ID-DRITT XXXV – Procedural requirements for counterclaims under the ICSID Convention: Consent and co

1. Introduction to counterclaims under the ICSID Convention

A counterclaim is an independent claim made by a respondent party against a claimant, which reacts and is incidental to the original claim, while also having an objective which extends beyond the mere dismissal of that original claim.¹ Therefore, a counterclaim has both a defensive quality which aims at defeating the primary claim, as well as an offensive quality with the simultaneous filing of an independent claim relating to the subject-matter of the dispute.²

The ICSID Convention provides for the bringing of counterclaims under its Article 46, which holds that:

Except as the parties otherwise agree, the Tribunal shall, if requested by a party, determine any incidental or additional claims or counterclaims arising directly out of the subject-matter of the dispute provided that they are within the scope of the consent of the parties and are otherwise within the jurisdiction of the Centre.³

Similarly, Rule 40 of the ICSID Arbitration Rules states:

Except as the parties otherwise agree, a party may present an incidental or additional claim or counter-claim arising directly out of the subject-matter of the dispute, provided that such ancillary claim is within the scope of the consent of the parties and is otherwise within the jurisdiction of the Centre.⁴

From the above, it can be inferred that there are three main procedural requirements for a counterclaim to succeed under the ICSID Convention, these being that the counterclaim: a) must arise directly out of the subject-matter of the dispute; b) be within the scope of the consent of the parties; and c) be within the jurisdiction of the International Centre for Settlement of Investment Disputes.⁵

As indicated in the travaux préparatoires of the ICSID Convention, the intent behind the counterclaims clause was not the expansion of the jurisdiction of the tribunal.⁶ Rather, it was to maintain continuity in the ongoing proceedings. This means keeping the same parties and the tribunal engaged in the proceedings that were initiated with the request for arbitration and the original claim, while also incorporating the counterclaim within the same proceedings, without needing to start a separate legal action.⁷ Indeed, counterclaims are considered as constituting a fundamental element of the respondent state’s right to put forward their case on a par with the investor.⁸ In this regard, counterclaims enhance procedural economy, ensure consistency in decisions, and thereby facilitate improved administration of justice through the establishment of mutual obligations for the parties.⁹

Nevertheless, counterclaims are not widely resorted to in investor-state arbitration. While it is typical for a respondent state to incorporate criticisms of the investor’s actions within its defence to claims in investment arbitration, such arguments are seldom presented as counterclaims seeking affirmative relief. The hesitation of states to frame their arguments as counterclaims might stem from their inclination to address affirmative claims within their domestic courts. Additionally, deemed limits to the jurisdiction of international tribunals to hear counterclaims might be a major contributing factor.¹⁰

Notably, the use of counterclaims in investment arbitration presents greater challenges in treaty claims when compared to contract claims.¹¹ This is mainly due to the asymmetrical nature of investment treaties, having as their main purpose the protection of investors’ rights.¹² On the other hand, contract claims are less problematic since they generally give either party the possibility of asserting claims for breaches made by the other party. These difficulties have contributed to making the upholding of jurisdiction by tribunals over counterclaims rare, and the success of counterclaims even rarer. Indeed, author Ana Vohryzek-Griest has described the history of host state counterclaims as ‘30 years of failure’.¹³

Against this background, this article will examine issues of jurisdiction and admissibility concerning counterclaims, with a specific focus on the requirements of consent and a connection between the original claim and the counterclaim. In exploring these issues, a number of important decisions will be examined with the purpose of understanding how tribunals have addressed counterclaims based on these criteria. Finally, these insights will be rounded off with some concluding observations on the broader role of counterclaims in investment arbitration.

2. The parties’ consent

2.1. Establishing consent

In line with Article 46 of the ICSID Convention and Rule 40 of the ICSID Arbitration Rules, for a tribunal to establish jurisdiction over a counterclaim, there must be the consent of both the host state and the investor. Since it is typically the host state that files the counterclaim, its consent does not usually pose any problems. Indeed, such consent might be included in the relevant investment treaty or in the counterclaim itself. The challenge is then obtaining or determining the existence of the consent of the investor to the counterclaim.¹⁴

As mentioned above, ICSID treaty cases with counterclaims differ from ICSID contract cases with counterclaims. Contracts that include an ICSID arbitration clause are bilateral, where both parties agree to fulfil the obligations specified in the contract and consent to ICSID arbitration if disputes arise. Therefore, in ICSID contract cases, the obligations and consent to arbitration are based solely on the contract between the host state and the investor itself.¹⁵ As such, the consent of the investor to any counterclaims raised by the host state is easily ascertainable.

On the other hand, the legal relationship between the parties in treaty cases is not strictly bilateral. In fact, a key feature of treaty counterclaims is the involvement of three parties: a) the investor’s home state; b) the host state; and c) the investor. Unlike in contract cases where the investor’s obligations, and therefore the basis for the counterclaims, are clearly outlined in the contract itself, counterclaims in treaty cases depend on identifying obligations imposed on the investor under the relevant investment treaty. This may prove difficult given that investment treaties often primarily focus on obligations for the host state. Hence, in treaty cases, consent to arbitration stems from the treaty and the investor’s arbitration request, not a contract.¹⁶ The question that arises here is whether the investor’s consent to arbitration, as expressed through the arbitration request, also amounts to consent to counterclaims. It is, however, difficult to imagine why an investor would agree to open themselves up to counterclaims at the stage of submitting the arbitration request, unless they would have given such consent in advance.¹⁷

Establishing the investor’s consent to counterclaims is not just a theoretical issue; it can have practical implications for the validity, recognition, and enforcement of the resulting award. Indeed, one of the grounds for annulling an ICSID award, as stated in Article 52 of the ICSID Convention, is the tribunal’s manifest excess of powers. If a tribunal adjudicates a claim outside the scope of the parties’ consent, it may be considered to have manifestly exceeded its authority.¹⁸ Nevertheless, none of the provisions under the ICSID Convention specify the manner in which the existence of the investor’s consent in treaty cases should be established. There are three options to consider in relation to this. The three options propose that consent can be secured or determined through: a) the investor’s explicit consent when filing a claim; b) the treaty’s language implying mutual consent for both claims and counterclaims; or c) the investor’s initiation of arbitration automatically implying consent to counterclaims.

