0121 - The ‘Close Connection’ Test and State Counterclaims in Investment Arbitration – When is ‘Close’ Close Enough?

0121 - The ‘Close Connection’ Test and State Counterclaims in Investment Arbitration – When is ‘Close’ Close Enough?



Investment arbitrations typically entail an investor alleging breaches by the respondent State of protections conferred on the investor by the State in an investment treaty. However, it is rare for States to assert counterclaims, and even more rare for tribunals to determine those counterclaims on the merits. While there may be practical reasons why States prefer not to raise counterclaims before investor-State tribunals (including a preference to litigate those issues in their national courts), counterclaims are also limited by jurisdictional and admissibility hurdles that do not apply to the investor’s claims. One such admissibility impediment is the requirement that the counterclaim be “closely connected” to the investor’s claim.

Investor-State tribunals have historically taken a restrictive approach to assessing the “close connection” test. There are various formulations of the test, including that the counterclaim be “indivisible” and “interdependent” on, and have a factual and legal nexus with, the investor’s claim. As a result, while tribunals have accepted jurisdiction over counterclaims that arise from the same (or related) investment contract that gives rise to the investor’s claims, counterclaims alleging violations of the State’s domestic laws, and international treaties and customs have been less successful. However, an award rendered in December 2016 in Urbaser v Argentina adopted a less restrictive approach and suggested that a mere “factual link” between the counterclaim and the investor’s claim would be sufficient to satisfy the “close connection” requirement.

This article explores the rationale and evolution of the “close connection” test, why the restrictive approach to assessing “close connection” is incorrect, and the positive implications of the decision in Urbaser v Argentina.

The Requirement for a “Close Connection” Between the Investor’s Claim and the State’s Counterclaim

The constitutive instruments of various international tribunals expressly provide that a counterclaim must share a “connection” with the original claim. For instance, the Rules of Court of the International Court of Justice (“ICJ”) provide that a counterclaim must be “directly connected with the subject-matter of the claim” (Rule 80(1)), while the Algiers Accords that established the US-Iran Claims Tribunal states that a counterclaim must “arise[] of out of the same contract, transaction or occurrence that constitutes the subject matter of the national’s claims” (Article II(1)). 

In contrast, investment treaties and applicable arbitral rules rarely contain an express requirement for any connection between a counterclaim and the original claim (most investment treaties are in fact entirely silent on counterclaims). One notable exception is the ICSID Convention which provides that counterclaims must arise “directly out of the subject-matter of the dispute” (Article 46). However, notwithstanding the absence of any express “connection” based requirements, investment arbitration tribunals – whether under ICSID or otherwise – regularly evaluate the degree of closeness between the original claim and counterclaim to determine the admissibility of a counterclaim. According to the UNCITRAL tribunal in Saluka v Czech Republic, the “close connection” element is now “customary” under international law (at para. 61).

The “close connection” test serves an important function in narrowing the scope of disputes before the Tribunal to claims that are related, thereby preventing a respondent from raising unrelated claims as a tactic to obstruct or delay resolution of the claimant’s claims. As the ICJ explained in in its decision on counterclaims in the Bosnia Genocide Case, a respondent cannot “impose on the Applicant any claim it chooses, at the risk of infringing the Applicant’s rights and of compromising the proper administration of justice” (at para. 31).

Assessing the “Close Connection” –  From “Essential Unity” To A “Factual Link”

The most frequently cited “close connection” test in investment arbitration was laid down by the tribunal in Saluka v Czech Republic. In Saluka, the Czech Republic filed counterclaims in relation to the investor’s breaches of domestic law. After reviewing primarily Iran-US Claims Tribunal and two ICSID precedents, the Saluka tribunal found there was an insufficient connection between the original claims and the counterclaims. In reaching its conclusion, the tribunal considered whether there existed “an interdependence and essential unity of instruments on which the original claim and counterclaim were based” (para. 70).

The strict test formulated in Saluka, which focuses on the closeness of the legal basis and instruments underlying the original claims and counterclaims, has since been applied by other tribunals that have similarly rejected counterclaims by respondent States (see e.g. Paushok v Mongolia). However, the Saluka test has been criticized for being wrong on both precedent and principal (Douglas, The International Law of Investment Claims, Cambridge; Cambridge University Press (2009), at p. 260-262).

