Yesterday, 3 November 2015, the London Court of International Arbitration ("LCIA") released costs and duration data regarding actual cases administered by the LCIA under the LCIA Rules. It is thought that this is the first time an arbitral institution has released such information. This move is no doubt part of a growing trend of, and indeed demand for, increased transparency in international arbitration. It remains to be seen how other arbitral institutions – and even perhaps some less transparent domestic commercial courts competing for big ticket commercial disputes –  will react and whether they will follow suit.

The data

The data provided is based on all LCIA arbitrations which progressed to a final award in the period 1 January 2013 to 15 June 2015 and were for a quantified sum.  When analysing these arbitrations, the LCIA calculated the total amount paid to the LCIA for administering the arbitration together with the arbitrators' fees.

The key takeaway figures and conclusions on costs are as follows:

  • The median and mean costs of an LCIA arbitration are US$ 99,000 and US$ 192,000, respectively.
  • On an overall comparison of arbitration costs when using median values, the LCIA concludes that the costs of an LCIA arbitration are (i) substantially lower than those of the International Chamber of Commerce ("ICC") and the Singapore International Arbitration Centre ("SIAC"), and (ii) comparable to those of the Hong Kong International Arbitration Centre ("HKIAC").
  • The LCIA also concludes that US$ 1 million is the 'turning point' at which the LCIA hourly rate system would be preferable to the ICC and SIAC's ad valorem systems. That is, for disputes under US$ 1 million the costs of LCIA, ICC and SIAC arbitrations are comparable. For disputes exceeding US$ 1 million, on the basis of this analysis, LCIA arbitration would appear cheaper and more cost-effective.

With respect to costs, the data highlights to users one of the key features and differentiators of LCIA arbitration: the hourly rate system by which parties are billed for the time actually spent by the arbitrators and the LCIA on their case. This is in contrast to the ICC and SIAC's ad valorem approaches whereby the costs of the arbitration are a percentage of the total amount in dispute. The LCIA says that this revelation could give 'food for thought' to those users who might have expected an hourly-based system to be cost effective only in the very largest of cases.

With regard to duration, the LCIA calculated the full period between the date on which the Request for Arbitration was received by the LCIA and the date of the final award. This included any stay of the arbitration whether informal or formal. The conclusions reached were that the median and mean durations of an LCIA arbitration are 16 and 20 months, respectively.  The LCIA was unable to compare its duration data to other institutions' because no such comparable data is presently available. However, the LCIA says that it hopes that other arbitral institutions will follow suit in making such information available to potential users.


The publication of this data is a welcome step towards increased transparency by arbitral institutions. Such information benefits users as it enables them to make more informed decisions regarding their choice of institution. In the longer term, the prospect of being objectively compared to other institutions may also provide a catalyst for institutions to improve the quality of services they provide. Such increased transparency would also reflect positively on the arbitral process and profession more generally, at a time when additional transparency in international arbitration is being actively sought by end users and society at large. Given the increasingly competitive market in which not only arbitral institutions but also national commercial courts (e.g. the English Commercial Court and the Singapore International Commercial Court) are competing for users and international cases, it will be interesting to see what the reaction is to the LCIA's publication of this data.