The first option is that the investor consents to counterclaims upon the filing of their claim against the state. While this is the clearest option, it is uncertain how the state could secure the investor’s consent to a future counterclaim at this stage.¹⁹

The second option is by arguing that the investment treaty expressly provides for the consent of both parties to the treaty to the submission of both claims and counterclaims. In the absence of such express provision, it can alternatively be argued that consent could be implied from the language of the treaty’s consent clause, where the use of phrases such as “any dispute” or “all disputes” could be taken to mean that consent is given in relation to counterclaims too. This analysis was in fact made in Saluka v. Czech Republic (“Saluka”)²⁰ and Paushok v. Mongolia (“Paushok”).²¹ In this option, the consent by the state in the treaty to the submission of both claims and counterclaims to ICSID arbitration, then becomes the investor’s consent with their submission of a request for arbitration.

This approach is based on the idea that the arbitration agreement has two instruments; an offer and an acceptance, as highlighted in Roussalis v. Romania (“Roussalis”).²² First, there is an offer to resolve investment disputes through claims and counterclaims under ICSID arbitration – which offer is materialised through the investment treaty entered into between the host state and the investor’s home state. Second, there is an acceptance by the investor of the offer to resolve the dispute through ICSID arbitration – which acceptance materialises through the submission for the request for arbitration.²³

The core issue with this approach is determining specifically whether the investor’s readiness to engage in ICSID arbitration by filing a claim also implies their consent to the submission of counterclaims by the host state. Two primary concerns arise in addressing this issue. The first pertains to the clarity of the treaty’s language, and particularly to whether the treaty unambiguously states that disputes can be presented as both claims and counterclaims, or whether the treaty’s wording could be considered as indirectly encompassing both claims and counterclaims. The second concern questions whether the investor needs to explicitly state their consent for the host state to submit a counterclaim.²⁴

The third option for establishing the investor’s consent is based on the premise that the state and the investor both give separate consents for arbitration, in the treaty and in the request for arbitration. This option suggests that even if the treaty’s wording is not very clear, the simple act of initiating the arbitration process implies the investor’s consent to any counterclaims that may be raised by the respondent state.²⁵ Therefore, according to this option, and as held in Goetz v. Burundi II (“Goetz II”)²⁶, the language of the treaty is not very important.²⁷ Moreover, as outlined by Professor Michael Reisman in his dissenting opinion in Roussalis, the necessary consent would be ipso facto imported upon the institution of ICSID arbitration by the investor.²⁸

The main concern with this approach is whether ICSID tribunals would feel comfortable with accepting that an investor’s consent to counterclaims is automatically incorporated through the initiation of the arbitration process, without a detailed interpretation of the treaty.²⁹ In fact, a possible interpretation of Article 25 of the ICSID Convention, which requires that the disputing parties provide written consent to submit their dispute to ICSID arbitration, is that, under Article 46 of the ICSID Convention (which is based on Article 25 of the ICSID Convention), consent to the submission of an ICSID counterclaim must also be explicit and in writing, rather than implied.³⁰

2.2. ICSID case law on consent in investment treaty counterclaims

2.2.1. Early ICSID treaty cases

In early ICSID treaty cases involving counterclaims – such as Genin v. Estonia³¹, Mitchell v. Democratic Republic of the Congo³², and Desert Line v. Yemen³³ – tribunals refrained from examining whether the counterclaims met the jurisdictional requirements of the ICSID Convention.³⁴ This is despite their clear obligation to do so, as outlined in Article 46 of the ICSID Convention and the Convention’s travaux préparatoires.³⁵ On the other hand, the tribunals in Sempra v. Argentina³⁶ (“Sempra”) and Hamester v. Ghana³⁷ (“Hamester”) began to explore this issue more deeply, setting the stage for later, more detailed analysis.

The tribunal in Sempra expressly referred to its obligation to examine whether the counterclaims met the jurisdictional requirements of the ICSID Convention while dealing with Argentina’s arguments regarding the losses it expected the investor to bear. The tribunal stated that Argentina had the right to raise a counterclaim, provided that any allegations fell within the tribunal’s jurisdiction and could potentially constitute a breach of the relevant bilateral investment treaty (“BIT”). However, the tribunal observed that Argentina did not exercise this right to submit a counterclaim.

