EJIL: Talk! | 22 October 2025
Time and compromise in UNCITRAL’s Working Group III
by Ladan Mehranvar and Tarald Gulseth Berge
During the week of 22 September 2025, States once again met in Vienna under Working Group III (WGIII) of the United Nations Commission on International Trade Law (UNCITRAL) to deliberate reforms to investor–State dispute settlement (ISDS). Since receiving its mandate in 2017, WGIII has been tasked with identifying and addressing concerns about the ISDS system embedded in investment treaties.
In the years since, WGIII has acknowledged persistent legitimacy concerns: the inconsistency and unpredictability of arbitral decisions; arbitrator independence and impartiality; cost and duration of proceedings; and systemic risks like third-party funding. WGIII has also identified a set of cross-cutting concerns—e.g., exhaustion of local remedies, counterclaims, regulatory chill, and valuation of damages. Though it has adopted some incremental reforms—a Code of Conduct for Adjudicators and a Statute for an Advisory Centre on International Investment Dispute Resolution—progress on the more fundamental issues has been slow (for background, see posts by Anthea Roberts and Taylor St John). As WGIII enters its final stretch, time is pressing and the stakes are high.
On the agenda in Vienna were procedural matters, including security for costs (SfC), and the design of a standing dispute settlement mechanism. The SfC debate is an important issue in the reform effort: it seeks to address long-standing concerns—especially among developing countries—that respondent governments often cannot recover costs from investors after obtaining cost orders in their favour. The debate over SfC thus sits at the intersection of fairness, access to justice, and systemic imbalance—offering a microcosm of the broader reform struggle within WGIII. The standing mechanism, meanwhile, reflects the European Union’s proposal to replace ad hoc arbitration with a permanent, court-like structure, as seen in the EU–Canada Comprehensive Economic and Trade Agreement.
Pending a decision by UNCITRAL’s governing body, there may be only a few WGIII meetings left to finalize outstanding issues. This post examines how the scarcity of time is shaping WGIII’s negotiation dynamics on the issue of SfC. We look at how delegations leverage time to define what counts as a reasonable compromise; how the very term ‘compromise’ can entrench rather than reduce asymmetries; and what these framing battles reveal about State influence and leverage.
Compromising When Time is Scarce
In negotiations, time is a bargaining asset. The party less constrained by time tends to hold greater leverage. In bilateral settings, more time-pressed parties will seek to steer discussions towards closure, often proposing package deals to settle outstanding issues. Less time-pressed parties can push for packages with more favourable terms. In multilateral forums like WGIII, individual States have less control over pace; instead, international organizations and negotiating chairs shape timelines based on institutional constraints and budgetary considerations. As deadlines approach, ‘time’ becomes a tool of pressure—used to induce ‘compromise’ or force closure.
In Vienna, the Chair’s repeated references to time underscored this dynamic: “We are going to need to rely on the ability of this group to find compromises … We are on the final stage of our reform process here, moving towards negotiating some of the last substantive reforms on our agenda … We have discussed this for years. What I am looking now to see is where the compromise might be to allow our work forward.” By mid-week, he warned that at the current rate, it would take 20 more days to complete just the procedural rules, a pace he implied was unaffordable. He urged delegates to accelerate discussions and “reach compromises quickly, without repeatedly revisiting previous unresolved positions.” On the final day, he declared: “I look at the calendar and time is up. It is time to take decisions.”
When time becomes an explicit constraint, it reshapes negotiation dynamics. Ideally, multilateral negotiation is deliberative: parties present arguments grounded in normative principles and empirical reasoning. In that space, power lies in persuasion rather than coercion. But under compressed timelines, the deliberative space contracts and reasoned debate is curtailed as parties feel pressure to move quickly. Urgency moves the process away from developing regulative solutions to collective action or coordination problems (‘positive sum’) and toward closure-driven compromise and transactional bargaining (‘zero-sum’). In this mode, the “islands” of deep deliberation shrink, leaving a terrain increasingly shaped by framing battles and procedural optics.
