I am grateful to Professor Dominic Regan and Nicholas Bacon QC for sending me a copy of the decision in Essar Oilfields -v- Norscot [2016] EWHC 2361 (Comm).A decision of His Honour Judge Waksman QC sitting as a Judge of the High Court. (Judgment available here essar-oilfields-v-norscot-15-09-16jud-1-2)


  • The phrase “other costs of the parties” in s.59(1)(c) of the Arbitration Act 1996 includes the costs of obtaining funding.
  • Whether to award those costs was always a matter of discretion for the arbitrator.

This case probably marks a major difference between costs recoverable in arbitration and in litigation. It is unlikely that the costs of litigation funding will be found to be recoverable as costs in litigation.


Essor was seeking to set aside part of an arbitrator’s award whereby the costs of litigation funding was allowed as part of the arbitration.The arbitrator held that he was entitled so to order in his discretion, because such litigation funding costs were “other costs” for the purpose of s.59(1)(c) of the Arbitration Act 1996, which refers to “legal or other costs of the parties”.


The judge’s primary finding was that the decision was that there was no serious irregularity in the arbitrator’s construction of “other costs”. This was sufficient to deal with the matter.  However the judgment went on to consider the arguments in relation to construction.


The Context
48. This is the underlying substantive point of Essar’s complaint, which is that the expression “other costs” does not, as a matter of construction, include the costs of obtaining litigation funding in respect of the arbitration in question. I have already set out the relevant provisions in the Act and the ICC Rules.
49. As a preliminary point, Essar contends that the relevant sections in the Act must be construed essentially by reference to what a court would or could allow by way of costs in litigation under the CPR. I reject that. Of course, both arbitration and litigation are forms of formal dispute resolution and there are many similarities, but it is crucial to keep in mind that the Act was designed to be and is a complete code as to the conduct of arbitration, subject to some well-established exceptions. In particular, s.81(1) of the Act provides that: “Nothing in this Part shall be construed as excluding the operation of any rule of law consistent with the provisions of this Part, in particular, any rule of law as to – (a) matters which are not capable of settlement by arbitration; or (b) the effect of an oral arbitration agreement; or (c) the refusal of recognition or enforcement of an arbitral Award on grounds of public policy”.
50. Mr. Hogan submits that this provision means that the Act and, in particular, ss.61 and 59 are to be read subject to whatever the common law rules as to recoverable costs under the CPR are. I reject that also. Section 81 is concerned with the particular matters as set out within it, all of which are fundamental questions going to jurisdiction, enforcement and the like. It is not concerned with procedural differences which may exist between the arbitral and CPR regimes on matters such as costs. Section 81 is, therefore, irrelevant.
51. The difference between these contexts is made abundantly clear where s.59(1)(c) defines the costs of arbitration as including, not just legal costs, but “other costs” too. There is no parallel provision in the CPR. The CPR’s own definition of costs in 44.1 is clearly more limited. It is true that the arbitrator did refer in the Award and the addendum to CPR 44, but it is plain that this was essentially because of the provisions dealing with standard and indemnity costs and the different bases of assessments, as opposed to going to the meaning of “other costs”. Accordingly, the approach taken by the courts under the CPR as to what can and cannot be Awarded by way of costs is of little direct relevance here. The relevant context is thus the Act itself and the wide scope of procedural powers conferred upon the arbitrator.
The language
52. In fact, of course, the correct starting point is not the wider context, but the language of the provisions themselves. Sections 63(3) and 61(1) allow the arbitrator to determine the recoverable costs of the arbitration as he sees fit. Section 59(1)(c) then deliberately includes a head of costs, other than legal costs.
53. Essar seeks to cut down on the scope of the provision by saying that the governing expression is really “costs of the arbitration” and that in itself would exclude the costs of third party funding, since the latter is not the cost of the arbitration, but the costs of funding it. However, that is the wrong way round. It is the collection of items in s.59(1) itself which defines what the costs of the arbitration are. “Costs of arbitration” is not some prior limiting definition.
54. I accept, of course, that “other costs” has to be seen as other costs which relate to the arbitration proceedings. But, in my judgment, that does not help Essar very much, because the question then is what such costs are or might be. Certainly, where a party to an arbitration is funding it by obtaining specific litigation funding which is now available in a variety of forms, so as to enable him to specifically enforce his legal rights, it is very hard to see how that is excluded for all purposes from the expression “other costs”. Indeed, Mr. Hogan in his submissions came close to accepting that when he said that other costs connoted “something necessary to get the arbitration off the ground or on the road”. Here, at least, that could be said to include the costs of third party funding.
55. Essar, of course, has to volunteer something for “other costs” to cover otherwise theexpression would be wholly meaningless. It says that it would include some costs not available in a court-based costs order, for example, internal expert fees or managerial time, but without extending to litigation funding. That is a somewhat arbitrary distinction to draw. The better view, as noted above, is to look at the expression functionally.
