THESIS ON JURISDICTIONAL AND PROCEDURAL CHALLENGES IN INVESTOR-STATE DISPUTE SETTLEMENT

Abstract. This thesis advances a single claim: the legitimacy crisis of investor-State dispute settlement is rooted not only in substantive standards such as fair and equitable treatment, but in the system’s compressed conception of consent. Contemporary ISDS treats a complex jurisdictional sequence State offer, investor acceptance, tribunal construction, and limited review as though it were equivalent to a single bilateral agreement. That compression obscures how sovereignty is front-loaded into treaty design, how procedural discretion reallocates the meaning of consent during arbitration, and why reform debates now focus on jurisdiction, appointment, review, and balance rather than only on merits standards. Methodologically, the thesis combines doctrinal analysis of primary legal materials most importantly the ICSID Convention, major awards, and UNCITRAL reform documents with a comparative reading of the leading scholarly schools: formalist, functionalist, critical, and reformist. Its central conclusion is that ISDS should be reconceived as a regime of layered public authorization. On that view, the core task of reform is not to abolish consent as a foundation, but to discipline it through clearer drafting, anti-abuse rules, reciprocal procedure, and stronger legitimacy safeguards. [1]

INTRODUCTION

The thesis begins from a simple observation: every major jurisdictional controversy in ISDS is, at bottom, a controversy about consent. Article 25 of the ICSID Convention requires written consent and forbids unilateral withdrawal once consent is perfected. Yet that seemingly simple rule has generated disputes over treaty versus contract claims, nationality and corporate control, MFN-based access to arbitration, local-remedies clauses, collective proceedings, counterclaims, provisional measures, and post-award review. The problem is not that the legal system ignores consent. The problem is that the system invokes consent at a level of abstraction too high to explain how it actually works in practice. [1]

The thesis uses a doctrinal and critical-comparative method. Doctrinally, it analyzes the texts that most directly constitute consent investment treaties, Article 25 of the ICSID Convention, Article 46 on ancillary claims, Article 47 on provisional measures, and Article 52 on annulment as well as leading awards. Critically, it compares those texts and awards against the major scholarly narratives of the field: the view that arbitration without privity is a justified extension of sovereign will; the view that treaty arbitration is a developing branch of public international law; the view that the regime suffers from foreign-investor privilege and jurisdiction beyond authentic consent; and the view that legitimacy can be repaired through institutional reform rather than abandonment. [2]

The thesis is organized into three chapters. The first reconstructs consent as a jurisdictional design that reallocates sovereignty. The second shows how procedural controversies repeatedly destabilize that design. The third examines legitimacy and reform, arguing that consent should be treated as a sequenced public-law structure rather than as a once-and-for-all private agreement.

CONSENT AS JURISDICTIONAL DESIGN

The first chapter argues that sovereignty in ISDS is not eliminated; it is front-loaded. Official UNCTAD materials stress that mere participation in the ICSID Convention creates no obligation to arbitrate and that States remain free to decide whether, and to what extent, they will consent. They may do so through contracts, legislation, or treaties. That point is doctrinally crucial. It means the system is not anti-consensual at its inception. States design the offer. The difficulty is that, once they do so, they lose meaningful control over who accepts it, when it is accepted, and how its procedural scope is later interpreted. Sovereignty is therefore exercised primarily at the drafting stage, not at the dispute stage. [16]

The difference between contract-based and treaty-based consent illustrates this front-loading. Contract arbitration resembles ordinary privity: the parties negotiate the arbitral clause in relation to a concrete investment relationship. Treaty arbitration does not. Vivendi’s annulment committee famously insisted that treaty claims are analytically distinct from contract claims, and Salini reinforced the idea that treaty jurisdiction does not simply absorb every contractual grievance. Yet the same jurisprudence also shows that once treaty consent exists, it can survive contractual forum choices and reach facts that are deeply embedded in contractual performance. The first chapter’s conclusion is that treaty consent is best understood as a sovereign regulatory decision to internationalize certain categories of disputes, not as a simple replication of commercial arbitration by other means. [14]

Standing doctrine then exposes how the identity of the consenting investor has become a second-order sovereignty issue. Tokios Tokelės [11] and Aguas del Tunari [4] privileged formal nationality and treaty drafting over economic nationality, effectively allowing protected status even where the investor’s ownership structure diluted or complicated the intuition of true foreignness. TSA Spectrum [12] moved back toward economic reality and denied jurisdiction by piercing the corporate veil. CMS [6] and Sempra [10] further broadened access by confirming standing for minority or indirect shareholding interests. These cases reveal that consent in ISDS is not just about a State’s decision to arbitrate with foreign investors; it is also about who counts as foreign enough to invoke that decision. The boundary between sovereignty and standing is, in practice, porous.

