It is often said that managing an arbitration means also managing a client's expectations, including the client's costs in pursuing its case in arbitration. 1 As the legal costs incurred by parties in international arbitrations continue to rise, it becomes increasingly important for parties to assess the costbenefit and risk involved in opting for arbitration. It is of course the parties-and the counsel they retain-who are primarily responsible for the costs incurred in the conduct of an arbitration. However, arbitral tribunals, too, can play a significant, albeit indirect, role in controlling the costs of arbitration through their management of the arbitral process. 3 This is particularly true in ICC arbitration.

In this article, four ways in which ICC arbitral tribunals may help to control costs at different stages of the proceedings will be discussed. They concern:

(1) the drawing up of the procedural time table, (2) the provision of guidanceon what issues to brief, (3) the awarding of legal costs prior to the final award, and (4) and the awarding of costs related to document production. The first three points should be considered as the arbitration unfolds, while the fourth is a matter for the final award. A fifth and final point that will be considered is the tribunal's decision on costs, the assessment and apportionment of which are[Page179:]

settled in the final award. Decisions on costs, although often difficult to make, deserve careful thought, given the stakes involved. 4 Many tribunals are indeed aware of this and give considerable care to discussing the reasons for their decisions on costs. 5

1. Procedural timetable

According to Article 18(4) of the ICC Rules of Arbitration ('the Rules'), 6 at the time of drawing up the Terms of Reference or as soon as possible thereafter, the arbitral tribunal 'shall establish in a separate document a provisional timetable that it intends to follow for the conduct of the arbitration'. The provisional timetable typically sets out the procedural steps that are expected to be followed in the conduct of the arbitration and the time limits for achieving these, generally up to and including the dates of hearings. In some cases, when jurisdiction is challenged or liability is treated separately from quantum, the timetable may deal with one part of the arbitration and be updated as the arbitration progresses to the next part. The Rules also envisage the possibility of 'subsequent modifications' of the timetable.

Article 18(4) clearly seeks to encourage the acceleration of the arbitration process by having the arbitrators and the parties focus at an early stage on the procedural needs of the dispute. Although that aim has an impact on costs, cost considerations rarely seem to play a role or, at least, remain largely ignored in this context. For instance, each additional round of written submissions and/or witness statements adds to the overall cost. While written witness statements are often very valuable and have become the norm in most international arbitrations, a second round of witness statements is not always really necessary or worth the money. 7 Written rebuttal statements add to the costs of witness testimony and may prolong the proceedings by a month, if not more. [Page180:]

At the outset, it is generally possible to anticipate the necessary procedural steps and a reasonable time limit for achieving them, and therefore to achieve a workable schedule. While the flexibility of the timetable and the possibility of modifying it are important for ensuring that the parties have a proper opportunity to be heard, the arbitration often takes on a momentum of its own once it is under way.As a result, it is not uncommon for delays to build up and additional procedural steps to be scheduled, both of which have an impact on costs, all too often overlooked. For example, a request is made by one of the parties for extensive discovery. Rather than streamlining their submissions according to the key issues, the parties submit lengthy briefs in which they revisit every single issue in detail and accompany these at each round with several volumes of exhibits. Or the parties request extensions of time that jeopardize the hearing dates. Even the most seasoned arbitration counsel may have a tendency to submit more rather than less, on the assumption that they cannot take the risk of failing to deal with an issue (in some cases, even dealing with an issue more than once ). 8 Often, when faced with voluminous exhibits, arbitral tribunals will request parties to provide a common bundle of key documents, which at the end of the day may be, but are not necessarily, those truly relevant to the issues to be decided. 9 It is for the tribunal to ensure a balance between affording the parties a fair hearing in the face of unexpected procedural issues and keeping as close as possible to the objectives and time limits of the initial procedural timetable.

Tribunals may have to cope with differences in the parties' cultural expectations. At the closing stages, for instance, counsel from a legal system with a strong oral tradition may seek to plead his case at length by oral argument, whereas another counsel may rather seek to have posthearing briefs, without any closing oral presentations. In either case, parties should restrict themselves at this stage to synthesizing their respective cases. Astonishingly, they often see the final round as a further opportunity to plead at length, asking for closing hearings of a week or more or submitting posthearing briefs running to a hundred or more[Page181:]

pages. Parties-or rather their lawyers-would do well to step back from the 'battle' in order to better focus their efforts. A wellprepared tribunal can assist in this process by providing them with guidance as to the areas where further clarification or argument is required. Imposing or, even better, agreeing upon a page limit for posthearing briefs may force the parties to be focused and thus possibly save costs.

