Briefing

Diverging approaches to security for costs across arbitration seats and institutions

Background

Security for costs, long relied upon to protect a party’s ability to recover its legal costs should it ultimately prevail, is widely considered an essential procedural tool. Despite such popularity, there is no uniform approach to security for costs in international arbitration, be it in terms of threshold, evidentiary expectations, or appeal of the decision ordering security.

This article outlines critical differences on security for costs across selected seats and institutional frameworks, and the related and growing trends.

Differing approaches according to the arbitral seat

Although arbitral tribunals generally apply a comparable set of considerations when deciding applications for security for costs based on the arbitral seat and Institutional Rules, approaches differ in terms of the weight given to the proof and criteria required.

In brief, security for costs pre-supposes:

a credible prospect of a costs award;
a serious risk of nonpayment of costs;
a timely, well-evidenced application; and
whether an order for security is fair in all circumstances

1. The Swiss System: (Chapter 12 PILA)

Swiss-seated tribunals generally treat security for costs as a form of interim or conservatory relief included in their power to order interim measures. As such, an arbitrator’s right to order security for costs is derived from Art. 183(1) of the Swiss Private International Law Act (PILA).

Considered a provisional measure, orders for security for costs cannot be challenged before the Swiss Federal Supreme Court (SFSC) unless the measure would cause irreparable harm to the party ordered to pay. Pecuniary harm would of itself rarely be considered irreparable, and therefore challenges to security for costs in Swiss-seated arbitrations are unlikely to succeed.

In terms of the criteria for an order, Swiss law does not codify a detailed, seat-specific test for security for costs. Instead, it leaves tribunals significant discretion to balance the requesting party’s interest in cost protection against due process and access to justice concerns. Jurisprudence and doctrine nonetheless require: i) evidence of a reasonable degree of certainty of a future claim for reimbursement of costs, and ii) an imminent risk to the reimbursement of costs.

As to quantum, parties must limit themselves to the costs estimated at the time of filing.

2. England & Wales (Arbitration Act 1996, 2020, 2025)

In England and Wales, the arbitral tribunal’s power to order security for costs is expressly stated in Section 38(3) of the Arbitration Act 1996, which provides that “[t]he tribunal may order a claimant to provide security for the costs of the arbitration”.

An order for security for costs is a procedural order, not an award. Therefore, a challenge is possible only in exceptional circumstances, notably via Section 68 of the Arbitration Act 1996 (as amended by the Arbitration Act 2025), where the manner in which the order was made constitutes a serious irregularity causing substantial injustice. In addition, and due to the procedural nature of the order, it is not enforceable via the New York Convention 1958, nor is it subject to scrutiny by the Institution, as it would be if it were an award.

Influenced by litigation-style analysis and detailed costs submissions, arbitral tribunals seated in England and Wales commonly require itemised and well-supported budgets, scrutinising whether an order would be fair and proportionate in light of the claimant’s ability to pursue the claim.

As opposed to Swiss-seated arbitrations, parties can include all costs from the commencement of the arbitration, as well as costs covering future phases of the proceeding. Section 59(1) Arbitration Act 1996 (as amended) provides that costs can include: the arbitrators’ fees and expenses, the arbitral Institution’s fees and expenses, and the legal or other costs of the parties

Differences in institutional rules: Codification and reliance on interim powers

While there is a clear trend to recognise the right to order security for costs in international commercial arbitration, and a consensus in considering it a provisional measure, arbitral Institutions still differ in how they address the issue.

Owing to its increasing prevalence in proceedings, several Institutions have recently elected to address security for costs expressly in the latest revisions of their rules. For example, Article 38, Stockholm Chamber of Commerce (SCC) introduced a specific provision on security for costs in its 2017 Rules, followed by the London Court of International Arbitration (LCIA) in Article 25.2 of its 2020 Rules, and last year the Singapore International Arbitration Centre (SIAC) included provision in Rule 48 of its 2025 Rules.

Where Institutional rules are silent, arbitral tribunals have authority to order security for costs from their traditional broad interim-measure powers. This is notably the case for arbitrations in the Swiss Arbitration Centre (SAC), where Article 29 of the 2021 Rules will be relied upon, as is the case for arbitrations under the International Chamber of Commerce (ICC), with arbitrators relying on Article 28 of the 2021 Rules or Article 29 of the 2026 Rules that come into force on 1 June 2026.

The historic discrepancy in a codified approach has meant some jurisdictions were reluctant, if not opposed, to recognising a right to order security for costs. With the global international arbitration world having seemingly accepted an arbitral tribunal’s right to order security for costs, this divergence should no longer be regarded as cause for concern by parties.

Security for costs and prejudgment on the merits

An order of interim measures usually implies a successful “prima facie case on the merits” test, in which the requesting party must convincingly demonstrate a probability of succeeding on its claim. Arbitrators may shy away from ordering these measures, particularly when they seek a provisional grant of the relief sought in the final award, for fear of casting doubt on their impartiality.

Security for costs, despite being an interim measure, does not appear to trigger the same reticence in arbitrators, at the very least in Switzerland.