Moreover, in Hamester the tribunal briefly considered whether the parties’ consent to ICSID arbitration under Article 46 of the ICSID Convention might also include consent to counterclaims by the host state. Although Ghana had not provided enough legal foundation for its counterclaim, the tribunal hinted that, based on a structural reading of Article 12 of the Germany-Ghana BIT, the possibility of hearing the counterclaim existed.³⁸ The tribunal first recognised that the arbitration clause in Article 12(1) of the BIT was limited to disputes concerning the treaty obligations of one party (the host state) in relation to an investment of a national or company of the other party (the home state). However, it also acknowledged that Articles 12(3) and (4) of the BIT, through the use of the phrase ‘aggrieved party’, allow either party, including the host state, to submit disputes to arbitration.³⁹ Although the tribunal ultimately denied jurisdiction over the counterclaim, it preliminarily analysed the scope of consent to arbitration from the BIT’s language. This suggests that the tribunal deemed that the counterclaim could potentially have been heard based on an analysis of the relevant BIT provisions, and not due to any implicit or automatic right to submit counterclaims.⁴⁰

Sempra and Hamester serve as a significant prelude to the more recent and pivotal ICSID treaty counterclaim cases. Since 2011, investment treaty cases have thoroughly examined the jurisdictional requirements for hearing counterclaims. A review of such cases, nevertheless, shows that the options for establishing the investor’s consent discussed in Section 2.1 have led to diverging interpretations by tribunals as to whether general consent to ICSID arbitration could be interpreted as also constituting ipso facto consent to counterclaims. The tension principally lies between: a) the view held in Roussalis, Saluka, Paushok, and Inmaris v. Ukraine (“Inmaris”)⁴¹ that the presence, or otherwise, of consent to counterclaims in the arbitral agreement necessitates a case-by-case examination; and b) the view held in Goetz II where the doctrine of ipso facto importation of consent is followed. This tension between interpretations will be more effectively illustrated in the following sections through a discussion of primary cases on the matter, namely Roussalis and Goetz II.

2.2.2. Roussalis v. Romania

In Roussalis, the majority of the tribunal found that the parties had not consented to having the state’s counterclaims arbitrated. This decision was based on the narrow wording of the dispute resolution clause of the Greek-Romanian BIT. The tribunal held that through the wording ‘disputes… concerning an obligation of the latter’,⁴² Article 9(1) of the Greek-Romanian BIT limited the jurisdiction of the tribunal to claims brought by investors with regards to obligations of the host state. Because of this, it was found that counterclaims by the respondent state with regards to obligations of the investor were not provided for under the underlying BIT.⁴³ The tribunal also held that when the relevant BIT places obligations solely on the contracting states and designates the BIT itself as the applicable law, the tribunal lacks jurisdiction over counterclaims.⁴⁴ It clarified that for its jurisdiction to extend towards state’s counterclaims, ‘the arbitration agreement should refer to disputes that can also be brought under domestic law.’⁴⁵ Moreover, it was held that the presence of an umbrella clause in the BIT did not alter this stance.⁴⁶ As a result, the tribunal rejected jurisdiction over the counterclaim.

In this case, the tribunal meticulously adhered to the principles of textual interpretation, interpreting terms within the underlying BIT based on their ordinary meanings.⁴⁷ However, one arbitrator, Professor Reisman, issued a dissenting opinion arguing that the tribunal should have recognised its jurisdiction to hear the counterclaim. Reisman noted:

[I]n my view, when the States Parties to a BIT contingently consent, inter alia, to ICSID jurisdiction, the consent component of Article 46 of the Washington Convention is ipso facto imported into any ICSID arbitration which an investor then elects to pursue.⁴⁸

Moreover, Reisman highlighted the importance of recognising that the tribunal’s jurisdiction over the counterclaim benefits both the respondent state and the investor. He argued that by rejecting ICSID jurisdiction over counterclaims, a neutral tribunal, chosen by the investor, essentially compels the respondent state to pursue its claims in its domestic courts, where the investor, initially seeking an external forum, becomes the defendant. This could lead to situations where, following an adverse judgement, the investor initiates another BIT claim. In Reisman’s opinion, apart from resulting in added costs and inefficiencies, such outcomes seem counterintuitive to the purpose of international investment law.⁴⁹ Considering all this, one notes that while the majority of the tribunal in Roussalis based its decision with regards to jurisdiction primarily on the phrasing of the dispute resolution clause of the underlying BIT, the dissenting arbitrator relied on Article 46 of the ICSID Convention and emphasised broader policy considerations.⁵⁰

2.2.3. Goetz v. Burundi II

Shortly after the decision in Roussalis, Goetz II sparked a new debate. This is largely because both cases dealt with the same issue but reached completely different outcomes. The Goetz II tribunal adopted Professor Reisman’s approach – relating to the doctrine of ipso facto importation of consent.⁵¹

In this case, the tribunal determined that Burundi’s counterclaim satisfied the requirements under Article 46 of the ICSID Convention and Rule 40 of the ICSID Arbitration Rules. It reasoned that by entering into the BIT, which includes in its Article 8 a provision for the settlement of disputes through ICSID arbitration, Burundi accepted that disputes would be settled according to the conditions and procedures contained in the ICSID Convention. This encompasses presenting ancillary claims or counterclaims during the proceedings. The tribunal further held that by accepting the offer to arbitrate, the claimants also accepted the same framework.⁵²

The tribunal emphasised that the absence of a provision in the BIT explicitly empowering the tribunal to examine counterclaims was irrelevant. Citing Professor Reisman’s dissenting view in Roussalis, it affirmed that when state parties to a BIT consent to ICSID jurisdiction, the consent element of Article 46 of the ICSID Convention is automatically applied to subsequent ICSID arbitrations pursued by the investor.⁵³ The tribunal also concluded that diversion from this approach would contradict both the letter and the spirit of the ICSID Convention. This is because it would encourage a host state to resort to domestic courts for settling counterclaims that relate to the same dispute presented before the ICSID tribunal by the investor. In turn, this could compel dissatisfied investors to challenge the judgements obtained by states in this manner through fresh requests for arbitration, which would complicate the settlement of investment disputes.⁵⁴ Given this rationale, the tribunal affirmed its jurisdiction over the counterclaim and held it admissible.