Article 18(4) does not mention a time limit for rendering the award after the close of the proceedings. 10This is to be regretted. Too many arbitrators nowadays take much longer to finalize and submit their awards than the parties had been led to expect. Delivery of the award may take up to nine months or even longer after the close of the hearings, often for no other reason than the arbitrators' busy schedules. 11

Arbitrators normally have nothing to fear-other than the frustration of the parties and their counsel at the final stage of the proceedings-if they do not submit their draft award to ICC within the time limit they indicated to the parties. 12 This is again about managing expectations and the parties' costs. If, for every week the arbitrators overrun the scheduled time limit for rendering the award, their fees were to be reduced by one per cent, it should be possible to shorten the time taken to render awards and certainly to make the truly excessive delays (of one year or more) exceptional. For a variety of reasons, the ICC Court would be unlikely to implement such a policy, 13 but why should not arbitrators themselves voluntarily agree to such an incentive to abide by their promise to render the award by a certain date? The many arbitrators who normally render their awards within the time limits indicated to the parties could set the example, as they would have no fee reduction to fear. [Page182:]

2. Streamlining the issues under the tribunal's guidance

Under Article 18(1)(d), it is no longer a requirement for arbitrators to include a 'list of issues to be determined' when drawing up the Terms of Reference. Where it appears from the parties' initial submissions that the issues to be decided are fairly straightforward, it may be appropriate to draft a list of issues, although this may be of little practical use if the issues are extremely simple. There is certainly no harm in doing so, as it will at least serve as a checklist for the Court when scrutinizing the final Award pursuant to Article 27.

As a case progresses, the dispute crystallizes and the arbitrators begin to see the issues more clearly. It can therefore be very helpful if, prior to and/or after the evidentiary hearing, the arbitrators let the parties know which areas of the dispute they are most concerned with. This guidance will assist parties in streamlining the presentation of their case. On many occasions, secondary and/ or irrelevant points are pursued by the parties with the same-if not more- energy than they give to the real issues to be decided. Pursuing points that the tribunal does not consider to be essential increases the overall cost-and especially each party's costs-to no benefit. If, however, in advance of an evidentiary hearing or prior to the submission of posthearing briefs, the tribunal were to point to the issues that it considers most relevant, parties may (but do not have to) take this as an encouragement to deal only-or mainly-with those points, 14 thereby saving time and money. This of course means that the tribunal would have to start its deliberations at a rather early stage in the proceedings. Were it to wait until it has heard the closing arguments and/or received the posthearing briefs, then such guidance would no longer be possible.

3. Proactive use of Article 31(2), second sentence

Under the 1998 ICC Rules of Arbitration arbitrators are expressly authorized to award costs, other than those fixed by the Court, at any time in the proceedings, i.e. before the final award (Article 31(2), second sentence). The costs involved are those mentioned in Article 31(1), that is to say party costs- 'the reasonable legal and other costs incurred by the parties for the arbitration'. Generally, in accordance with Article 31(3), which requires that the 'final Award[Page183:]

shall fix the costs of the arbitration and decide which of the parties shall bear them or in what proportion', once the Court has finally determined the costs of the arbitration, the arbitrators decide on the apportionment of both these procedural costs and the parties' costs when rendering the final award. The arbitrators' decision may fully or partly reflect the ultimate outcome of the case, unless each party is to bear its own costs.

In practice, it would seem that arbitral tribunals have so far made limited use of the discretion they have to decide upon party costs at an earlier stage of the proceedings, since decisions on such costs are routinely left to the final award. Interim decisions often say that the decision on costs is reserved, even though it may have been possible and appropriate for the tribunal to deal with the party costs at that stage.