Attribution sharpens the same theme from the respondent side. In Maffezini, the tribunal was willing to treat the state-linked entity SODIGA as a prima facie State actor for jurisdictional purposes, while leaving questions of imputability for the merits. The lesson is that the State may not be able to compartmentalize sovereign consent by devolving investment relations to formally separate bodies. If a treaty authorizes claims against the State, then questions of attribution operate as a bridge between jurisdictional consent and substantive responsibility. The State’s autonomy in designing the offer thus coexists with a broad international-law apparatus that may attribute acts of public bodies, agencies, or state-linked entities back to it. [15]

The first chapter therefore concludes that consent in ISDS is already multilayered before procedure begins. It contains a sovereign drafting choice, a standing filter, and an attribution framework. This is why the thesis rejects the familiar claim that ISDS is either purely consensual or obviously coercive. It is neither. It is a hybrid system in which the State voluntarily internationalizes disputes, but then cedes significant control over the identification of claimants and the characterization of public acts. [17]

PROCEDURAL BREAKDOWN OF CONSENT

The second chapter argues that procedure is where the system’s compressed conception of consent breaks down most visibly. MFN disputes are the clearest example. Maffezini and Siemens treated MFN language as capable of extending more favorable access to arbitration, including by bypassing waiting periods or local-courts requirements. Plama, Wintershall and Daimler rejected that approach and insisted on clear and unambiguous consent to arbitrate. The chapter reads this split not as a technical disagreement over one doctrine, but as a disagreement about the nature of sovereignty itself. On the expansive view, sovereignty lies in the State’s treaty network as a whole. On the restrictive view, sovereignty lies in the precise text of the basic treaty alone.

The same instability appears in local-remedies and pre-arbitration conditions. BG Group, in the context of annulment and enforcement review before the U.S. Supreme Court, treated the local-litigation clause as a matter for arbitrators to interpret in the first instance rather than as a pure gateway question for domestic courts. Metalpar, RosInvestCo and Sempra belong to the same larger conversation about whether preconditions to arbitration are true jurisdictional limits, merely procedural filters, or default rules capable of being displaced by treaty interpretation. Once again, consent does not disappear, but its legal effect becomes contingent on the institutional vantage point from which it is analyzed. [18]

Treaty shopping and post hoc restructuring bring the temporal dimension of consent into focus. Phoenix Action, [9] Pac Rim [7] and Philip Morris [8] confirm that even formally valid treaty nationality may fail when it is created to internationalize a dispute that is already existing or highly foreseeable. Pac Rim’s formulation that abuse becomes relevant when a party can foresee a specific future dispute as a very high probability captures the right intuition: the legal problem is not foreign structuring as such, but manipulative sequencing. The second chapter therefore develops a “critical date” approach. Consent should be denied or discounted where the protected status of the claimant is created only after the relevant dispute has matured into a concrete legal controversy.

Mass procedure raises a different version of the same concern. Abaclat’s majority held that Argentina’s treaty consent could extend to mass claims by large numbers of bondholders, while the dissent argued that this effectively altered the procedural architecture of the ICSID framework without party agreement. The significance of the case is not confined to sovereign debt. It shows that even where claimant identity is individually protected, the collective aggregation of claims may transform the kind of proceeding to which the State can fairly be said to have consented. In this part of the chapter, consent is analyzed not merely as access to a tribunal, but as consent to a certain procedural form. [3]

Counterclaims reveal the problem of reciprocity. ICSID’s own materials underline that ancillary claims and counterclaims must lie within the scope of the parties’ consent. But the cases are divided. Roussalis treated the investor’s acceptance as too asymmetrical to support the State’s counterclaim. Saluka took a relatively open view of counterclaim jurisdiction while emphasizing the need for a sufficient connection. Urbaser accepted jurisdiction over Argentina’s counterclaim because the arbitration clause was broad and the dispute legally connected, [13] and Burlington demonstrated that counterclaims can produce concrete liability for investors when the procedural basis is secure. [5] The chapter’s argument is that one-way consent is a central source of the system’s legitimacy deficit, because it lets the investor activate international jurisdiction while often denying the State an equally robust route to investor accountability.