2.2.4. Comparing the diverging interpretations in Roussalis and Goetz II

From a strict textual analysis, the BIT in Goetz II appears to offer more flexibility for a tribunal to determine its jurisdiction over the host state’s counterclaims than the one in Roussalis. The Belgium-Burundi BIT holds that the state parties consent to ‘any’ investment dispute, implying that this encompasses counterclaims directly related to the original dispute’s subject matter.⁵⁵ Contrastingly, there were challenges to reaching a similar conclusion as the one in Goetz II when interpreting the language of the Greek-Romanian BIT. Specifically, such BIT raises questions about whether initiating arbitration under it also means consenting to the tribunal’s jurisdiction to consider counterclaims from the host state related to obligations owed by the investor.⁵⁶

Having said this, the Goetz II tribunal did not cite the contrasting treaty language between the Belgium-Burundi BIT and the Greece-Romania BIT as justification for its conclusions and departure from Roussalis.⁵⁷ Indeed, while the majority in the Roussalis tribunal considered that Article 9(1) of the Greece-Romania BIT restricts jurisdiction to claims by investors concerning obligations of the host state,⁵⁸ the Goetz II tribunal grounded its jurisdiction on Article 46 of the ICSID Convention and Rule 40 of the ICSID Arbitration Rules.⁵⁹

This divergence is noteworthy, and the trajectory of this issue in the future remains uncertain. Having said this, so far, most tribunals appear to have preferred the approach of examining whether the treaty provides for consent to counterclaims by analysing its language.⁶⁰

3. The connection between the claim and the counterclaim

3.1. The requirement of a close connection

Apart from the consent requirement, it is generally accepted that there must be a connection between the claim and the counterclaim in an investment arbitration. In this regard, Article 46 of the ICSID Convention and Rule 40 of the ICSID Arbitration Rules require that a counterclaim arises ‘directly out of the subject-matter of the dispute’.⁶¹ This requirement is clarified in the Notes to the ICSID Arbitration Rules, which hold that:

[t]he test to satisfy this condition is whether the factual connection between the original and the ancillary claim is so close as to require the adjudication of the latter in order to achieve the final settlement of the dispute.⁶²

A distinction is not always drawn by tribunals as to whether the connection requirement is a jurisdictional or admissibility requirement. Nevertheless, as will be discussed, various decisions have established this requirement as one of the prerequisites for the acceptance of jurisdiction by a tribunal over a counterclaim⁶³ or for the counterclaim to be deemed admissible.

3.2. Establishing a close connection

3.2.1. Challenges in establishing a close connection

Assessing whether a counterclaim arises directly from the subject-matter of the dispute involves first identifying the core subject-matter of the dispute and then determining whether the counterclaim is connected to that same subject-matter. To identify the subject-matter, a logical approach is to analyse the dispute and pinpoint the investments at issue in such dispute. This may include various types of assets typically outlined in investment treaties, such as contracts, concessions, shares, and interests in companies.⁶⁴

Certain counterclaims make it more straightforward for the tribunal to determine that they are closely connected to, or that they arise directly out of the subject-matter of, the original dispute. In fact, the challenge of establishing whether the counterclaim stems from the subject-matter of the original dispute is generally straightforward when the dispute involves a contract allegedly breached by the host state, and the counterclaim asserts that it was, in fact, the investor who breached the contract. Similarly, if the counterclaim relates to damages or costs arising from alleged non-compliance with a contract relied on by both parties, the link is clear. These scenarios were indeed evident in Inmaris and Goetz II.

In Inmaris, the tribunal noted that the respondent’s counterclaim for costs incurred in maintaining a vessel during the winter months in Ukraine was tied to the original dispute submitted by the claimant, which concerned a commercial use agreement over a vessel.⁶⁵ As such, it was clear that the counterclaim arose directly from the subject-matter of the original dispute.⁶⁶ In Goetz II, both the claim and the counterclaim were connected by a common instrument: a bank operating certificate. The claimant argued that the unfair revocation of the certificate violated the investment treaty, while the respondent counterclaimed that the claimant’s bank had failed to meet its obligations under the same certificate. Consequently, the tribunal found a direct link between the subject-matter of the dispute and the counterclaim, both revolving around the bank operating certificate.⁶⁷

However, the complexity increases when the dispute involves an ICSID treaty claim rather than an ICSID contract claim, and the counterclaim is not based on an instrument, such as a contract, already invoked by the investor.⁶⁸ Tribunals encounter even more difficulty when the counterclaims raised by the state respondent allege breaches of domestic law. In fact, the decisions of tribunals on this matter have been diverse. As will be explored in more detail in the following section, in cases like Roussalis, Saluka, and Paushok, the tribunals held either that; a) the dispute and the counterclaim were distinct, or b) there was a failure to meet the close connection criterion when the respondent states alleged breaches of their domestic laws.⁶⁹

3.2.2. The examination of the connection requirement by tribunals

Although Saluka is a UNCITRAL case, its relevance to the current discussion stems from its extensive citation in other decisions concerning the connection requirement for counterclaims and the substantial criticism it garnered. In Saluka, the tribunal rejected jurisdiction over the counterclaim based on a lack of a close connection between the original claim and the counterclaim.⁷⁰ Additionally, it declined jurisdiction as the counterclaim related to obligations under Czech law rather than the underlying BIT,⁷¹ and because the Share Purchase Agreement allegedly breached referred to a different forum.⁷²

