By Aryan Sharma.
Aryan Sharma is a 4th-year undergraduate law student at Maharashtra National Law University Mumbai.
International arbitration promises finality and efficiency, yet permissive cost regimes in major seats allow losing parties to mount weak challenges with minimal financial risk. This creates incentives for delay and settlement tactics by forcing award creditors to bear heavy legal expenses. In this regard, this article argues for stronger fee-shifting rules through a comparative analysis of jurisdictions such as Singapore, France, the United States (U.S.), England, Hong Kong, and Dubai International Financial Centre (DIFC). It advocates for a presumptive indemnity costs model to prevent frivolous setting-aside applications and at the same time preserving access for meritorious challenges.
Keywords: set-aside applications, costs, fee-shifting, frivolous challenges
International arbitration is hailed as an efficient mode of dispute resolution for cross-border commercial disputes. Parties opt for arbitration because of its finality and enforceability under the New York Convention. However a paradox undermines these very principles, i.e., in major arbitration seats, losing parties face minimal financial consequences when mounting unmeritorious challenges to arbitral awards. Such an imbalance generates perverse incentives.
The party who faces an adverse ruling can delay enforcement for months through court proceedings and recover little of the award creditor’s legal costs, even when the challenge fails. On the other hand, the award creditor ends up paying heavy legal costs despite prevailing in the arbitration as well as the subsequent court proceedings.
This article examines the costs regimes across major arbitration seats and argues that current approaches in most jurisdictions deter frivolous challenges which undermines arbitration’s core purpose. It draws on comparative analysis and proposes that courts should adopt more robust fee-shifting mechanisms.
i. Permissive Approach
The majority of the seats are best described under what may be labelled a “permissive” cost regime. In Singapore, recoverable costs are awarded on a standard basis and recover between 40-60% of actual lawyers’ fees incurred. The Singapore International Commercial Court (SICC) has marginally improved recoveries, but even then, indemnity cost awards are rare. In France, there is wide discretion in allocating costs [including legal fees and indemnity] by applying a “costs follow the event” approach. This is given under Article 700 of the French Code of civil procedure.
The U.S. is an extreme example of this approach in practice. After the “American rule,” each party pays their costs in full, without recovering anything in successful cases. This applies similarly in the courts in DIFC in onshore courts, where only nominal court fees can be recovered.
England and Wales (E&W) occupy middle ground. The general rule there is that the losing party pays the winning party’s costs and costs are usually assessed on the standard basis, where the receiving party must show costs are reasonable. Indemnity costs may be awarded in limited circumstances, such as unreasonable conduct.
ii. Deterrent Approach
Hong Kong stands alone among the major arbitration seats in adopting a salutary practice: indemnity costs are awarded against unsuccessful challengers to arbitral awards, in case no special circumstances exist. This emerged from A v R, which held that parties should not be permitted to undermine awards through unmeritorious challenges.
A similar approach is taken by the DIFC courts, where substantial cost awards are regularly granted. This accords with not only the common-law underpinning of the DIFC but also its objective of attracting international commercial parties.
Such divergent approaches produce divergent outcomes. Hong Kong, despite being consistently ranked as the 3rd most preferred arbitration seat globally [and having a comparable caseload to Singapore International Arbitration Centre (SIAC)], sees fewer setting-aside applications. Over a recent seven-year period, Hong Kong heard just 27 reported challenges compared to Singapore’s 95.
The DIFC reports minimal challenges to awards, with zero successful applications in recent years. By contrast, Singapore has witnessed growth in setting-aside applications. Parties invoke creative grounds, mainly allegations of natural justice breaches that blur the line between procedural irregularity and problem with tribunal’s substantive reasoning.
i. Finality
Finality is arbitration’s foundational promise. The UNCITRAL Model Law provides only narrow grounds for challenging awards precisely because parties chose arbitration to avoid prolonged litigation. The U.S. Supreme Court emphasized in Hall Street Associates v. Mattel, Inc., judicial review of arbitral awards must be “exceedingly deferral.”
Yet permissive costs regimes undermine this policy. When an award debtor can mount a challenge knowing it will cost the award creditor US$150,000 in legal fees [of which perhaps US$60,000 might be recovered], the calculus changes. The debtor gains 18 months of delay at relatively low cost which creates leverage for settlement negotiations that extract concessions from the rightful winner.