It is submitted that tribunals should be more active in taking advantage, where appropriate, of the power they are given by Article 31(2), second sentence. Parties often incur significant legal costs in the course of an arbitration, and it is preferable and possible to deal with the costs of certain preliminary decisions- whether procedural or substantive-at the time they are made. The parties' costs consist for the most part of the legal fees of their counsel. These can be clearly identified, especially where lawyers bill their clients on an hourly basis and issue monthly invoices. The more applications counsel has to deal with in an arbitration, the higher will be the cost for the client. If clients know that by having their counsel pursue an unsuccessful application on their behalf they may have to bear not only their own costs in that connection but also those of their opponent (possibly with immediate effect, rather than later in the final Award as part of a global assessment), then they are likely to think twice about whether a defence should be raised and pursued.

The second sentence of Article 31(2) can therefore be a useful tool in managing and streamlining party costs, as it may discourage parties from pursuing too many applications and motions before the arbitral tribunal. To avoid taking the parties by surprise, the arbitral tribunal could include in the Terms of Reference an express reference to the powers it is granted under this provision and indicate that it will ask the parties to submit their costs as the case goes on. A 'payasyougo' approach will not always be appropriate, and will require a good sense of timing on the part of the tribunal. It will be necessary for the parties to submit their relevant costs before the tribunal renders its decision. Generally, since such costs consist mainly of legal fees and expenses, they can be readily established when required. [Page184:]

The tribunal will have to decide whether the amount claimed by a party is reasonable. This should be relatively easy as the costs concern a specific application or phase of the arbitration. The reasonableness should be determined from the perspective of the tribunal, which has to assess what was reasonably required of a party in order to make its case or present its defence. If a party attends a hearing with a team of five lawyers and makes a highly sophisticated PowerPoint presentation in defending, for instance, a request for the production of documents, 15 the tribunal, whilst appreciating the effort made, may consider that it was not warranted and that the associated costs, or part thereof, should not therefore be borne by the other side. Other factors, such as the procedural conduct of the parties, may also be taken into account when making a decision pursuant to the second sentence of Article 31(2). 16

Where a tribunal decides that it lacks jurisdiction over one of the respondents, the costs of that respondent (no longer a party to the arbitration) should be dealt with there and then rather than in a final award that may come several months or even years later. Where the tribunal confirms the locus standi of a party, the same would apply to the reasonable legal costs incurred in establishing locus standi, which are separate from and do not necessarily depend on the final outcome concerning the merits. Where an application is made for interim or conservatory measures, it would likewise be preferable for the tribunal to decide immediately on parties' costs connected with that application rather than reserve the decision on costs to the final award. When parties require a separate award on the law applicable to the merits, a decision on costs may in exceptional cases also be warranted. The basic justification for adopting such an approach in these instances is the need to sanction unsuccessful applications or requests that should not have been made.

While the Rules do not indicate what form early decisions on costs should take, the natural tendency is for the decision on costs to take the same form as the decision on the matter at issue. However, this might not be appropriate. If a party is to pay the costs immediately, then the tribunal's decision on costs will usually be presented as final, in which case it should take the form of an interim[Page185:]

or partial award and be subject to the scrutiny of the Court. If the tribunal wishes to reserve the right to reconsider all costs generally, including those relating to a specific application, then the decision will be of a preliminary nature and arguably should be set out in a procedural order so as to avoid the risk of being considered res judicata. However, this would be an invitation to reargue the issue and perhaps make the decision less likely to be enforced. The logical result is to prefer an interim or partial award, which would reinforce the 'payasyougo' principle and encourage both parties to limit their applications.

4. Documentary disclosure requests

Requests for the production of documents are becoming increasingly popular in international commercial arbitration, among parties from all quarters. They lead to numerous exchanges of letters, with counsel often hotly contending whether or not this or that document request is relevant and material or has been satisfied. As a result, tremendous costs are generated without any visible added value for the arbitral process. In the author's view, the more precise the request for document production, the lower the costs incurred, and the more likely it is that the document will be relevant to the issues to be decided by the tribunal. If both parties prefer a more general approach, then presumably the costs will be shared by both sides. However, where one party adopts a general approach to a request for production and the other adopts either a very focused approach or does not request production of documents at all, then the tribunal could address the issue of costs on two levels: first, whether the party succeeds in the documentary request and, second, whether the opposing party complies with the request for the production of documents and produces documents that prove to be material to the outcome of the case. If parties know that the discovery procedure may have a direct impact on costs in accordance with these two levels, then this may indeed encourage them to focus on the issues to be decided rather than secondary issues (although, as already noted, since the issues often crystallize only in the course of the proceedings, the gratuitousness of certain document requests may only become apparent later on). When assessing costs in the final award, the tribunal should establish whether the documents one party requested from the other were material for the outcome of one or more claims. 17 In other words, it must determine whether there was a correspondence between the effort of document production and the result obtained in the arbitration. This will be a difficult task, but should not be thought impossible. [Page186:]