In elucidating that a close connection is a ‘condition customarily governing the relationship between claims and counterclaims’,⁷³ the tribunal highlighted that this requirement is a prevalent legal principle found in the ICSID Convention, the UNCITRAL Rules, and the Iran-US Claims Settlement declaration, even though these instruments employ distinct counterclaim terminology.⁷⁴ On this basis, it relied heavily on previous decisions by ICSID tribunals and the Iran-US Claims Tribunal, despite that the arbitration was governed by the UNCITRAL Arbitration Rules.⁷⁵

Based on the arbitration provision within the relevant BIT, the tribunal determined that the parties consented to the bringing of counterclaims for its consideration. However, it concluded that the specific disputes prompting the respondent’s counterclaim were either addressed by a distinct arbitration agreement within the parties’ contractual agreement or lacked a close enough connection to the subject matter of Saluka’s original claim.⁷⁶ Consequently, the tribunal ruled that it lacked jurisdiction over those specific counterclaims.

The restrictive approach taken in Saluka faced criticism, mainly due to the tribunal’s heavy reliance on the rationale set out in Klöckner v. Cameroon (“Klöckner”). In Klöckner, the tribunal had concluded that a close connecting factor existed between the claim and the counterclaim since their subject-matters were ‘indivisible’ and ‘interdependent’.⁷⁷ Taking a straightforward interpretation and by endorsing the terminology used in Klöckner, the tribunal in Saluka determined that the dispute underlying the counterclaim should be resolved using Czech law procedures rather than procedures arising from treaty-based investment protection. This decision was taken because the counterclaim was not considered as: a) forming an ‘indivisible whole’ with the original claim; b) invoking obligations that shared with the original claim ‘a common origin, identical sources, and an operational unity’; or c) being ‘interdependent’ with the original claim for the purposes of achieving a singular objective.⁷⁸

Another reason for the criticism directed towards Saluka was that the UNCITRAL Rules which governed the proceedings did not include the specific requirements of indivisibility and interdependence as highlighted in Klöckner,⁷⁹ making the test applied in Saluka too stringent. As will be discussed in Section 3.3., insisting on strict legal unity between the claim and the counterclaim for the purposes of the connection requirement seems unwarranted when taking into consideration the predominant unilateral nature of obligations within investment treaties.⁸⁰ This is why critiques of Saluka argue that the tribunal should have prioritised the terms of the investment treaty, which refer to claims ‘concerning an investment’.⁸¹ When considering such wording, one notes that there is no indication that a stricter interdependence test with the original claim is essential.⁸² Therefore, the broad declarations from Klöckner should not have been viewed as a ‘general expression of the law’.⁸³

Notwithstanding the criticisms received, the principles established in Saluka still influenced subsequent rulings. For example, the rationale of the Saluka tribunal was extensively cited in Paushok. In this case, the arbitrators determined that they must examine the jurisdiction to entertain counterclaims based on whether there exists a close connection between the counterclaims and the original claim, or whether the counterclaims fall under the general laws of the respondent state.⁸⁴ The view taken in Paushok was that a counterclaim rooted in a host state’s domestic law cannot be closely linked to an investor’s claim. Therefore, it was found that there is no justification for simultaneously addressing such a claim and counterclaim together.⁸⁵ On this basis, the tribunal found no close connection⁸⁶ and, as a result, declined jurisdiction over the counterclaims.

The same approach was also followed by the tribunal in Amto v. Ukraine⁵⁰ (“Amto”). In this case, the tribunal noted that its jurisdiction over a state party’s counterclaim under an investment treaty depends on the specific dispute resolution provisions of the treaty, the nature of the counterclaim, and the connection between the counterclaims and the claims in the arbitration.⁵⁰ The tribunal highlighted that, according to Article 26(6) of the Energy Charter Treaty, the applicable law includes the treaty itself, along with the relevant rules and principles of international law. Therefore, it held that for the counterclaim to be successful it must be based on the same legal basis.⁵⁰

3.3. The factual and legal basis for the connection between the claim and the counterclaim

From the above discussion, it is clear that according to Saluka, Paushok, and Amto the connection between the claim and the counterclaim must not only be on a factual basis, but also on a legal basis, meaning that the claim and the counterclaim must arise out of the same legal instrument or cause of action.⁹⁰ However, it is noted that while the investor’s claim is usually based on a breach of a treaty obligation, counterclaims are often based on the investor’s violation of contractual provisions or the law of the host state. Since the obligations in investment treaties are mostly unilateral, states usually have no other choice than to base their counterclaims on a different source.⁹¹ This means that legal connectedness rarely exists in practice.⁹² Therefore, a strict application of the criteria for admitting counterclaims in this manner, that is, by requiring an indivisible connection between a claim and a counterclaim on the basis of both facts and law, creates a problem. This makes the connection requirement too stringent, and it will almost be impossible for the tribunal to admit a state counterclaim in investment treaty arbitration by following this approach.

Fortunately, it seems that both arbitral practice and scholarship have shifted to a point where strict legal identity between claims is no longer necessary for the requirement of a close connection to be fulfilled.⁹³ According to one view, what tribunals should deal with is counterclaims related to the same investment, this being the subject-matter of the dispute.⁹⁴ This appears to have indeed been the approach in Goetz II, Urbaser v. Argentina⁹⁵ (“Urbaser”), and Burlington v. Ecuador⁹⁶ (“Burlington”).

