Between Treaty Text and Arbitral Reasoning: Overlap and Divergence in the Core Investor-Protection Standards
This paper examines the core standards of investor protection in international investment law, namely fair and equitable treatment, full protection and security, most-favoured-nation and national treatment, expropriation, and umbrella clauses. Adopting a doctrinal approach, it argues that the content of these standards cannot be derived from treaty drafting alone. Open-textured clauses acquire practical shape through tribunal reasoning, which has prompted states to recalibrate treaty language. Proceeding standard by standard, the paper maps dominant treaty drafting approaches and traces how tribunals have interpreted them across a representative sample of awards selected for doctrinal salience, forum diversity, and the availability of reasoned decisions. It shows that fair and equitable treatment is the principal site of interpretive divergence, that full protection and security has expanded from physical protection towards legal security, and that the procedural reach of most-favoured-nation clauses remains contested, while expropriation and umbrella-clause doctrine test the boundaries between regulation and takings, and between treaty and contract. Overlap among standards emerges as a systemic feature of investor-state dispute settlement: a single measure may be pleaded under several standards, affecting thresholds of breach, standards of review and remedies, and risking double counting. Deference varies, inconsistency often reflects methodological variation more than disagreement over outcomes, and treaty practice is responding through calibrated drafting, closed lists, carve-outs, annexes and procedural filters. The paper concludes that the standards show partial convergence on rule-of-law principles such as legality, non-arbitrariness, due process and non-discrimination, alongside persistent fragmentation.
Introduction
This paper examines the core standards of investor protection through a jurisprudential lens, integrating the evolution of treaty text with the arbitral practice that has given those standards operational meaning.1,2 The organising premise is that the content of investor-protection standards cannot be derived from treaty drafting alone, because open-textured clauses acquire practical shape through tribunal reasoning, and, conversely, tribunal reasoning has prompted states to recalibrate treaty language in newer instruments.3 Accordingly, the paper proceeds standard by standard, first identifying the dominant drafting approaches in modern treaties and then analysing how arbitral tribunals have interpreted and applied those standards across a representative set of awards.4 It concludes by synthesising cross-standard overlaps and the blurring of doctrinal boundaries, and by identifying patterns of deference, inconsistency and normative direction.5
Method of case selection and the analytical approach
The jurisprudential sample is constructed to be representative rather than exhaustive. It draws on ICSID and UNCITRAL awards as the principal ISDS fora, and includes disputes decided under treaty frameworks associated with NAFTA or the USMCA and the Energy Charter Treaty (ECT), given their influence on the development of interpretive approaches.6 Cases are selected using three criteria. First, doctrinal salience: the award is repeatedly cited for a proposition central to fair and equitable treatment (FET), full protection and security (FPS), most-favoured-nation (MFN) or national treatment (NT), expropriation, or umbrella clauses.7 Second, forum and geographic diversity: the sample includes awards arising from different treaty families and regulatory contexts, including measures adopted for public-interest objectives.8 Third, the availability of reasoned decisions, because the paper is doctrinal and relies on articulated reasoning rather than empirical aggregation.9 The objective is not to “count” approaches but to trace how interpretive moves recur, how lines of authority diverge, and how drafting models have emerged partly in reaction to these trajectories.10
Fair and equitable treatment: drafting models and divergent arbitral approaches
Fair and equitable treatment is often described as the regime’s most influential standard because it functions as a general discipline on the exercise of public power and can integrate rule-of-law concerns such as non-arbitrariness, due process, transparency and legitimate expectations.11 Its breadth, however, explains why it has been the primary site of interpretive divergence and why states have increasingly sought to define it more tightly.12 Five drafting models capture the dominant treaty approaches.