Provisional measures and review mechanisms expose the final procedural layer. ICSID’s official guidance makes clear that tribunals may recommend provisional measures to preserve rights, and real disputes have extended that power into contested terrain such as security for costs. Yet the stronger doctrinal point lies on the back end: annulment under Article 52 is expressly limited and is not an appeal. ICSID’s own background paper repeats that proposition, while current reform discussions in UNCITRAL contemplate standing mechanisms or appellate structures that may recalibrate, and in some proposals partly displace, the existing review model. The chapter concludes that procedure in ISDS is no longer a neutral envelope. It is the principal site at which consent is stretched, disciplined, or institutionally repaired. [1]

LEGITIMACY AND REFORM

The third chapter turns from doctrine to theory and institutional design. Its starting point is the observation, now reflected in UNCITRAL Working Group III, that the legitimacy debate is not reducible to ideology. States have identified recurring concerns about inconsistency, coherence and correctness of decisions, independence and impartiality of decision-makers, and cost and duration. Those concerns are not external to consent; they are symptoms of what happens when a regime built on thin ex ante authorization is asked to perform functions closer to public-law adjudication. In other words, the system’s legitimacy crisis is a consent problem in institutional form.

The third chapter then maps the main scholarly schools. Formalists, associated with scholars such as Christoph Schreuer [22] and Zachary Douglas [17], emphasize explicit written consent, the decisive role of text, and the need to avoid jurisdictional inflation. Functionalists and system-builders, associated with figures such as Stephan Schill [21], Martins Paparinskis [19] and Anthea Roberts [20], defend treaty arbitration as a coherent extension of public international law, even if they disagree on detail. Critical and TWAIL-oriented scholars such as M. Sornarajah [23] and Gus Van Harten [24] argue that the regime suffers from foreign privilege, jurisdiction beyond real consent, and structural distortions borrowed from private arbitration. Reformists, finally, focus not on conceptual rejection but on institutional redesign: appointments, codes of conduct, advisory support, appellate review, and investor accountability.

The thesis sides with none of these camps in full. It accepts the formalist point that consent cannot be created by mere policy preference. It accepts the functionalist point that modern investment arbitration cannot be understood solely through analogies to private contracts. It accepts the critical point that the system distributes power asymmetrically and often privileges foreign investors in ways domestic actors cannot replicate. And it accepts the reformist point that legitimacy can be improved through institutional design. Its own contribution is to synthesize these insights through a sequenced-consent model. On this model, what matters is not the presence of a single jurisdictional moment, but the integrity of the entire route from ex ante State offer to ex post award review.

That model yields specific reform implications. Treaties should state expressly whether MFN clauses reach dispute settlement, whether umbrella clauses convert contractual obligations into treaty obligations, and whether local-remedies requirements are jurisdictional or procedural. Ownership transparency and denial-of-benefits clauses should be aligned with anti-abuse standards to reduce opportunistic restructuring. Counterclaims should be addressed expressly, rather than left to strained inference from investor-centered consent language. Review structures should remain limited, but they should be calibrated more closely to jurisdictional distortion, particularly where tribunals transform the procedural form or scope of proceedings. None of these proposals requires abandoning consent. All of them require taking consent more seriously.

CONCLUSION

The thesis concludes that the central challenge of ISDS is not that consent is absent, but that consent is over-compressed. The legal system presents a multilayered structure as if it were a single bilateral agreement, and the resulting mismatch fuels disputes over sovereignty, procedure and legitimacy. Reframing ISDS as layered public authorization does not solve every problem. It does, however, clarify why jurisprudence on MFN clauses, treaty shopping, counterclaims, collective procedures and review has become so unstable: each controversy is an effort to reconstruct the limits of a consent that the system has rendered too abstract. The most promising path forward is therefore a jurisprudence and treaty practice that make those limits explicit.