In Goetz II, as discussed above, the tribunal determined that the counterclaim stemmed from the same bank operating certificate, which was also central to the investment under contention, directly linking it to the dispute’s subject matter.⁹⁷ The same approach was followed in Urbaser, wherein the respondent raised a counterclaim that was not based on the BIT governing the claimant’s claim, but instead relied on human rights law, specifically the international right to water.⁹⁸ In this case, the tribunal found that it had jurisdiction to hear the counterclaim and deemed it admissible on the basis of the existence of a factual connection between the original claim and the counterclaim, both of which revolved around the same investment, or the alleged lack of investment, in relation to the same concession.⁹⁹ Here, the tribunal also emphasised that it would be entirely inconsistent to decide on the claimant’s claim concerning their investment in one manner and then address the counterclaim related to the same investment through a separate proceeding in a different manner.¹⁰⁰ Furthermore, it was stressed that such a potential inconsistency in outcomes is incompatible with the principle of reasonable administration of justice.¹⁰¹ Similarly, in Burlington, the tribunal confirmed that the respondent’s counterclaims satisfied the requirements of Article 46 of the ICSID Convention since they ‘arose directly out of the subject-matter of the dispute, namely Burlington’s investment”.¹⁰²

An alternative view suggests that the close connection criterion is fulfilled when there exists a factual nexus between the original claim and the counterclaim.¹⁰³ This view is supported by Pierre Lalive and Laura Halonen, who hold that:

there is no reason to imply a requirement of a legal connection into the term ‘subject-matter’… It should be enough that the claims and counterclaims arise directly out of the same subject-matter as a matter of fact.¹⁰⁴

To bolster their stance, Lalive and Halonen cite the Notes to the ICSID Arbitration Rules which state that the connection requirement is satisfied when there is a close factual connection between the original claim and the counterclaim.¹⁰⁵ This view is also shared by Zachary Douglas who maintains that ‘the scope of a tribunal’s jurisdiction to determine counterclaims by the host State ultimately depends upon a factual nexus between the dispute submitted by the claimant and the counterclaim’.¹⁰⁶ This perspective, therefore, aligns more closely with the wording of Article 46 of the ICSID Convention.¹⁰⁷

4. Conclusion

The issue of counterclaims in investment arbitration remains complex and contentious, as demonstrated by the evolving jurisprudence and ongoing policy debates. While it is clear that respondent states have the right to raise counterclaims, successfully doing so – both in terms of jurisdiction and on the merits – has been rare. The jurisprudential landscape underscores two pivotal criteria that tribunals must grapple with to assert jurisdiction over a counterclaim: a) the mutual consent of the parties involved and b) a close connection between the original claim and the counterclaim. However, the interpretation of these requirements, particularly in treaty cases, exhibits notable discrepancies across tribunals.¹⁰⁸

A significant determinant for tribunals’ interpretations lies in the wording of individual treaties and the breadth of their dispute resolution clauses. The differing approaches in cases such as Roussalis and Goetz II highlight how tribunals’ interpretations of treaty language can greatly influence the fate of a counterclaim. However, the inherent asymmetry of investment treaties, which primarily focus on protecting the rights of investors, is a key challenge in this regard. Indeed, this imbalance makes it difficult for states to bring counterclaims, as they often lack treaty provisions that impose obligations on investors. Even when such provisions exist, limitations in the treaty’s jurisdictional scope – whether ratione materiae or ratione personae – often prevent tribunals from hearing counterclaims. Apart from this, the most challenging conundrum emerges when one considers broader implications of hearing the counterclaims, such as certain policy considerations. In this regard, it is noteworthy that the investor-state dispute resolution (“ISDS”) system faces increasing scrutiny for its perceived tilt in favour of investors, constraining states’ possibility of asserting counterclaims effectively.¹⁰⁹

To recalibrate this perceived imbalance, a plausible solution entails revising existing treaties. By infusing explicit provisions that allow counterclaims and detail investor obligations, a more equitable framework might emerge. Yet, such treaty modifications demand meticulous deliberation and are inevitably time intensive. As this evolution unfolds, it is paramount to heed esteemed voices like Professor Reisman’s, who underscores the significance of policy nuances. Amidst this backdrop, the intrinsic merits of counterclaims – fostering procedural efficiency, bolstering legitimacy, and upholding the sanctity of the rule of law – should remain central to deliberations by tribunals. The inclusion of counterclaims in investment arbitration could potentially enhance the legitimacy of the ISDS system by allowing both parties to present their claims and avoid inconsistencies in parallel proceedings. Nevertheless, it is essential to navigate this matter with caution, and keep in mind that such policy considerations should not be given priority over clear treaty language preventing or limiting counterclaims.¹¹⁰

It is also relevant to note that, on the other side of the coin, there are significant policy arguments for limiting counterclaims by states in investment arbitration. Since state claims do not typically arise from obligations in BITs (since these do not generally place direct obligations on investors), counterclaims might be viewed as circumventing carefully negotiated contractual dispute resolution provisions. Moreover, counterclaims could draw tribunals into disputes governed primarily by local law rather than international law, raising concerns about tribunals’ legitimacy, as they may lack the same expertise in national law as local courts. Additionally, allowing extensive counterclaims could deter investors from pursuing arbitration against states, thereby potentially undermining the very purpose of BITs as a tool to reassure investors by offering them a reliable forum to bring their claims when disputes arise.¹¹¹

In conclusion, while counterclaims offer a potential means to correct the perceived imbalance in the ISDS system, their broader adoption requires careful treaty drafting and consideration of both the procedural and substantive challenges they present. The balance between efficiency, fairness, and the risk of undermining the ISDS system’s legitimacy will remain at the heart of this debate.

Disclaimer: Ganado Advocates is responsible for contributing to this article but was not in any way involved as legal advisor for the parties discussed herein. This article was first published in ‘ID-Dritt’ in 2025.