This may be explained by the fact that a “prima facie case on the merits” was not traditionally part of the criteria for an order of interim measures in Switzerland. Indeed, building on Art. 183(1) PILA’s basis, and drawing from Swiss civil procedural law, Swiss doctrine considers that the requirement for an order of security for costs should be a “potential future claim for the reimbursement of costs worthy of protection”. Therefore, to succeed respondents must demonstrate with a reasonable degree of certainty that they will be awarded reimbursement of their costs should they win. This requirement does not consider the likelihood of succeeding on the merits but rather focuses on the assessment of cost allocation should – regardless of its actual chances – the respondent win.

Switzerland’s approach cuts short any potential pre-judging and assorted threats of bias, ultimately making it far friendlier to security for costs than at the start of the millennium.

In jurisdictions, or under Institutions, requiring a prima facie case on the merits, arbitrators should nonetheless not give way to reticence for fear of appearing biased. An increasingly large portion of the literature convincingly argues that prima facie should not be regarded as potentially prejudging the merits of the case as “it is purely a provisional assessment based upon incomplete submissions and evidence, without preclusive effects”.

Looking ahead: Trends and developments

In a number of recently revised Institutional rules, security for costs is at the forefront of developments in International Arbitration. Of note is also the rising importance of third-party funding, already well established in England and often used in arbitration proceedings either for a single funded case or via a portfolio facility to cover multiple claims.

While most practitioners would consider third-party funding as a non-issue, some debate remains as to whether it should nonetheless constitute a prima facie case for security for costs.

The initial discomfort around third-party funding appears to have stemmed from the, yet to truly be substantiated, fear that claimants would be unable to comply with adverse cost awards and the funding agreement would not have provided for the funder to be liable for the adverse costs – otherwise referred to as a so-called “arbitral hit and run”. Some practitioners and authors have therefore argued that the presence of third-party funding should be treated as prima facie evidence of impecuniosity, thereby supporting an order for security for costs. Another position notably shared by Institutions views third-party funding as a “relevant circumstance to be assessed” when security for costs is requested. Authors and practitioners rallying to the former consider that external funding alone should not be interpreted as impecuniosity of a claimant, since financially stable parties may rely on third party funding as a risk-management tool.

There is some anecdotal evidence that where a claimant has obtained funding, respondents will be keen to seek security for costs against them- on the basis that this evidences their lack of funds, unless they have adequate adverse costs (ATE) insurance, which can satisfy any costs award against them. However, that is oversimplistic as claimants may choose to take out funding for reasons other than financial limitations, for example to de-risk or share the risk of the legal costs, or for strategic or tactical reasons e.g. to show that their case is strong enough to secure that level of investment. Nonetheless, sensitive to the possibility of insolvent claimants being unable to comply with an adverse award on costs when supported by external funders, Institutions and arbitrators alike have been pushing for clarity on the terms of funding agreements.

Initial concerns, mostly regarding impartiality and independence between the arbitrators and external funders, led to rule revisions by Institutions such as the ICC, SIAC, and HKIAC to include an obligation to reveal the existence of a funding agreement and the identity of the funder. Further, in its 2024 updated Rules, the IBA included a requirement for the disclosure of the identity of those who may have “a direct economic interest in the prosecution or defence of the case in dispute, a controlling influence on a party to the arbitration, or influence over the conduct of proceedings, including the selection of arbitrators”, which has been deemed to include funding and funders. Finally, the recent CIArb Guidelines on Third Party Funding, recommends early disclosure that there is a funding agreement in place and of the identity of the funder. However, it is interesting to note that in their last Rules update, the LCIA did not include a provision for disclosure of a funding agreement or funder; these Rules are currently under review and so the position may change.

The debate has since shifted to the disclosure of the funding agreement’s terms, particularly where the extent of the financial coverage and withdrawal terms are concerned. While this mandatory disclosure has been observed in investment arbitration, commercial arbitration tribunals have – to our current knowledge – yet to order it.

Although the negative bias against third-party funding has been discussed at length within the community, the possible positive bias of arbitrators against externally funded claims has not been the subject of the same level of scrutiny, aside from the impartiality-led disclosure obligations. Considering prospective claimants undergo a stringent vetting process, at the end of which only 5-10% may see their claims funded, there could be a legitimate argument as to whether an arbitrator could be biased towards a claim which has been scrutinised and deemed worthy of financial investment.

To date, no clear consensus has emerged in international practice. Most arbitral tribunals nonetheless appear to be assessing third-party funding on a case-by-case basis, considering it as a factor among others, and rejecting the presumption that funding alone implies an inability to meet an adverse costs order (or even impecuniosity) and justifies an order of security for costs.

As third-party funding rises in popularity, with a growing number of jurisdictions endorsing the practice as being commercially sensible, more guidance will be issued by Institutions.

Commentary

Despite the current wave of codification ensuing from the increased prevalence of security for costs applications, this interim measure remains a very discretionary and fact-sensitive application, with limited possibilities for review.

As considered above, Institutions and arbitral seats (governing the process) offer parties varying degrees of flexibility, especially where the quantification and degree of detail as to costs are concerned. Parties and practitioners alike should nonetheless be cautious of the absence of rigid statutory rules, as granting arbitral tribunals discretionary powers will inevitably cause them to balance interests and fairness, and consequently place a significant evidentiary burden on the parties.

Marie Widman, Paralegal, assisted in the preparation of this briefing.