First, unqualified FET clauses promise fair and equitable treatment without definition. These formulations have permitted tribunals to build an autonomous standard that sometimes extends beyond the customary minimum standard. Tecmed13 is frequently associated with an expectations-centred reading, in which the tribunal treated the investor’s reliance and the state’s conduct as central to evaluating fairness, and in which regulatory instability and administrative inconsistency featured prominently in the breach analysis. Saluka14 is often treated as a refinement, recognising that investors may expect a stable framework but emphasising that legitimate public-interest regulation is not inherently unfair and that the standard does not freeze the legal order.
Second, customary minimum standard formulations explicitly tie FET to customary international law.15 This approach, associated with NAFTA practice after the Free Trade Commission’s 2001 Notes of Interpretation and later reflected in U.S. model texts, is designed to prevent FET from becoming an open-ended “good governance” clause. Waste Management (No. 2)16 is widely cited for consolidating minimum-standard content and for framing breach in terms of egregious arbitrariness, manifest unfairness or denial of justice.
Third, illustrative list formulations specify that FET includes certain wrongs, such as denial of justice or fundamental breach of due process, without necessarily stating that the list is exhaustive. This model structures argument around recognised categories while preserving limited flexibility.
Fourth, closed-list formulations, exemplified by CETA, provide an exhaustive list of conduct that constitutes breach, such as denial of justice, fundamental breach of due process, manifest arbitrariness, targeted discrimination or abusive treatment. This drafting style reduces tribunal discretion by confining FET to enumerated wrongs and by narrowing legitimate expectations to cases grounded in specific, attributable representations.17
Fifth, “customary-law-based and enumerated formulations”, exemplified by instruments such as India’s 2016 Model BIT, avoid an express FET clause while defining protected treatment through customary international law and specified forms of prohibited conduct. The doctrinal effect is to raise the threshold for breach and to shift the focus towards denial of justice, fundamental breach of due process, targeted discrimination and manifestly abusive treatment rather than regulatory change as such.18
The divergence among FET approaches is visible in how tribunals handle legitimate expectations and regulatory autonomy. Henckels argues for cabining expectations within a rule-of-law frame, limiting protection primarily to expectations created by specific representations.19 Paparinskis20 similarly cautions against using rule-of-law language to justify expansion beyond what treaty text and state practice support, and emphasises due process and denial of justice as firmer anchors. These debates are reflected in the contrast between expectations-heavy awards and later treaty drafting that increasingly restricts or specifies expectations.
Full protection and security: from physical protection to legal security
Full protection and security historically refers to the host state’s duty to exercise due diligence to protect investors and their property from physical harm, particularly during civil unrest or violence.21 AAPL22 is often cited for treating FPS as a due diligence standard rather than one of strict liability, requiring reasonable measures to prevent harm. AMT23 similarly treats the obligation as one of reasonable protection, and has been invoked to support the view that FPS does not guarantee complete immunity from harm.
A more contested trend is the expansion of FPS into “legal security”, where tribunals treat the clause as extending beyond physical protection to encompass the stability and effectiveness of the legal system, including administrative and judicial protection, although the scope and doctrinal basis of that extension remain contested. Mantilla Blanco24 traces this shift and shows that it draws FPS closer to FET, thereby increasing the risk of overlap and potentially undermining a state’s attempt to cabin FET through closed lists or tethering to custom. The jurisprudence also shows variation in the intensity of review. Tribunals have examined whether the state failed to provide effective protection in the face of threats,25 while in disputes such as South American Silver,26 arguments about protection and security intersect with broader issues of social conflict, police conduct and regulatory response, illustrating how factual complexity can blur the boundary between FPS as protection against violence and FET as a broader fairness discipline.
MFN and national treatment: substantive reach and the procedural controversies of MFN
National treatment and MFN are equality disciplines, but their practical application involves recurring interpretive problems: the identification of comparators, the meaning of “like circumstances”, the role of regulatory purpose, and the evidentiary thresholds for showing differential treatment. Weiler’s27 historical analysis argues that investment equality cannot simply borrow WTO tests, because investment treaties confer individual rights and damages remedies rather than operating as inter-state compliance mechanisms. Whitsitt28 similarly emphasises that non-discrimination norms should be read as rule-of-law protections that safeguard effective equality while accommodating differentiated treatment grounded in legitimate policy objectives.