¹ Anna De Luca and Crina Baltag, ‘Counterclaims in Investment Arbitration: Reflections on UNCITRAL WG III Reform’ (Kluwer Arbitration Blog, 5 November 2021) <https://arbitrationblog.kluwerarbitration.com/2021/11/05/counterclaims-in-investment-arbitration-reflections-on-uncitral-wg-iii-reform/> accessed 24 October 2024.

² Elise Ruggeri Abonnat, ‘Counterclaims’ (Jus Mundi, 31 July 2023) <https://jusmundi.com/en/document/publication/en-counterclaims#:~:text=A> accessed 24 October 2024.

³ Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (opened for signature 18 March 1965, entered into force 14 October 1966) 575 UNTS 159 (ICSID Convention), art 46.

⁴ ICSID, Rules of Procedure for Arbitration Proceedings (Arbitration Rules), rule 40.

⁵ Anne K. Hoffmann, ‘Chapter 36: Counterclaims’, in Meg Kinnear, Geraldine R. Fischer, et al. (eds), Building International Investment Law: The First 50 Years of ICSID (Kluwer Law International 2015) 505, 508.

⁶ ICSID, History of the ICSID Convention. Documents Concerning the Origin and the Formulation of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States, Washington DC, ICSID, Vol. IV, 160; José Antonio Rivas, ‘ICSID Treaty Counterclaims: Case Law and Treaty Evolution’ in Jean E. Kalicki and Anna Joubin-Bret (eds), Reshaping the Investor-State Dispute Settlement System: Journeys for the 21st Century (Nijhoff International Investment Law Series 2015) 791.

⁷ ICSID (n 6) 103; Rivas (n 6) 781.

⁸ Crina Baltag and Ylli Dautaj, ‘Regime Interaction in Investment Arbitration: Counterclaims’ (Kluwer Arbitration Blog, 11 January 2022) <https://arbitrationblog.kluwerarbitration.com/2022/01/11/regime-interaction-in-investment-arbitration-counterclaims/> accessed 24 October 2024.

⁹ ibid.

¹⁰ Jean E. Kalicki, ‘Counterclaims by States in Investment Arbitration’ (IISD, 14 January 2013) <https://www.iisd.org/itn/en/2013/01/14/counterclaims-by-states-in-investment-arbitration-2/> accessed 24 October 2024.

¹¹ Ruggeri Abonnat (n 2).

¹² ibid.

¹³ Ana Vohryzek-Griest, ‘State Counterclaims in Investor-State Disputes: A History of 30 Years of Failure’ (2009) 15 International Law: Revista Colombiana de derecho Internacional 83.

¹⁴ Rivas (n 6) 782.

¹⁵ ibid., 780.

¹⁶ ibid.

¹⁷ ibid., 782.

¹⁸ Hoffmann (n 5) 508.

¹⁹ Rivas (n 6) 782.

²⁰ Saluka Investments B.V. v. Czech Republic, UNCITRAL, Decision on Jurisdiction over the Czech Republic’s Counterclaim (7 May 2004) (Saluka v. Czech Republic), para 39.

²¹ Sergei Paushok and others v. The Government of Mongolia, UNCITRAL, Award on Jurisdiction and Liability (28 April 2011) (Paushok v. Mongolia), para 689.

²² Spyridon Roussalis v. Romania, ICSID Case No ARB/06/01, Award (7 December 2011) (Roussalis v. Romania), para 866.

²³ Rivas (n 6) 783.

²⁴ ibid.

²⁵ ibid.

²⁶ Antoine Goetz and others v. Republic of Burundi, ICSID Case No ARB/01/2, Award (21 June 2012) (Goetz v. Burundi II), para 278.

²⁷ ibid., para 279.

²⁸ Spyridon Roussalis v. Romania, ICSID Case No ARB/06/1, Declaration of Professor W. Michael Reisman (28 November 2011) (Roussalis v. Romania, Declaration of Prof. Reisman).

²⁹ Rivas (n 6) 784.

³⁰ ibid

³¹ Genin v. Estonia, ICSID Case No ARB/99/2, Award (25 June 2001).

³² Mitchell v. Democratic Republic of the Congo, ICSID Case No. ARB/99/3, Excerpts of Award (9 February 2004).

³³ Desert Line v. Yemen, ICSID Case No. ARB/05/17 (6 February 2008).

³⁴ Rivas (n 6) 793.

³⁵ ICSID, History of the ICSID, Vol. II, 810.

³⁶ Sempra Energy Int’l v. Argentine Republic, ICSID Case No. ARB/02/16, Award (28 September 2007).

³⁷ Gustav F W Hamester GmbH & Co KG v. Republic of Ghana, ICSID Case No. ARB/07/24, Award (18 June 2010) (Hamester v. Ghana).

³⁸ Rivas (n 6) 796.

³⁹ Hamester v. Ghana, paras 353-354.

⁴⁰ Rivas (n 6) 797

⁴¹ Inmaris Perestroika Sailing Maritime Services GmbH and others v. Ukraine, ICSID Case No ARB/08/8, Excerpts of Award (1 March 2012) (Inmaris v. Ukraine).

⁴² Greece-Romania BIT (23 May 1997), art 9(1).

⁴³ Roussalis v. Romania, para 869.

⁴⁴ ibid., paras 870-871.

⁴⁵ ibid., para 871.

⁴⁶ ibid., para 873.

⁴⁷ Rivas (n 6) 801.

⁴⁸ Roussalis v. Romania, Declaration of Prof. Reisman.

⁴⁹ Roussalis v. Romania, Declaration of Prof. Reisman.

