Cost allocation in investment arbitration: Back toward diversification


Columbia FDI Perspectives

Perspectives on topical foreign direct investment issues by the Vale Columbia Center on Sustainable International Investment


Cost allocation in investment arbitration: Back toward diversification


Baiju S. Vasani and Anastasiya Ugale*

In 2006, the Thunderbird tribunal, operating under the UNCITRAL Arbitration Rules, called for the harmonization of cost-allocation approaches in commercial and investment arbitration.[1]Subsequent tribunals appear to be heeding Thunderbird’s call paving a trend in favor of the so-called “costs follow the event” (CFtE) approach and its variations.[2] Generally, this approach prescribes the shifting of arbitral costs and reasonable legal fees to the unsuccessful party (or based on parties’ relative success) and has historically been prevalent in commercial arbitration. By contrast, the more traditional approach in investment arbitration has been to share the costs of arbitration equally, save for special circumstances, with each party covering its own legal fees (traditional approach).[3] In the wake of what appears to be an emerging trend in favor of a default CFtE custom, it is time to revisit the idea of whether a single harmonized approach to cost allocation is really appropriate. We suggest that it most likely is not.

The two above-referenced rules of cost allocation serve different purposes, having divergent implications for the parties. From the claimants’ perspective, taking into account absolute cost of arbitration, CFtE is largely a deterrent, while the traditional approach encourages arbitration.[4] Simply put, a claimant is less likely to initiate an arbitration when it risks paying not only its own expenses, but also those of its opponent.[5] CFtE makes arbitration less appealing to claimants (and would-be third-party funders), more risky and/or outright economically unviable, if a claim’s value is lower than the absolute cost of arbitration. Under the traditional approach, on the other hand, the absolute cost is likely to be significantly lower and, above all, more certain, as each party is responsible for its own portion of the costs.[6] As a result, by encouraging the parties to try for their day in court, the rule leads to a more dynamic development of arbitral jurisprudence and broader access to justice.

These, and potentially other, implications might not necessarily resonate with the goals of a particular arbitral forum. In the context of ICSID, particularly, the CFtE approach is less likely to meet that forum’s goals. As a dispute-resolution center, ICSID is unique, as it sets an agenda “attaching particular importance” to the “availability” of the arbitration facility to the parties.[7] No set of commercial arbitration rules explicitly affirms such goals. The negotiating history of the Convention also demonstrates that the parties voiced the concerns the traditional approach seeks to eliminate.[8] On the contrary, the arbitration rules of many commercial arbitral institutions promote an arbitration mechanism impartial to the general availability of the forum to any party.[9]

Therefore, where the traditional approach as a baseline might meet the priorities of the forum, CFtE might fail them entirely, and vice versa. Consequently, the default customs of cost allocation based not on the relative success but on matching the policy goals of a forum and the effects of the customs might be a better lodestar for arbitral discretion. Once ascertained, the tribunals should consistently apply the default custom instilling confidence in the arbitration system. Of course, in certain instances costs should be shifted regardless of the default rule due to such factors as party conduct during the arbitration, the egregiousness of the respondent’s actions, or the fraudulent/frivolous nature of the claims. But, a default CFtE custom in the context of ICSID seems inapposite just at a time when it appears to be gaining popularity.