MFN clauses generate an additional controversy: whether they allow investors to import more favourable procedural or substantive provisions from third-party treaties. Maffezini29 is commonly cited as permitting procedural importation in certain circumstances, treating dispute settlement as part of “treatment” unless clearly excluded. This line of authority has been sharply contested. Plama30 and later awards resist importing dispute settlement rights absent a clear textual basis, emphasising that consent to arbitration must be expressed and cannot be presumed through MFN. Other cases, including RosInvest31 and Berschader,32 are frequently invoked in debates about the capacity of MFN to expand jurisdiction or to import particular dispute-settlement arrangements, revealing a persistent tension between broad investor access and the consent-centred structure of arbitration. In response, modern treaty drafting increasingly includes MFN carve-outs that expressly exclude dispute settlement or limit MFN to post-establishment substantive treatment. India’s 2016 Model BIT goes further still: it contains no MFN clause at all, and national treatment is its only relative standard of treatment.33
Expropriation: legality tests, indirect takings and compensation
Expropriation doctrine has shifted from disputes about direct takings towards disputes about indirect or creeping expropriation through regulatory measures. Reinisch’s account of the legality of expropriations shows that, under general international law and treaty restatements, lawful expropriation requires public purpose, non-discrimination, due process and compensation.34 Treaty language often codifies these requirements explicitly, but tribunals still diverge in how they test indirect expropriation and how they distinguish regulation from taking.
Santa Elena35 is frequently cited for the proposition that even a public-purpose expropriation requires compensation, because public purpose relates to legality, not to the duty to compensate. ADC36 is often associated with a stricter legality analysis, treating the absence of due process and of a genuine public purpose as rendering the expropriation unlawful, with corresponding implications for damages. Burlington37 illustrates how expropriation claims can intersect with regulatory measures in sensitive sectors, raising questions about the intensity of review and the valuation of loss. Tethyan Copper38 is significant because it demonstrates how treaty standards, domestic regulatory decisions and investor reliance can interact in expropriation and FET reasoning, and how damages analysis can become central to the practical consequences of doctrinal classification.
Bungenberg’s analysis shows how modern treaties and tribunals increasingly use hybrid tests, combining the severity of economic impact with factors such as expectations, duration and the character of the measure, and how annex drafting attempts to standardise the analysis by clarifying that bona fide, non-discriminatory public-welfare regulation is generally non-compensable.39 Methanex40 remains a benchmark in the police-powers line, especially where non-discriminatory environmental regulation adopted in the public interest was held not to constitute a compensable taking. The broader doctrinal point is that expropriation analysis is one of the regime’s principal sites for managing regulatory autonomy, and treaty drafting innovations are in part an attempt to stabilise this balance.
Umbrella clauses: contractual elevation, limiting doctrines and the investment-nexus test
Umbrella clauses are controversial because they potentially elevate breaches of contract or other undertakings into treaty breaches, thereby allowing investors to plead contract-type grievances as treaty violations.41 The jurisprudence remains fragmented. The SGS trilogy42 is often used to map the spectrum of approaches: some tribunals adopt a broader “elevation” view, while others insist on narrower readings or on coordination with contractual forums and forum-selection clauses. El Paso43 is frequently cited for resisting the automatic elevation of contract breaches into treaty breaches and for emphasising the separation between treaty obligations and contractual obligations, absent clear language. Noble Ventures44 illustrates the possibility of umbrella clauses operating as a conduit for enforcing specific undertakings where the text supports such a reading and where the obligation is sufficiently connected to the investment.
