Authors: Manu Sanan*, Jay Malwade**
| Jurisdiction: | Topics:
|
In August 2024, a violent uprising in Bangladesh (dubbed the ‘Monsoon Revolution’) led to a political regime change, ending the fifteen-year term of Bangladesh’s previous Prime Minister. Taking stock of Bangladesh’s economy on priority, the new interim government (supported by the military and with constitutional approval by the Bangladeshi Supreme Court) recently issued a 400-page White Paper on State of the Bangladesh Economy, highlighting the previous government’s alleged lapses (stating that over US$240 billion had been embezzled and laundered over fifteen years) and identifying projects that had operated/been awarded on questionable terms and regulatory exemptions. The interim government plans to review and renegotiate previous contracts and projects, and clearer next steps will likely emerge through Bangladesh’s forthcoming economic mid-term plan (2025 – 2027).
Meanwhile, Indian investments in Bangladesh face an uncertain future, not least given a prominent “India Out” protest in Bangladesh in 2024, before the uprising. India is the largest exporter of goods to Bangladesh (approximately US$11 billion, FY 2023 – 2024), and Indian companies have invested heavily in Bangladesh (around US$3 billion in FDI). These Indian companies are reportedly now facing payment delays and investment protection issues. As a prominent example, Indian power companies continue to be owed large sums as unpaid dues under power-purchase agreements, the terms of which will extend into the next decade, with at least one contract under investigation by Bangladesh. The regime change has also impacted commitments with Indian entities under MoUs for long-term rail connectivity, pipeline projects, and ongoing infrastructure projects in the energy sector.
The situation calls into focus the text of the India-Bangladesh BIT 2009 (the “2009 BIT”), the impact of the India-Bangladesh Joint Interpretative Notes 2017 (the “2017 JIN”), and the broader issue of the interpretive value of joint interpretative agreements in international law.
The 2009 BIT comprises relatively strong investment protections: a broad FET clause (covering ‘investments and returns of investors’), a broad National Treatment clause (covering ‘investments of investors’ and ‘return in respect of their investments’), an MFN clause (covering treatment to ‘investors’ but excluding matters relating wholly to taxation), an Expropriation clause (protecting against measures ‘equivalent to nationalization or expropriation’ except for a ‘public purpose’ and against ‘fair and equitable compensation’), and a Transfers clause (protecting free transfers of ‘all funds of an investor … related to an investment’).
Specifically, in the context of payments owed by Bangladeshi state entities, the 2009 BIT describes a covered ‘investment’ (Article 1(b)) as ‘every kind of asset invested by an investor’ including ‘rights to money or to any performance under contract having a financial value’ as well as ‘business concession conferred by law or under contract, including concessions to search for and extract oil and other minerals’. Based on a similarly worded standard, in Deutsche Bank AG v. Sri Lanka (ICSID, 2012), the tribunal found an ‘oil hedging agreement’ to be ‘a claim to money’ that had been ‘used to create an economic value’ and therefore concluded that it was an ‘investment’ under the Germany-Sri Lanka BIT (2000). In contrast, other tribunals assessing contractual debts have been more circumspect in their analysis. For instance, in Nova Scotia Power v. Venezuela (ICSID, 2014), a contract for the sale and purchase of coal did not qualify as an ‘investment’. Similarly, in Romak v. Uzbekistan (UNCITRAL, 2009), a wheat supply contract was found to be lacking the inherent characteristics of an ‘investment’ i.e., a contribution of capital over a certain duration and entailing certain risk.
Relevantly, the 2017 JIN (the first JIN by India following its 2015 Model BIT) reads down several provisions of the 2009 BIT, and qualifies the term ‘investment’ as follows:
[i]n accordance with Article 1(b), the minimum characteristics of an “investment” are (a) the lasting contribution of capital or other resources; (b) the expectation of gain or profit; (c) the assumption of risk by the investor; and (d) significance for development of the Contracting party receiving the investment.