In WGIII, this dynamic is increasingly visible. The Chair’s emphasis on time constraints and the need for compromise has, at times, led to a pattern in which delegations supporting certain draft provisions begin their intervention by thanking the Secretariat for its ‘compromise’ text. This rhetorical opening is one way leverage is exercised—by establishing favourable framing at the outset, which constrains what counts as reasonable before substantive debate even begins. Whether such texts reflect a fair or just compromise—correcting the imbalances WGIII was meant to address—is unclear. In theory, a just compromise should move beyond an unjust status quo to redress past inequities. In practice, the language of ‘compromise’ has become a tool of rhetorical closure, inhibiting argumentative reasoning.
The Discussion Around Security for Costs in WGIII
The SfC debate offers a vivid example of how reform goals can be reshaped under procedural pressure. Early in the reform process, many delegations emphasized the difficulty States face in recovering costs from investors in ISDS cases, even when they prevail. In 2017, the inability to recover costs was described as “exemplif[ying] an imbalance between the parties” (para 68, 34th Session Report). The problem was later reiterated as emblematic of systemic imbalance. While States must defend every claim, even frivolous ones, investors can become insolvent or simply refuse to pay (para 130, 36th Session Report).
UNCITRAL’s first Working Paper (WP) on SfC acknowledged this imbalance: “While there have been many instances where States have requested security for costs, there have been few decisions in which tribunals have granted security” (para 12, WP192, January 2020). A more predictable SfC framework could protect States and deter frivolous claims, the Secretariat argued, though safeguards should not unduly restrict investor access to arbitration (para 16, WP192). Thus, the language of ‘balance’ entered the discussion. While intended to reconcile asymmetrical risks, it opened the door for later reinterpretations of ‘balance’ as ‘symmetry’, a different framing from the one initially used by States and even the Secretariat.
The question of whether claimants should also be able to request SfC was introduced parenthetically in the same document (para 6, WP192), despite the consensus that the mechanism’s purpose was to protect respondents (para 70, 39th Session Report, October 2020). Over time, this parenthetical note became central to the reinterpretation of ‘balance’ as ‘symmetry’. A small but notable exception arose when discussing SfC vis-à-vis an appellate mechanism (para 59, WP202, November 2020). The session report records that “security for costs should only be imposed on the investor, not on the State … because States were likely to be involved in several disputes unlike the investors” (para 111, 40th Session Report, February 2021).
By September 2022, that shift was visible in the draft text. The Secretariat proposed applying SfC to “a party making a claim (including a counterclaim…)” (para 20, WP219, July 2022), to ensure “equitable treatment of claimant investors and respondent States” (para 21). The scope was thus subtly redefined as an issue of ‘equal treatment’ rather than ‘corrective protection’. Subsequent drafts retained the same framing. In WP231 (July 2023), both claims and counterclaims were explicitly included. One year later, in WP244, square brackets around ‘counterclaim’ were introduced, but the annotations hinted at tension: “[t]he Working Group may wish to consider whether only claimants should be ordered security for costs, or if a respondent State making a counterclaim may also be ordered security for costs” (para 18, WP245, July 2024).
By the end of 2024, however, the record was clear: a substantial majority—including the EU and its Member States, Algeria, Argentina, Brazil, Chile, Colombia, the Dominican Republic, Ecuador, Mexico, Panama, and Viet Nam—opposed extending SfC to counterclaims. A small minority (Australia, Singapore, and Switzerland) invoked a need for ‘equal treatment’ of claimant and respondents on SfC (Compilation of Comments). Despite overwhelming opposition, the Secretariat’s June 2025 draft (WP253) retained this broad scope. Paragraph 1 in Draft Provision 5 stated that a tribunal may “order any party making a claim to provide security for costs,” with annotations clarifying that ‘claim’ includes counterclaims (para 24, WP254, June 2025). In effect, a minority position became the default formulation.