Success rates for setting-aside applications hover around 15-25% across most jurisdictions. Across major arbitral seats, the vast majority of challenges to awards are unsuccessful, with success rates often below 40%, and in several jurisdictions below 10%. For example, only about 38% of challenges succeed in E&W, and single-digit success rates [8–11%] were observed in New York, Bahrain and onshore UAE courts. This means majority of challenges fail, yet in most seats, these unsuccessful challengers face limited financial consequences. The system thus subsidizes at the expense of award creditors.
ii. Fair Compensation
Award creditors already incur significant costs in the process of obtaining an enforceable arbitral award. However, if they must defend their arbitration award in court, it costs them significantly in attorneys’ fees. Even if the party succeeds entirely, it only offsets a percentage of the costs.
The situation worsens when the issue is considered in the context of access to justice. Well-financed award debtors can conduct an act of economic warfare through setting-aside actions, knowing that the costs of defending themselves [even if partly recoverable] will force creditors into an unfavourable settlement. This is even more concerning in disputes where parties are on a very different economic standing.
Moreover, certain grounds for setting-aside awards are prone to manipulation, such as violations of natural justice, due process, and public policy, as they can be alleged in basically any case. Although it is true that courts consistently reject such arguments when there is lack of genuine circumstances, but each allegation demands a response. In absence of a deterrence like imposition of costs, risk-averse award debtors that face substantial awards have no reason to try.
In France, where the practice of recovery of costs is limited, “public policy” challenges were invoked in 137 out of 222 cases examined [yet succeeded in only 6 cases, 4% success rate]. This shows over-pleading of a ground that rarely succeeds.
The first argument is that indemnity costs could deter meritorious challenges, which would result in a “chilling effect” on parties having valid grievances. Should parties seeking to set aside an award know that they faced a possibility of incurring the other party’s legal costs in full, would they still pursue a meritorious claim in challenging an award?
However, this concern is overstated. It is contradicted by empirical evidence from Hong Kong. Despite having a strong costs regime, their success rate for setting-aside applications is only 22%, very close to Singapore’s 23%. This suggests that a stricter costs regime does not necessarily produce a significant chilling effect on meritorious challenges. However, success-rate comparisons alone cannot conclusively establish deterrence, as that would require behavioural evidence such as survey-based data.
Moreover, truly meritorious cases involve genuine irregularities or jurisdictional defects that should be apparent from the award and record. Parties with legitimate claims can make informed assessments of their prospects.
Second argument pertains to the fact that in the U.S., the tradition of parties bearing their own costs is a different conception of access to justice. Fee-shifting is seen as deterring rights-enforcement and creating barriers.
This argument has less weight in the setting-aside context. Setting-aside proceedings are a form of appeal from private adjudication that parties contractually chose. The policy considerations that favour broad access to initial dispute resolution do not apply with equal force to what is essentially an appeal of a privately-rendered decision. Also, even American law recognizes exceptions for bad-faith litigation. Courts possess inherent authority to sanction parties for frivolous filings under Federal Rule of Civil Procedure 11.
Courts in leading arbitration seats should adopt a presumptive indemnity costs rule for unsuccessful setting-aside applications, subject to carefully defined exceptions. The default rule should provide that unsuccessful challengers pay the successful party’s reasonable legal costs in full which can be assessed on an indemnity basis. Costs should be determined by reference to the complexity of the case, the amounts at stake, and the prevailing rates for qualified counsel in the market.
The presumption should yield in specifically defined circumstances. Firstly, where a challenge raises questions of first impression of an issue of law or involves genuinely unsettled law, courts should retain discretion to reduce [or eliminate] adverse costs. Secondly, where a challenge succeeds on some grounds but fails on others, courts should apportion costs accordingly. Thirdly, where the costs claimed are genuinely excessive relative to the case’s complexity, courts should award only reasonable amounts.
Additionally, cases that involve factual questions about the validity of the reasoning used in the tribunal’s decision which are disguised as procedural issues can be summarily dismissed, with the consequence of enhanced costs. Moreover, courts should be clearer in their reasoning in setting the costs in setting-aside matters.
The experience in Hong Kong illustrates that robust fee-shifting is a discouraging factor for frivolous claims, without chilling meritorious ones. Other arbitration seats should emulate this model by adopting presumptive indemnity costs for unsuccessful challenges against awards.
Such reform would strengthen the core principles of finality and efficacy that arbitration promises. It would also deter abusive tactics and ensure that parties who have successfully defended against arbitration are not punished for that success.
Copyright© 2026 Milon K. Banerji Centre for Arbitration Law instituted at NALSAR University of Law, Hyderabad. All rights reserved.
