. 31, 2011).

[2] PCA Case No. 2008-8, Award on the Merits, at ¶ 464-465 (July 29, 2014).

[3] Ben Mostafa, ‘The Sole Effects Doctrine, Police Powers and Indirect Expropriation under International
Law’ 15 Australian International Law Journal 267 (2008).

[4] R. Dolzer & C. Schreuer, Principles of International Investment Law 112-114 (2nd ed. 2012).

[5] Burlington v. Ecuador, ICSID Case No. ARB/08/5, Decision on Liability, ¶ 400-401 (Dec. 14, 2012).

[6] American Law Institute, U.S. Restatement of the Law (Third), Foreign Relations Law of the United
States (1987), § 712 comment (g) (1987).

[7] Dolzer & Schreuer, supra note 4, at 99.

[8] U. Kriebaum, Expropriation, in International Investment Law: A Handbook 959, 963-964 (Bungenberg et al. eds., 2015).

[9] Middle East Cement Shipping and Handling Co. S.A. v. Arab Republic of Egypt, ICSID Case No. ARB/99/6, Award, at ¶ 143 (Apr. 12, 2002).

[10] С. C. Hyde, International Law Chiefly as Interpreted and Applied by the United States at ¶ 491–92 (1922).

[11] B.E. Chattin (United States) v. United Mexican States, 4 R.I.A.A. 282, Award, at ¶ 286 (July 23, 1927).

[12] Harvard Research Draft, Art. 9, quoted in Ian Brownlie, Principles of Public International Law 506-507 (6th ed. 2003).

[13] Robert Azinian, Kenneth Davitian, & Ellen Baca v. The United Mexican States, ICSID Case No. ARB (AF)/97/2, Award, at ¶ 102-103 (Nov. 1, 1999).

[14] Loewen Group, Inc. and Raymond L. Loewen v. United States of America, ICSID Case No. ARB(AF)/98/3, Award, at ¶ 132 (June 26, 2003).

[15] Hamid G. Gharavi, Discord Over Judicial Expropriation, 33 ICSID Review 349 (2018).

[16] Azinian, supra note 13.

[17] Ares International S.r.l and MetalGeo S.r.l. v. Georgia, ICSID Case No. ARB/05/23, Award at ¶ 8.3.7 (Feb. 26, 2008).

[18] MNSS B.V. and Recupero Credito Acciaio N.V. v. Montenegro, ICSID Case No. ARB(AF)/12/8, Award, at ¶ 370 (May 4, 2016).

[19] Eli Lily and Company v. The Government of Canada, ICSID Case No. UNCT/14/, Final Award, at ¶ 223 (Mar. 16, 2017).

[20] Id.

[21] Rumeli Telekom AS and Telsim Mobil Telekomunikasyon Hizmetleri AS v. Republic of Kazakhstan, ICSID Case No. ARB/05/16, Award, at ¶ 702–704 (July 29, 2008).

[22] Id. at ¶ 706.

[23] Saipem v. Republic Of Bangladesh ICSID Case No. ARB/05/07, Award, at ¶151 (June 30, 2009).

[24] Krederi Ltd. v. Ukraine, ICSID Case No. ARB/14/17, Award, at ¶ 713 (July 2, 2018).

[25] Id.

[26] GPF GP S.à.r.l. v. Republic of Poland [2018] EWHC (Comm) 409 (Eng.).