It is extremely rare for parties to be asked to make separate submissions on the costs they have incurred in order to satisfy their opponent's requests for document production. The costs associated with some of the documentary production effort should be recoverable, not least because the time spent on this by management and inhouse counsel represents a significant cost in addition to the costs and expenses of outside lawyers. Generally speaking, ICC arbitrators are not sympathetic to the inclusion of management time in the parties' costs, as it is difficult to quantify, cost and allocate. However, in the context of requests for the production of documents, where managers at all levels are asked to search through their files, and/or make them available, the cost of the time devoted to that task should be recoverable, particularly if it can be identified and accounted for. Recovery of the costs incurred in document production should offer fair and adequate compensation for the time spent by a party's corporate officers on these tasks, and would also constitute an additional deterrent to the use of extensive discovery requests. However, as a caveat, it must be possible to provide proof of third party costs. One of the main problems with internal management costs is providing adequate proof as to (i) the time spent, (ii) that the time spent was related to the specific request and (iii) the real cost of the time spent. Therefore, if counsel wish to recover such costs, they should ensure that their clients maintain the type of documentary record on which a tribunal can comfortably rely for the purpose of awarding costs.

5. Assessment and apportionment of arbitration costs

In most ICC arbitrations, arbitrators apply the 'costs follow the event' rule for both the procedural costs and the party costs, including in particular attorneys' fees. 18Two issues regularly arise, but are often dealt with in a far too mechanical manner that may create unfair results.

(i) Assessment of the outcome of the case

If a party does not prevail with 100 per cent of its claims, it cannot usually expect to obtain 100 per cent of the costs. Thus, for example, if a party claims ten million dollars but is awarded only eight million, a tribunal may decide to award it only 80 per cent of the costs it has incurred. This is not necessarily the[Page187:]

right approach for either the procedural or party costs. As far as party costs are concerned, it may often be inappropriate to apply a percentage factor based on the outcome of the case. As was stated above, the vast majority of counsel engaged in ICC arbitration bill their services to their clients on a time basis. If one looks at how legal costs were incurred during the arbitration, one can quite easily identify those issues on which most time was spent. The basic principle should be not just that time was spent but that it was well spent. For example, the fact that a party wins on an issue is usually justification for awarding costs on that issue. However, the costs to be awarded should bear a relationship to the complexity of the issue and the time required to deal with it, as well as its ultimate relevance to the overall outcome of the case. In some cases, a distinction should be drawn between success on the claim and success on the measure of damages. If the issue of liability is hotly contested and a claimant succeeds, then it should be entitled to its costs with respect to the liability issue. In some cases, the quantum issues take up much less time than those relating to liability. In those cases, therefore, limiting the costs recovered to the percentage of the damages awarded may be inappropriate.

In fixing the arbitrators' fees, the ICC Court always takes into account the time spent. If the arbitrators have had to deal with preliminary issues, such as jurisdiction or the applicable law, and possibly held hearings on such issues and rendered partial awards, they will have spent proportionally more time on the case. The claimant's success on these issues can and often should be taken into account and, in the above example, the tribunal may therefore be justified in going above-or below-the 80 per cent award of costs. If one assumes in the above example that the chief focus of the proceedings was the claimant's first claim valued at eight million dollars, while the remaining twomillion dollar claim concerned, for instance, consequent loss of reputation, which was unsubstantiated from the outset and on which neither party, let alone the tribunal, spent any time, then the tribunal's decision to award the claimant up to 100 per cent of the procedural and party costs might indeed be justified. In other words, the more time needed to pursue a claim successfully, the more justifiable it would be to give corresponding weight to that success when deciding on costs. However, the tribunal will need to explain its reasoning more than ever, so as not to leave the parties guessing.

(ii) Division of costs

If a party is successful with up to sixty per cent of its claims, many arbitrators are inclined to award not only sixty per cent of the procedural costs fixed by the ICC Court, but also of the winning party's legal costs. What in that case[Page188:] should happen with the legal costs of the losing party, whose defence was successful to the extent of 40 per cent? In most cases, it would seem unfair not to take into account the loser's legal costs in the same proportion. There are at least two ways of dealing with this situation.