For the avoidance of doubt, an investor of one Contracting Party must make its investments “in the territory of” the other Contracting Party. This means, for example, that claims to money arising solely from cross-border commercial contracts for the sale of goods or services, or the extension of trade financing in connection with a cross-border commercial transaction, or other relationships or instruments not involving an investor’s actual investment project in the territory of the other Contracting Party, do not constitute covered investment. Furthermore, the mere fact that an investment “benefits” the Contracting Party in which it is made is insufficient to establish that it is an investment “in territory” of that Contracting Party.
The language of the 2017 JIN pointedly excludes ‘claims to money arising solely from cross-border commercial contracts for the sale of goods or services’. It would, in the present context, exclude payments owed by Bangladeshi state entities to Indian investors or suppliers of goods and services. Further, the 2017 JIN: a) excludes from the 2009 BIT’s protection ‘any law or measure regarding taxation including measures taken to enforce taxation obligations’; b) pegs the FET standard to the customary international law minimum; c) clarifies that the MFN standard does not alter the 2009 BIT’s substantive content via textual importation; and d) clarifies that a determination of indirect expropriation would require ‘a total or near total and permanent destruction of the value of the investment’ and would exclude measures taken for ‘public policy objectives’ such as those for the ‘protection and improvement of economic conditions and the integrity of the financial system’ and the ‘implementation of fiscal policy measures, including taxation’.
In short, the 2017 JIN likely affords Bangladesh critical executive flexibility against potential claims by Indian investors. This contrasts with claims that Bangladesh might otherwise face, such as the recent claim initiated by S. Alam, a Bangladeshi-origin Singapore citizen, under the Singapore-Bangladesh BIT (2004), following investigations into his businesses and asset freezes by Bangladesh.
Then again, the application and interpretive value of joint interpretative notes remains a debated question in international law, especially given the International Law Commission’s controversial observation (in its commentary to its 2018 ‘Draft conclusions on subsequent agreements and subsequent practice in relation to the interpretation of treaties’) that ‘subsequent agreements and subsequent practice that establish the agreement of the parties regarding the interpretation of a treaty are not necessarily legally binding’ (see here). Moreover, as the decisions reviewed below show, several tribunals have previously disregarded joint interpretations on various counts.
Consider NAFTA’s early example, which contained a mechanism for interpretation by the parties through a Joint Commission, famously exercised via a Note of Interpretation dated 30 July 2001, which clarified the FET standard in NAFTA as being aligned to the minimum standard of treatment in international law. The tribunal in Pope & Talbot v. Canada (UNCITRAL 2002), after ruling against Canada on the merits, observed in its damages award that it had a ‘duty to consider’ whether what ‘the Commission has stated to be an interpretation is one for the purposes of Article 1131(2)’ and that ‘were the Tribunal required to make a determination whether the Commission’s action is an interpretation or an amendment, it would choose the latter’. The Pope & Talbot tribunal ultimately sidestepped the application of the joint interpretation.
In another example, the EU Member States set out their shared understanding on the invalidity of intra-EU investment treaties via the ‘Declaration of the EU Member States on the Legal Consequences of the Achmea Judgment’ (15 January 2019) stating that ‘all investor-State arbitration clauses contained in bilateral investment treaties concluded between Member States are contrary to Union law and thus inapplicable.’ That understanding was nevertheless rejected based on the application of treaty interpretation by, among others: a) the tribunal in PNB Banka v. Latvia (ICSID, 2021), which dismissed jurisdictional objections on the basis that ‘deleting a clause in entirety cannot be described as interpretation’, and b) the tribunal in Strabag v. Poland (ICSID, 2020), which dismissed jurisdictional objections and pointedly observed that:
[i]n particular, Article 31(3) VCLT requires “extrinsic” elements such as (i) “any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions” and (ii) “relevant rules of international law applicable in the relations between the parties” to be “taken into account, together with the context” (emphasis added). From the text of Article 31(3) VCLT, it is evident that such “extrinsic” elements, while informative to the context of a treaty, cannot be used to rewrite the ordinary meaning of the text of the treaty under interpretation … [a]ccordingly, the Tribunal is not persuaded by the Respondent’s argument that the EU Treaties, the Achmea Judgment or the Achmea Declaration can “trump over other methods of interpretation” in Article 31(1) VCLT or even “amend the text of the treaty” … [s]ince Article 31 VCLT states that the treaty text must be given its “ordinary meaning”, extrinsic factors that may be taken “together with” the context of a treaty cannot prevail over the ordinary meaning of the text of that treaty.