Doctrinal writing has sought to provide limiting principles. De Nanteuil45 argues that umbrella clauses can enhance legal certainty and rule-of-law values by ensuring respect for state undertakings, while acknowledging the need for careful coordination to avoid making treaties general tools of contract enforcement. Naniwadekar frames the problem as one of deference and calls for a theory that respects the contractual allocation of dispute resolution while still policing abusive repudiation.46 A limiting doctrine that has emerged in practice is the investment-nexus test, which asks whether the breached obligation is sufficiently tied to the investment as such rather than being a purely commercial obligation.47 This test does not eliminate umbrella claims, but it can narrow their scope and mitigate overlap with FET and expropriation.
Cross-standard overlaps and synthesis of trends: deference, inconsistency and normative direction
This section draws the doctrinal threads together by focusing on overlap and trend. Petsche’s systemic account argues that overlap is embedded in ISDS because multiple standards regulate related concerns: legality, fairness, equality, protection of property, and compliance with undertakings.48 In practice, a single regulatory measure can be pleaded simultaneously as a breach of FET (arbitrariness or expectations), FPS (legal security), NT or MFN (discrimination), the prohibition of indirect expropriation (substantial deprivation) and, where a contract exists, an umbrella clause. Overlap increases the risk that tribunals will select different doctrinal lenses to address the same underlying grievance, which in turn affects thresholds of breach, standards of review and remedial frameworks. It can also produce double counting unless tribunals explicitly manage the interaction among standards.
Across the awards discussed, three broad patterns emerge. First, deference varies significantly. Some tribunals defer strongly to state determinations of public purpose, regulatory design or security assessments, while others apply more searching review, especially where due process failures, targeted discrimination or bad faith are alleged. Second, inconsistency often reflects methodological variation rather than pure disagreement about outcomes; tribunals differ in how they read consent, how they use systemic integration, and how they treat prior awards as persuasive authority. Third, the normative direction of treaty practice is towards calibrated drafting and institutional reform. Closed-list FET, MFN carve-outs, expropriation annexes and procedural filters reflect state attempts to reduce interpretive discretion and preserve regulatory space while maintaining a credible commitment to protect investments.
Conclusion
The jurisprudence surveyed in this paper shows that the core standards of investment protection are best understood as a dynamic interaction between treaty text and arbitral reasoning, rather than as fixed clauses with uniform meaning. Across FET, FPS, MFN and NT, expropriation, and umbrella clauses, tribunals have generated divergent lines of authority on thresholds of breach, standards of review, and the relationship between investor protection and legitimate regulation.
Modern treaty drafting has, in response, moved towards tighter definitions, annexes and procedural filters. The resulting picture is one of partial convergence on foundational rule-of-law ideas (legality, non-arbitrariness, due process and non-discrimination) alongside persistent fragmentation driven by overlap among standards, inconsistent interpretive methodologies, and contested conceptions of regulatory autonomy.
The significance of this interaction is not merely that tribunals interpret differently worded provisions differently. Rather, attempts to constrain interpretation through treaty specificity may themselves redistribute interpretive pressure across neighbouring standards. The effectiveness of drafting reform therefore depends not only on the precision of individual clauses but on the architecture of the treaty as a whole and the manner in which tribunals manage the interaction among its protections.
*****
Footnotes
1. This paper is adapted from Chapter 4 of the author’s dissertation on investor-protection standards under international investment law.
2. See generally Rudolf Dolzer & Christoph Schreuer, Principles of International Investment Law (Oxford Univ. Press 2d ed. 2012).
3. See, e.g., U.S. Dep’t of State, 2012 U.S. Model Bilateral Investment Treaty; Comprehensive Economic and Trade Agreement, Can.-EU, ch. 8, Oct. 30, 2016, 2017 O.J. (L 11) 23 [hereinafter CETA]; Gov’t of India, Model Text for the Indian Bilateral Investment Treaty (2016) [hereinafter Indian Model BIT].