One is for the tribunal to add up the legal costs incurred by both parties and then apply a percentage factor to them. If a rate of 60/40 is applied and the legal costs of each party amounted to 500,000 dollars, making a total of one million dollars, then the claimant would be awarded 200,000 dollars net (600,000 less 400,000) and the respondent nothing. 19 In the author's view, this approach is only appropriate when the legal costs of both sides are more or less identical, which they sometimes are. In such cases, the tribunal will often accept the amount of costs claimed by both parties as reasonable.

The other-preferable-approach is to apply the success rate percentage to the legal costs of each side. First, the tribunal will have to determine whether it considers the costs incurred by each side as reasonable. Then, in the above example, it will have to allow the claimant to recover 60% of its costs and the respondent 40% of its costs. 20 In light of the great discretion arbitrators enjoy when it comes to the apportionment of costs, adjustments of the type mentioned above can be made in appropriate circumstances. There would normally have to be some very good reasons why a respondent that is successful in challenging a good part of the claimant's claims should be left with 100 per cent of its own costs. 21 However, once again, it may be appropriate to distinguish between various types of claims (and the time spent on them) as well as the results as to quantum. [Page189:]

A claimant that succeeds in its primary claims, which took up much of the time and effort of the arbitration, may be entitled to recover a substantial portion of its costs, even if it fails on a number of secondary or ancillary claims. Similarly, if a claimant succeeds with its major liability claim and is awarded a significant amount of the damages sought under that claim, then it is reasonable to conclude that the claimant was in essence the successful party and is entitled to be treated as such. No rigid rule can satisfactorily solve the problem of fairly apportioning costs, and in some cases it may indeed be most appropriate to have each party bear its own party costs, or to have one party bear the entirety of the arbitration costs.

Concluding remarks

Decisions on costs other than the fixing of procedural costs (which is done by the Court on the basis of the scale of costs appended to the ICC Rules) are reserved for the arbitral tribunal, according to the Rules. The discretion that the Rules give to the arbitral tribunal in this regard, if well used, may help prevent arbitration costs from further spiralling to limits beyond which parties may feel they are no longer getting value for money. As always, finding appropriate solutions is much harder than raising critical questions, yet it is hoped that the above considerations may point towards ways in which improvements may be sought. [Page190:]

1
In an ICC arbitration, costs comprise: (i) procedural costs consisting of the ICC administrative expenses and the arbitrators' fees and expenses, and (ii) party costs consisting of the legal fees and expenses of counsel, costs for witnesses, and hearing and ancillary expenses. See the present author's article, 'Costs in ICC Arbitration: A Practitioner's View' (1992) 3 The American Review of International Arbitration 116, and, more recently, the article by his Swiss namesake Micha Bühler, 'Awarding Costs in International Commercial Arbitration: an Overview' (2004) 22 ASA Bulletin 249.

2
S.R. Bond, 'Counsel's responsibilities in regard to the speed of an international arbitration' in Improving International Arbitration: The need for speed and trust, Liber Amicorum Michael Gaudet (ICC Publishing, 1998) 113 at 115ff.; Y. Derains, 'La responsabilité des conseils', ibid., 119.

3
On the somewhat controversial subject of who controls the arbitration, see e.g. G. KaufmannKohler, 'Qui contrôle l'arbitrage? Autonomie des parties, pouvoir des arbitres et principe d'efficacité' in Liber Amicorum Claude Reymond : Autour de l'arbitrage (Litec, 2004) 153.

4
The process whereby all draft awards are submitted to the ICC International Court of Arbitration for scrutiny before being approved allows formal deficiencies in relation to costs to be remedied. For instance, if no reason has been given by the tribunal for the way the costs have been apportioned, the Court may, as a matter of form, require that reasons be added. Given the broad discretion arbitrators enjoy with respect to costs, the Court very rarely draws the tribunal's attention substantive points relating to costs.

5
Having been a member of the ICC Court since the inception of Robert Briner's first term in 1997, the author has had the opportunity to read hundreds of draft awards submitted to the Court for scrutiny and approval pursuant to Article 27 of the ICC Rules. In 2004, the Court approved a total of 345 awards, including 31 awards by consent.