In another instance, the joint interpretation effected via an exchange of letters between China and Laos (stating that the China-Laos BIT 1993 did not apply to Macau) was discounted by the Singapore Court of Appeal in Sanum v. Laos (SGCA 57, 2016). The court reasoned that: a) the diplomatic letters represented modifications rather than interpretations given that the China-Laos BIT 1993 was unclear as to whether it applied to Macau; b) the letters/joint interpretation were issued after the initiation of the arbitral proceedings, making them ‘post-critical-date-Evidence’ that could not be applied to ‘pre-critical-Date position’; and c) the example of the NAFTA joint-commission’s common interpretation was distinguishable, as NAFTA contained a mechanism provided for in the text of the treaty. That reasoning was heavily criticized for not being rooted in the interpretative framework of the VLCT.
Notwithstanding critical commentary on the above precedents (see, here and here), the substantive impact of joint interpretations under international law (especially when they are ad hoc and purport to be ‘binding’, as the 2017 JIN does) appears to turn on whether they are viewed as interpretative aids or disguised amendments. However, that difficult determination is only further compounded by the barely perceptible line between ‘interpretation’ and ‘amendment’—at best, commentators have distinguished ‘interpretation’ as an exercise giving meaning to a norm, in contrast to an ‘amendment’ that alters the norm altogether (see here). That is especially relevant when considering the 2017 JIN, the FET-clarifying provision of which has been commented on as an ‘amendment disguised as interpretation’ due to its seemingly substantive norm-altering character (see here).
More broadly, the practice of joint interpretative statements has become increasingly used by state parties to retain control over the treaty texts. For instance, several treaties now include explicit provisions on binding joint interpretative statements, such as CETA, CPTPP, USMCA, the Additional Protocol to the Pacific Alliance, the ASEAN-India Investment Agreement, and an increasing number of Model BITs, including the Indian Model BIT (2015). In more overt practice, Colombia has steadily employed joint interpretative statements over the past few years via the Colombia-France Joint Interpretative Statement (2014), the Colombia-Israel Joint Interpretative Statement (2020), and the US-Colombia Joint Interpretative Statement (2025). India, in addition to the 2017 JIN, has signed the India-Colombia Joint Interpretative Statement (2018) and the India-Mauritius Joint Interpretative Statement (2022)—the latter likely forming a prominent discussion in certain ongoing disputes.
Seen in the larger context, the present situation reaffirms the increasing prominence of the question of interpretive control over treaty texts in international law and what has been described as the rise of ‘shared interpretive authority’ (see here and here). That phenomenon is theoretically complex and spans diverse analytical axes—one being investor concerns over procedural unfairness and post-hoc amendments disguised as joint interpretations, a question that might come to be asked of the 2017 JIN.
*Manu Sanan is the founder of Sanan Law, a specialist law office in New Delhi with international credentials, and acts for private and sovereign clients in a variety of disputes before the courts in India and before international tribunals constituted under the ICC, SIAC, HKIAC, DIS, ICSID, and UNCITRAL rules, and before WTO Panels. Manu also serves as an arbitrator and is a Fellow of the Chartered Institute of Arbitrators (FCIArb).
**Jay Malwade is an advocate at Sanan Law. Jay advises clients on contentious and non-contentious matters, with specific experience in Indian financial regulations. Jay is also experienced in commercial disputes and arbitrations.