56. Subject to the question of assessments of, for example, proportionality and reasonableness and the like, conventional legal costs in the sense of lawyer’s fees and disbursements are incurred in order to bring or defend the claim in question. There can be other costs also incurred to the same end. These could be management time and they could also be the costs of obtaining funding for the dispute.
57. Mr. Hogan also contends that the expression “other costs” must be construed eiusdem generis with legal costs being a residuary class, and accordingly, it is to be construed narrowly so as to cover only those costs that are truly analogous to legal costs. That would allow only modest extensions therefrom the examples that have already been cited.
58. I do not accept that the eiusdem generis rule can be said to apply in this way here. “Legal costs” is not some defining “genus” whereby the use of “other” may be expressed simply to catch any costs which almost by accident, as it were, fall outside the definition of “legal costs”. The better “genus”, in my view, is the costs of the arbitration to be regarded in a broad sense. There are legal costs of the arbitration and there are other costs of the arbitration. The real limiting factor, in my view, is the functional one. Do the costs relate to the arbitration and are they for the purposes of it? If the costs have not been incurred in order to bring or defend the claim in question, I would accept that they fall outside the definition of “other costs” and they would not relate to the arbitration – but that is not the case here.
59. Mr. Hogan also points to the fact that s.59(1) does not refer expressly to costs of obtaining third party funding, but that is hardly surprising, for, if it were otherwise, s.59(1) would have to include each and every example of what could or did include “other costs”. The lack of a specific reference to third party funding is immaterial, in my view.
60. Mr. Hogan also contends that the true source, if any, of the arbitrator’s ability to award the costs of third party funding is not this part of s.63 and s59 on costs, but rather, its power to Award interest under s.49. He accepts that, if the current Award was set aside, it would have to be remitted to the arbitrator and Essar would have to deal with the merits of an argument that Norscot was entitled to such costs by way of an interest Award. I do not here analyse the scope of s.49. But, even if the arbitrator could so Award on that basis, I fail to see why that, in and of itself, narrows the scope of s.59(1)(c). Such entitlement on the part of Norscot could exist side by side. For the same reason, I do not see how the Award is somehow in fundamental conflict with the scheme of the Act.
61. Since Rule 31(1) of the ICC Rules is expressed in substantially the same terms as s.59(1)(c), the ICC Commission Report of 2015 headed “Decisions on Costs in International Arbitration” is relevant. Its preliminary note reads: “The considerations contained in this Report are intended to inform users of arbitration how tribunals may allocate costs in accordance with the parties’ agreement and/or any applicable rules or law. However, they should not be regarded as affecting a tribunal’s discretion to allocate costs”.
62. The Report did not endorse any particular approach to decisions on costs, nor establish guidelines or checklists.
63. Then, under the heading “Third Party Funded Costs”, at para.87 it said: “The successful party will itself ultimately be out of pocket upon reimbursing such costs to the third-party funder and may therefore be entitled to recover its reasonable costs, including what it needs to pay to the third-party funder, from the unsuccessful party. The tribunal will need to determine whether these costs were actually incurred and paid or payable. The fact that the successful party must in turn reimburse those costs is, in itself, largely immaterial”.
64. And at para.90: “If there is evidence of a funding arrangement that is likely to impact on the non-funded party’s ability to recover costs, that party might decide to apply early in the proceedings for interim or conservatory measures”.
65. Then, at para. 92, under “Success fees and uplifts”, it says that: “In reality, funding arrangements are rarely limited solely to the costs of the arbitration. Usually, the third-party funder will require payment of an uplift or success fee. … As a tribunal only needs to satisfy itself that a cost was incurred specifically to pursue the arbitration, has been paid or is payable, and was reasonable, it is feasible that in certain circumstances the cost of capital, e.g. bank borrowing specifically for the costs of the arbitration or loss of use of the funds, may be recoverable”.
66. And finally at para. 93: “The requirement that the cost be reasonable serves as an important check and balance in protecting against unfair or unequal treatment of the parties in respect of costs, or improper windfalls to third-party funders. Tribunals have from time to time dealt with this when assessing the reasonableness of costs in general, sometimes including the success fee in the allocation of costs and sometimes not, depending on their view of the case as a whole”.
67. I regard all of those observations as highly pertinent here. Of course, they are not determinative on the issue, nor are they authorities, but they do in their reasoning support the functional view that I have expressed above and the notion that the ICC regime is not to be regarded as subservient to the CPR regime on such costs issues.
68. Therefore, as a matter of language, context and logic, it seems to me that “other costs” can include the costs of obtaining litigation funding. The expression should not be confined by some legal straightjacket imposed by reason of what a court might or might not be permitted to order. All that this conclusion entails is that such litigation funding costs falls within the arbitrator’s general costs discretion. Whether and, if so, how the arbitrator exercises that discretion in any particular case is an entirely different matter. Indeed, the ICC bulletin at para.93 reminds one that the overall requirement of reasonableness can act as an important check and balance here.