The Saluka tribunal relied heavily on jurisprudence of the Iran-US Claims Tribunal. This approach was flawed. The tests formulated in those cases were a direct result of the Algiers Accords (the constitutive instrument of the Iran-US Claims Tribunal), which expressly required that the counterclaims arise from the “same contract, transaction or occurrence that constitutes the subject matter of the national’s claims.” The Iran-US Claims Tribunal was therefore bound by a more limited notion of connectedness. While well-suited to disputes arising out of the same instrument or suite of instruments, the narrow test in the investment arbitration context creates an onerous burden on States seeking to allege breaches by the investor of domestic or international law. Such claims necessarily have a separate legal basis to the investor’s claims which arise out of the investment treaty.

The test in Saluka is also problematic from a policy perspective. By narrowly construing the ambit of permissible counterclaims, the Saluka test precludes otherwise factually related claims from being resolved in the same forum, thereby reducing the efficiency of the dispute resolution process and increasing the risk of inconsistent decisions.  Furthermore, compelling States to litigate their counterclaims in domestic courts undermines the very premise of denationalizing disputes in investment law. As rightly noted by Professor Michael Reisman in his dissenting opinion in Roussalis v Romania:

“In rejecting, … jurisdiction over counterclaims, a neutral tribunal – which was, in fact, selected by the claimant – perforce directs the respondent State to pursue its claims in its own courts where the very investor who had sought a forum outside the State apparatus is now constrained to become the defendant (And if an adverse judgment ensues, that erstwhile defendant might well transform to claimant again, bringing another BIT claim.) Aside for duplication and inefficiency… it is an ironic, if not absurd, outcome, at odds, in my view, with the objectives of international investment law.”

In December 2016, however, the tribunal in Urbaser v Argentina appears to have shifted away from the narrow test in Saluka, choosing to focus instead on the factual link between the original claim and the counterclaim.

The dispute in Urbaser related to a concession for water and sewage services to be provided by the investor’s subsidiary in Buenos Aires. Certain fiscal measures taken by the Argentinean government to mitigate the financial crisis in 2001-2002 caused the investor significant financial loss. The investor subsequently commenced proceedings against Argentina alleging violations of the Spain-Argentina BIT. Argentina in turn filed a counterclaim alleging that the investor’s failure to finance the work necessary for the concession “violat[ed] its commitments and its obligations under international law based on the human right to water” (para 34).

The parties’ respective submissions on the “close connection” test canvassed precedent, including Saluka. The Urbaser tribunal held that the “factual link” between the original claim and the counterclaim was manifest because they are “based on the same investment, or the alleged lack of sufficient investment, in relation to the same Concession” (para. 1151). Reflecting the concerns voiced by Professor Reisman, the tribunal also went on to state that it would be inconsistent for these related disputes to be heard in separate fora, with the attendant risk of inconsistent decisions (para. 1151). On the facts of the case, the tribunal found that Argentina failed to establish an obligation on the investor to secure the human right to water.

The Urbaser tribunal’s reliance on a “factual link” to determine connectedness is a welcome improvement over the strict Saluka formulations of “interdependence” and “essential unity of instruments” between the claims. An emphasis on factual closeness grants tribunals a greater degree of flexibility at the admissibility stage and will therefore allow more respondent States to raise counterclaims in relation to conduct of the investor that violates domestic and international law. Furthermore, it is efficient and reduces the risk of inconsistent decisions for disputes relating to the same investment to be heard by the same tribunal.

While its conclusions are laudable however, the methodology of the Urbaser tribunal is susceptible to critique. Although raised by the parties, the tribunal did not engage with existing precedent, including Saluka, or explain the reason for its divergence from past case law. In addition, the Urbaser tribunal did not articulate the scope, and in particular, the limits of the “factual link” required to satisfy the close connection test, thereby potentially opening the door to frivolous counterclaims.


The “factual link” test advocated by the Urbaser tribunal is a positive development that could result in more related issues being resolved in the same forum, reducing the inefficiencies of litigating claims in multiple fora and the risks of inconsistent decisions. However, it remains to be seen whether future investor-State tribunals embrace the test proposed by the Urbaser tribunal. If so, it will be imperative for tribunals to articulate what constitutes a sufficient factual nexus between the counterclaim and original claim. A failure to do so risks respondent States raising unrelated counterclaims to obstruct proceedings, undermining both the premise and protections offered by the “close connection” test.


Burgerlijk recht, Verzekeringsrecht
Dharshini Prasad

Dharshini Prasad

advocaat WilmerHale Londen

website by Primosite