4. See generally Andrew Newcombe & Lluís Paradell, Law and Practice of Investment Treaties: Standards of Treatment (Kluwer Law Int’l 2009).
5. See Markus Petsche, Conceptual Overlaps Between Investment Protection Standards: Analysis of a Yet Unexplored Systemic Problem of ISDS, 41 J.L. & Com. 41 (2022), https://doi.org/10.5195/jlc.2022.240.
6. Convention on the Settlement of Investment Disputes Between States and Nationals of Other States, Mar. 18, 1965, 17 U.S.T. 1270, 575 U.N.T.S. 159; UNCITRAL Arbitration Rules (1976), as revised 2010, 2013; Energy Charter Treaty, Dec. 17, 1994, 2080 U.N.T.S. 95; North American Free Trade Agreement, Can.-Mex.-U.S., Dec. 17, 1992, 32 I.L.M. 289 (1993) [hereinafter NAFTA].
7. See generally Ioana Tudor, The Fair and Equitable Treatment Standard in the International Law of Foreign Investment (Oxford Univ. Press 2008).
8. See generally Florencia Sarmiento & Suzy Nikièma, Int’l Inst. for Sustainable Dev., Fair and Equitable Treatment: Why It Matters and What Can Be Done (IISD Best Practices Series, Nov. 2022), https://www.iisd.org/system/files/2022-11/fair-equitable-treatment-en.pdf.
9. See generally Todd Weiler, The Interpretation of International Investment Law: Equality, Discrimination and Minimum Standards of Treatment in Historical Context ch. 2 (Martinus Nijhoff 2013).
10. See generally Kenneth J. Vandevelde, Bilateral Investment Treaties: History, Policy, and Interpretation (Oxford Univ. Press 2010).
11. Tudor, supra note 7.
12. Martins Paparinskis, The Rule of Law and Fair and Equitable Treatment, in Investment Protection Standards and the Rule of Law ch. 2 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
13. Técnicas Medioambientales Tecmed, S.A. v. United Mexican States, ICSID Case No. ARB(AF)/00/2, Award, ¶ 154 (May 29, 2003).
14. Saluka Invs. B.V. v. Czech Republic, UNCITRAL, Partial Award, ¶¶ 305–306 (Mar. 17, 2006).
15. NAFTA, supra note 6, art. 1105(1); NAFTA Free Trade Comm’n, Notes of Interpretation of Certain Chapter 11 Provisions § B(1)–(2) (July 31, 2001); 2012 U.S. Model Bilateral Investment Treaty, supra note 3, art. 5(1)–(2), annex A.
16. Waste Mgmt., Inc. v. United Mexican States (No. 2), ICSID Case No. ARB(AF)/00/3, Award, ¶ 98 (Apr. 30, 2004).
17. CETA, supra note 3, art. 8.10(2), (4); see Caroline Henckels, Legitimate Expectations and the Rule of Law in International Investment Law, in Investment Protection Standards and the Rule of Law ch. 3 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
18. Indian Model BIT, supra note 3, art. 3.1; see Prabhash Ranjan & Pushkar Anand, The 2016 Model Indian Bilateral Investment Treaty: A Critical Deconstruction, 38 Nw. J. Int’l L. & Bus. 1 (2017), https://scholarlycommons.law.northwestern.edu/njilb/vol38/iss1/1.
19. Henckels, supra note 17.
20. Paparinskis, supra note 12.
21. Sebastián Mantilla Blanco, Full Protection and Security and the Rule of Law, in Investment Protection Standards and the Rule of Law ch. 5 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