⁵⁰ Hoffmann (n 5) 507.

⁵¹ ibid., 515.

⁵² Goetz v. Burundi II, para 278.

⁵³ ibid., para 279.

⁵⁴ ibid., para 280.

⁵⁵ Rivas (n 6) 803.

⁵⁶ ibid.

⁵⁷ ibid.

⁵⁸ Roussalis v. Romania, para 869.

⁵⁹ Goetz v. Burundi II, para 280.

⁶⁰ Rivas (n 6) 809.

⁶¹ ICSID Convention, art 46; Arbitration Rules, rule 40.

⁶² Notes to the ICSID Arbitration Rules (1968), Note B (a) to Rule 40, reprinted in 1 ICSID Reports 63, 100 (1993).

⁶³ Hoffmann (n 5) 513.

⁶⁴ Rivas (n 6) 784.

⁶⁵ Inmaris v. Ukraine, paras 270, 432.

⁶⁶ Rivas (n 6) 810.

⁶⁷ Goetz v. Burundi II, para 285.

⁶⁸ Rivas (n 6) 785.

⁶⁹ ibid.

⁷⁰ Hoffmann (n 5) 513.

⁷¹ Saluka v. Czech Republic, paras 47-82; Rivas (n 6) 805.

⁷² Saluka v. Czech Republic, paras 47-82; Christoph H. Schreuer and others, The ICSID Convention: A Commentary (Cambridge University Press 2009) 752.

⁷³ Saluka v. Czech Republic, paras 47-82.

⁷⁴ ibid., para 76.

⁷⁵ Mees Brenninkmeijer and Fabien Gélinas, ‘Counterclaims in Investment Arbitration: Towards an Integrated Approach’ in Meg Kinnear and Campbell McLachlan (eds), ICSID Review – Foreign Investment Law Journal (2023) 38 Oxford University Press 567.

⁷⁶ Saluka v. Czech Republic, para 61; Hoffmann (n 5) 513.

⁷⁷ Klöckner Industrieanlagen-Anlagen GmbH and others. v. United Republic of Cameroon and Sociėtė Camerounaise des Engrais, ICSID Case No ARB/81/2,Award (21 October 1983), paras 17 and 65.

⁷⁸ Saluka v. Czech Republic, para 79; Brenninkmeijer and Gélinas (n 75).

⁷⁹ Pierre Lalive and Laura Halonen, ‘On the Availability of Counterclaims in Investment Treaty Arbitration’ (2011) 2 CYIL 141; Zachary Douglas, The International Law of Investment Claims (1st edn, Cambridge University Press 2009) 260-263;

Hoffmann (n 5) 513.

⁸⁰ Lalive and Halonen (n 79).

⁸¹ ibid.; Dafina Atanasova, Adrián Martínez Benoit and Josef Ostřansky, ‘The Legal Framework for Counterclaims in Investment Treaty Arbitration’ (2014) 31 J Intl Arb 357, 381.

⁸² Atanasova, Martínez Benoit and Ostřansky (n 81) 383.

⁸³ ibid.

⁸⁴ Paushok v. Mongolia, para 693.

⁸⁵ Brenninkmeijer and Gélinas (n 75).

⁸⁶ Paushok v. Mongolia, paras 694-699.

⁸⁷ Limited Liability Company AMTO v. Ukraine, SCC Case No. 080/2005, Final Award (26 March 2008).

⁸⁸ ibid., para 118.

⁸⁹ ibid.

⁹⁰ Saluka v. Czech Republic, paras 79-80; Paushok v. Mongolia, para 693; Hoffmann (n 5) 514.

⁹¹ Brenninkmeijer and Gélinas (n 75).

⁹² Yiduo Gong, Jiang Hong and Ying Wu, ‘Counterclaims’ (Lexology, 21 July 2023) <https://www.lexology.com/library/detail.aspx?g=954d6282-3b92-4559-bbbc-d080aef7409a> accessed 24 October 2024.

⁹³ Brenninkmeijer and Gélinas (n 75).

⁹⁴ ibid.

⁹⁵ Urbaser S.A. and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v. The Argentine Republic, ICSID Case No ARB/07/26, Award (8 December 2016) (Urbaser v. Argentina).

⁹⁶ Burlington v. Ecuador, ICSID Case No ARB/08/5, Decision on Counterclaims (7 February 2017) (Burlington v. Ecuador).

⁹⁷ Goetz v. Burundi II, para 285; Brenninkmeijer and Gélinas (n 75).

⁹⁸ Urbaser v. Argentina.

⁹⁹ ibid., para 1151.

¹⁰⁰ ibid.

¹⁰¹ ibid.

¹⁰² Burlington v. Ecuador, para 62.

¹⁰³ Brenninkmeijer and Gélinas (n 75).

¹⁰⁴ Lalive and Halonen (n 79) (emphasis added).

¹⁰⁵ Notes to the ICSID Arbitration Rules (1968), Note B (a) to r 40, reprinted in 1 ICSID Rep 63, 100.

¹⁰⁶ Zachary Douglas, ‘The Enforcement of Environmental Norms in Investment Treaty Arbitration’ in Pierre-Marie Dupuy and Jorge Viñuales (eds), Harnessing Foreign Investment to Promote Environmental Protection: Incentives and Safeguards (CUP 2013) 430 (emphasis added).

¹⁰⁷ Brenninkmeijer and Gélinas (n 75).

¹⁰⁸ Hoffmann (n 5) 518.

¹⁰⁹ ibid., 519.

¹¹⁰ ibid., 520.

¹¹¹ Kalicki (n 10).

(c)