69. The arbitrator’s exercise of his discretion here to award to Norscot the costs of its third party funding, while, of course itself not under challenge, is nonetheless a telling example of the good sense of reading “other costs” in this way. This was a case, perhaps unusual, where the arbitrator ruled in detailed and robust terms that Essar drove Norscot into this expensive litigation because of its own reprehensible conduct going far beyond technical breaches of contract, in order to vindicate its rights. Further, as the tribunal found, Norscot had no option, but to obtain this funding from this third party funder. As a matter of justice, it would seem very odd and certainly unfortunate if the arbitrator was not entitled under s.59(1) (c) to include the costs of obtaining third party funding as part of “other costs” where they were so directly and immediately caused by the losing party.
70. In my judgment, therefore, I unhesitatingly conclude that the arbitrator’s interpretation of “other costs” was correct, in that it extended in principle to the costs of obtaining third party legal funding. Whether then to Award it is a matter of discretion.
71. There is a subsidiary issue between the parties as to whether, in fact, the court could not order by way of legal costs some of the items that Essar said it could not and which Norscot said it could or might. A further group of cases was debated here. I accept that there are instances where the court has been prepared to include by way of costs under the CPR somewhat more than Essar would allow, but nothing much turns on this for the reasons already given.
72. Thus the arbitrator was entitled to interpret “other costs” so as to include the costs of third party funding. There was therefore no error of law anyway. Given my decisions on the two key issues of Characterisation and Construction in favour of Norscot, I deal with the remaining issues somewhat more shortly.
73. Even if there had been serious irregularity, it must have caused substantial injustice to the party now complaining of it. In one sense, it can be said that it did, because, as a result, Essar has to pay something approaching £2 million, which, on this hypothesis, it could and should never have been ordered to pay by the arbitrator.
74. However, Norscot argues that there is no substantial injustice since it was Essar which, on the clear findings of the arbitrator, forced Norscot into purchasing the third party litigation funding in order to vindicate its rights anyway, and I have referred to the relevant paragraphs above. Norscot further contends that, in deciding if there has been substantial injustice, the court must enter into some sort of balancing exercise and pay regard to the justice or otherwise of the other party’s position if the serious irregularity is not set aside.
75. Mr. Karia referred me, in particular, to the decision of Burton J. in CNH v. PGN [2009] 1 CLC 807. Here, the arbitrator had power to award interest on damages from the date of breach, but, for some reason, only awarded them from the date of the claim. The arbitrator subsequently purported to correct the award by adding a further and very substantial amount for interest; i.e. doing what he could and should have done in the first place. Having found that the arbitrator had no power to take that course by means of a corrective power, Burton J. went onto consider the question of substantial injustice. He said here that:
“All that is required is to reverse the procedural irregularity. A reversal of this procedural irregularity [he said] would then cause that substantial injustice – namely [to pay the interest be removed]. In my judgment it cannot be possibly arguable that it would cause substantial injustice to the Claimant if the procedural irregularity were reversed and the correction of the howler prevented, if so doing, would cause, on the one hand, a substantial injustice to the Defendant and, on the other, a wholly undeserved windfall to the Claimant”
. He said much the same in para.43.
76. I follow all of that, but the core point on those facts surely was that there was no true injustice caused to the claimant in that case, because, had the arbitrator not made the original error, he would have made the payment of interest all the way through at the outset and it could not be just for the claimant to escape that by reason of the irregularity. The converse, of course, was that it would be unjust to deprive the other party of such interest. I do not read from that a principle that the court should, in fact, undertake a wideranging balancing exercise looking at justice in the round. The question of whether substantial injustice has been caused may arise more acutely where it is not clear if the irregularity has led to any loss at all. But, in our case, at least on the footing that the arbitrator was wrong and it was an irregularity, he could not have Awarded the third party funding in the first place. So, in setting that order aside, it could not be said that Essar was obtaining a windfall.
77. I accept that, on the arbitrator’s finding on the facts, Essar did cause Norscot to incur the third party funding costs. But, if, in truth, they were irrecoverable in the arbitration, then its conduct in this context is much less relevant. The more important direct point, therefore, is that, absent the irregularity, Essar would not have had to pay the substantial sum awarded against it. Wider questions of justice and fairness in respect of Norscot do not trump that, in my view. Accordingly, had s.68(2)(b) otherwise been made out, the serious irregularity here on substantial injustice would also have been made out. In the event, it is academic. “