22. Asian Agric. Prods. Ltd. v. Republic of Sri Lanka, ICSID Case No. ARB/87/3, Final Award (June 27, 1990).
23. Am. Mfg. & Trading, Inc. v. Republic of Zaire, ICSID Case No. ARB/93/1, Award (Feb. 21, 1997).
24. Mantilla Blanco, supra note 21.
25. See Asian Agric. Prods., supra note 22; Am. Mfg. & Trading, supra note 23.
26. South Am. Silver Ltd. v. Plurinational State of Bolivia, PCA Case No. 2013-15, Award (Nov. 22, 2018).
27. Weiler, supra note 9, ch. 9, at 415–54; Elizabeth Whitsitt, International Investment Law’s Non-Discrimination Norms and the Rule of Law, in Investment Protection Standards and the Rule of Law ch. 6 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
28. Whitsitt, supra note 27.
29. Emilio Agustín Maffezini v. Kingdom of Spain, ICSID Case No. ARB/97/7, Decision on Objections to Jurisdiction, ¶ 56 (Jan. 25, 2000).
30. Plama Consortium Ltd. v. Republic of Bulgaria, ICSID Case No. ARB/03/24, Decision on Jurisdiction, ¶ 223 (Feb. 8, 2005).
31. RosInvestCo UK Ltd. v. Russian Federation, SCC Case No. V079/2005, Award on Jurisdiction (Oct. 1, 2007).
32. Vladimir Berschader & Moïse Berschader v. Russian Federation, SCC Case No. 080/2004, Award (Apr. 21, 2006).
33. Indian Model BIT, supra note 3, art. 4; see Ranjan & Anand, supra note 18.
34. August Reinisch, Legality of Expropriations, in Standards of Investment Protection 171 (August Reinisch ed., Oxford Univ. Press 2008).
35. Compañía del Desarrollo de Santa Elena, S.A. v. Republic of Costa Rica, ICSID Case No. ARB/96/1, Final Award, ¶¶ 71–72 (Feb. 17, 2000).
36. ADC Affiliate Ltd. & ADC & ADMC Mgmt. Ltd. v. Republic of Hungary, ICSID Case No. ARB/03/16, Award (Oct. 2, 2006).
37. Burlington Res., Inc. v. Republic of Ecuador, ICSID Case No. ARB/08/5, Decision on Liability (Dec. 14, 2012).
38. Tethyan Copper Co. Pty Ltd. v. Islamic Republic of Pakistan, ICSID Case No. ARB/12/1, Award (July 12, 2019).
39. Marc Bungenberg, (Direct and Indirect) Expropriation and the Rule of Law, in Investment Protection Standards and the Rule of Law ch. 4 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
40. Methanex Corp. v. United States, UNCITRAL, Final Award of the Tribunal on Jurisdiction and Merits, pt. IV, ch. D, ¶ 7 (Aug. 3, 2005).
41. Mihir C. Naniwadekar, The Scope and Effect of Umbrella Clauses: The Need for a Theory of Deference?, 2 Trade L. & Dev. 169 (2010).
42. SGS Société Générale de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID Case No. ARB/01/13, Decision on Objections to Jurisdiction (Aug. 6, 2003); SGS Société Générale de Surveillance S.A. v. Republic of the Philippines, ICSID Case No. ARB/02/6, Decision on Objections to Jurisdiction (Jan. 29, 2004); SGS Société Générale de Surveillance S.A. v. Republic of Paraguay, ICSID Case No. ARB/07/29, Award (Feb. 10, 2012).
43. El Paso Energy Int’l Co. v. Argentine Republic, ICSID Case No. ARB/03/15, Award (Oct. 31, 2011); see also El Paso Energy Int’l Co. v. Argentine Republic, ICSID Case No. ARB/03/15, Decision on Jurisdiction (Apr. 27, 2006).
44. Noble Ventures, Inc. v. Romania, ICSID Case No. ARB/01/11, Award (Oct. 12, 2005).
45. Arnaud de Nanteuil, Umbrella Clauses and the Rule of Law, in Investment Protection Standards and the Rule of Law ch. 9 (August Reinisch & Stephan W. Schill eds., Oxford Univ. Press 2023).
46. Naniwadekar, supra note 41.
47. See id.
48. Petsche, supra note 5.