6
References to Articles are those of the Rules, unless otherwise indicated.

7
In many instances, a witness can make rebuttal comments on direct examination within a reasonable time.

8
Maybe rightly so, as tribunals sometimes take a rather liberal approach to the burden of proof where the respondent's denial is insufficient to have the claimant's claim rejected for lack of proof.

9
This is not to say that common hearing bundles do not have problems of their own, in particular as they are often costly to prepare (especially when, unbeknown to the tribunal, lengthy exchanges between counsel take place to agree on what documents should or should not be included), and in some respects a waste of the parties' resources. Computerliterate arbitrators who deal with computerfriendly counsel may avoid the problem by having all exhibits scanned and put on a CDRom or DVD.

10
The general sixmonth time limit for rendering the award referred to in Article 24(1) of the Rules has become rather meaningless in practice.

11
It is a mistake to suggest that if parties want to have the best arbitrators, delays are among the problems they have to accept. Delays can happen with arbitrators of any rank, for very few arbitrators are not busy otherwise.

12
The Secretariat of the Court rarely goes beyond soft monitoring of the tribunal's progress. A tighter control, with weekly followups by phone and the involvement of the Secretary General in person, might have a greater impact than sending the usual standard letters. See also J. Werner, 'Who controls speed? A few reflections on the relationship between parties and arbitrators in ICC arbitration' in Improving International Arbitration, supra note 2, 99 at 103.

13
The Court has, however, recently started to monitor the time lapse between the closing of the proceedings and the submission of the draft award to the Secretariat. When fixing the arbitrators' fees, the Court will therefore know the time taken to draw up the Terms of Reference, the time spent between the Terms of Reference and the close of the proceedings, and the time from the close of the proceedings to the draft award.

14
On the adjustment of the issues in dispute, see e.g. T. Webster, 'Terms of Reference and French Annulment Proceedings' (2003) 20 J. Int. Arb. 561 at 564-65.

15
PowerPoint presentations can be extremely useful, but are not needed in each and every case, and for every hearing. Stateoftheart PowerPoint presentations are costly to prepare, and parties should be aware that they may not necessarily recover these costs.

16
For other factors that may generally play a role in assessing costs, see M.W. Bühler & S. Jarvin, 'The Arbitration Rules of the International Chamber of Commerce' in F.B. Weigand, ed., Practitioner's Handbook on International Arbitration (Munich: Beck, 2002) Part 2, Art. 31, paras. 70-71.

17
See Article 4 of the 1999 IBA Rules on the Taking of Evidence in International Commercial Arbitration

18
Parties from legal systems where the normal rule is that each party bears its own attorney's fees should be aware that attorney's fees may be-and are regularly-awarded in ICC arbitrations. See M.W. Bühler & H. Conybeare Williams, 'New York Law: Awards of Attorney's Fees in International Arbitration' (2005) 20:5 Mealey's International Arbitration Report 36.

19
As opposed to 300,000 dollars, if the respondent's success in defending the claim were not taken into account.

20
Let us assume that the claimant incurred party costs amounting to 800,000 dollars and the respondent party costs amounting to 1,000,000 dollars, this would give the claimant a claim for 60 per cent of its costs (480,000 dollars), leaving it to bear 40 per cent of its costs, and the respondent a claim for 40 per cent of its costs (400,000 dollars), leaving it to bear 60 per cent of its costs. The net recovery of party costs for the claimant would amount to 80,000 dollars. If the costs were the other way round, the following result would be reached: (1,000,000 x 60% = 600,000) less (800,000 x 40% = 320,000) = 280,000, or 28 per cent of the claimant's claim for party costs. While in both cases the respondent would have to bear 100 per cent of its own costs, it would be charged with a much lower net amount for the claimant's party costs than if the claimant is awarded 60 per cent of its party costs.

21
This has nonetheless happened in the past more than once. For instance, in a recent case, two of the claimant's five claims were fully dismissed by the tribunal on the merits and a third claim partially dismissed. The claimant ended up being awarded less than 40 per cent of its total claims, with interest from the date of the award rather than the breach, which occurred eight years earlier. Without providing any reason, the tribunal awarded the claimant 60 per cent of the procedural costs and 50 per cent of its party costs, leaving the respondent with 100 per cent of its own party costs.