Negotiating the ‘Compromise’
At the September session in Vienna, several States objected to the formulation in WP253. Argentina noted that applying SfC to counterclaims would invert the provision’s logic: States are not flight risks and public budgets cannot be mobilized to post securities. Panama, the Philippines, Nigeria, and Türkiye agreed, warning that such a rule would deter States from filing counterclaims and deepen existing asymmetries. Viet Nam and India reminded the room that SfC originated from the problem of cost recovery by States, not concerns about investors’ exposure to counterclaims.
Nevertheless, Switzerland praised the inclusion of counterclaims as a “welcome compromise,” invoking ICSID’s Rule 53 and its principle of “equal treatment of the parties”. Canada followed with similar language: “We can live with this compromise … It should be … neutral in its form for the reasons expressed by Switzerland and providing for equality of the parties.” The United States loosely aligned itself with Canada and called on the Working Group to “maintain the compromises that have already been made”.
In response, several States, including the Russian Federation and the Republic of Korea, pushed back against the ‘compromise’ narrative, supporting Argentina’s point that expanding the scope of the SfC provision risks reintroducing the very imbalance the reform sought to correct. They also noted that many States operate under strict budgetary constraints and lack the flexibility to commit public funds on short notice to satisfy cost orders.
Viet Nam and India re-entered the discussion to underscore that SfC was conceived to address the difficulty of cost recovery by States, not to impose new obligations on them. Moreover, they noted that even if a State loses an arbitration, the cost award against it would be negligible compared to potential damages. India reinforced this logic: “The purpose of an order for security … is to preserve the balance of convenience in favor of the respondent State, which is often compelled to defend itself in frivolous arbitrations initiated by claimants. … Expanding the definition of claims to cover counterclaims would instead shift that balance of convenience towards the claimants bringing such frivolous proceedings.” Several States concurred.
The EU and its Member States, while formally opposed to extending SfC to counterclaims, proposed a ‘compromise’: limiting SfC against States to ‘exceptional circumstances’. But without defining what ‘exceptional’ means, this formulation could expand tribunal discretion and paradoxically shift power away from States. By week’s end, the Chair echoed the compromise narrative: “… a number of States would like States to be exempt entirely … others … don’t think States should be exempt. … There is an equality … of the parties here.” However, the draft provision allowing SfC against “any party making a claim” was not the product of consensus or of deliberative compromise; it was the product of rhetorical drift, where ‘balance’ had been recast as ‘sameness’.
What began as an effort to redress asymmetry now risks deterring States from filing counterclaims altogether. The so-called ‘compromise’ has become a mirror image of the imbalance it was meant to fix.
Deliberation, Framing Battles and the Re-Emergence of State Power
The Vienna discussions reveal two broader points about multilateral lawmaking. First, as time pressure intensifies, States engage less in deliberation (argumentation) and more in discussions over what qualifies as a ‘compromise.’ In these framing battles, less attention is paid to the normative and empirical grounds of proposals and more to anchoring competing positions between which compromise can be claimed. Second, as negotiations near their end, reasoned arguments wane while traditional forms of State power become more visible. In this context, the political weight and influence of delegations often matter more in determining which compromises advance than the strength or merits of the arguments behind them—arguments that might otherwise help define what a genuine just compromise would look like.
It is telling that in Vienna this September, delegations from Canada, Switzerland, and the U.S. repeatedly invoked ‘compromise’, ‘equality’, or ‘neutrality’, when advancing for the contested SfC text. These actors have both strategic interests and structural power—economic, institutional, and political—that afford them leverage in framing what counts as an acceptable compromise. By contrast, smaller or less-resourced States, with less structural leverage, must defend each tweak to the text proposed by the Secretariat, often relying on coalition-building or procedural maneuvers to demonstrate that their objections are not obstructionist but reasoned. While the SfC deliberations in WGIII are not finalized, it seems increasingly likely that the final outcomes will reflect not only legal reasoning but also the balance of leverage and control over the reform narrative.
