The arbritration files

The use of commercial arbitration can be traced back far beyond the beginnings of the court system, and up until 1998 (when the Housing Grants, Construction and Regeneration Act 1996 came into force) it was the most commonly used private method of dispute resolution. Yet despite this, there remains little research into this grand old father of dispute resolution.

My recent study has been undertaken with feedback from 36 arbitrators sitting on the CIArb, RICS, ICE, RIBA, IChemE and the Law Society panels, and the collection of statistical information from approximately 100 construction and civil engineering arbitrations across the UK (most of which ended during the last few years). The combined experience of those who assisted with the research amounted to approximately 2,450 construction and civil engineering arbitrations, an average of 68 per arbitrator.

Protagonists

Given that it is main contractors who enter into the highest number of contracts and thereby are most at risk of experiencing a dispute, it is not surprising to find that main contractors appeared as either claimant or defendant in 72% of the arbitrations. Employers were involved in 54% and sub-contractors in just under a quarter.

Interestingly, consultants only appeared in 3% of the arbitrations. Other parties involved in these arbitrations included various statutory bodies, water authorities, a coal authority and a rail company. Also identified were developers, building owners, tenants and the National House-Building Council.

Furthermore, the claimant was often lower in the contractual chain than the defendant and the most likely protagonist combination, representing 36% of arbitrations, was the main contractor as claimant versus the employer as defendant.

Just 14% of arbitrations involved the employer as claimant against a main contractor. The second most likely combination of protagonists, taking part in 17% of arbitrations, was a claimant sub-contractor versus a defendant main contractor. Only 2% of arbitrations involved the main contractor as claimant against a sub-contractor.

Causes

Main contractors appeared as claimant or defendant in 72% of cases

The study sought to classify the nature of the dispute under one or more of the following causes: variations; extensions of time; loss and/or expense; workmanship; design; or other.

Over half of the cases could be attributed to more than one of these categories. While 44% involved a single category, 22% consisted of two categories and 24% concerned three. The remaining 10% dealt with between four and all six of the above.

Variation issues were most prone to dispute (54% of the cases). Extensions of time and loss and/or expense issues generated similar incidences of 36% and 34% respectively. Not surprisingly, disputes concerning extensions of time and loss and/or expense tended to occur together in the same case. Workmanship issues, which appeared in 31% of arbitrations, were twice as likely to occur as design issues (16% of cases).

Arbitration length

While the average duration of the arbitrations was found to be just over 14 months, this fluctuated from as short as one month up to a staggering nine years. However, the second longest duration was only four years and four months.

The nine year case involved a £1m claim. The nature of the dispute related to both variations and loss and expense. The arbitrator said it was extremely hard-fought with barristers on both sides. The claim was put under a number of headings, which were heard and decided separately. The duration was protracted because each award was appealed in the High Court, which dismissed all but one that was referred back to the arbitrator regarding the rate used for an interest calculation. The arbitrator eventually awarded £493,000 to the claimant.

It was surprising to find, however, that the existence of a counterclaim had little impact upon the duration of the cases researched, lasting only 0.6 months longer on average. Of those arbitrations resolved by an arbitrator's award, the average duration was 15.6 months, with the average duration for those settled prior to award at 10.4 months.

Counterclaims had little impact upon the length of cases

Arbitrations involving larger claims tend to last longer. Those with claims worth over £500,000 were associated with an average length of 27 months. Those with a claim size of up to £50,000 had an average length of 9.6 months - a mere third of the average length of the largest claim sizes. Those with a claim size of between £50,000 and £500,000 generated an average length of 11.2 months.

Claim/counterclaim sizes

Four out of five cases related to a claim size of less than £1m, while just 2.3% involved claims over £10m. One third of the claims did not exceed £100,000 and a further 29.9% fell between £100,000 and £250,000 (see factfile).

Whenever variations were involved as a single dispute, the average claim size was a staggering 10 times the size of those concerning workmanship. Although variations issues cost more than workmanship issues, the proportion of costs awarded to the claim size was over two and a half times as high for disputes regarding workmanship than for variations. This suggests that in relation to claim size, workmanship disputes give rise to greater costs than those relating to variations.

Concerning counterclaims, over 60% of arbitrations did not include a counterclaim. But of those that did, 79.5% related to a counterclaim size of less than £1m and 65.8% involved a counterclaim size that was worth less than £250,000 (see table below). These proportions are similar to those found for the claim sizes detailed in the first table. While the largest bracket of claim size encountered was for between £100,000 and £250,000, the largest bracket of counterclaim size was between £10,000 and £50,000, in which 28.9% of the counterclaims fell.

Tribunal costs/expenses

Under the Arbitration Act 1996, the arbitration panel, which usually comprises just one arbitrator, is referred to as the ‘tribunal'. The arbitrators did not supply details of their costs and expenses for all of the arbitrations analysed, but where they did it can be seen from the table below that arbitrations with a claim size of no more than £50,000 involved a mean charge from the arbitrator of £3,261. The three bands of claim sizes between £50,000 and £500,000 generated mean tribunal charges of between £9,873 and £13,932 - roughly three and a half times as much as the £3,261 for the smallest band of claims. The mean tribunal charge was subject to a further increase for claims over £500,000, which involved a mean charge of £52,942.

Case lengths ranged from one month to a staggering nine years

Amount claimed for costs

The table below shows that the average level of costs claimed by the successful party was £54,129, but this varied significantly depending on the size of claim.

Arbitrations with claim values up to £50,000 generated an average costs claim of £7,420, and those worth between £50,000 and £100,000 were not significantly different, at £9,000.

A substantial difference was identified in those arbitrations involving claims of between £100,000 and £250,000 generating an average claim for costs of approximately £62,600. The upward trend continued for arbitrations with a claim value between £250,000 and £500,000 with average cost claims of £76,500. Arbitrations for a claim in excess of £500,000 created an average cost claim of £115,333.

Amount awarded for costs

The lowest proportion of the successful party's claim for costs awarded by the arbitrator was 50% and on three occasions the amount awarded for costs was 100% of that claimed. Looking at the table below, the average amount awarded for costs was approximately 76.9% of that claimed. The highest average encountered, 81.2%, was within the smallest band of claim size of up to £50,000.

Settlement prior to award

I don’t think arbitration has recovered from the adjudication onslaught. Most people are seeing the benefits of adjudication

David Blake, founder of Blake Newport

Industry observers suggest that a significant number of disputes referred to arbitration are settled prior to the final award. This is a fair comment: 42% of all of the 2,450 arbitrations were settled prior to the final award.

However, individual arbitrators reported widely divergent settlement percentages. Three arbitrators reported a 0% settlement rate, while two others reported 100%.

Even where only the settlement rates reported by the more experienced arbitrators are considered, such ranged between 13% and 87%. No obvious reason was evident to explain the differing experiences of those who took part.

Size of award

For those arbitrations that are not settled early, the average proportion of the amount finally awarded compared to the size of the claim was 54.7%. In four cases the arbitrator awarded the whole claim. While three of the four involved claim sizes of less than £20,000, the other was £2.9m. The table below examines the mean percentage awarded for various categories of claim size.

Who receives the award?

The arbitrator's award will order one party to recompense the other and in this respect the successful party is referred to as having ‘received the award'. Of the arbitrations that reached final award, the claimant received the arbitrator's award in 60% of instances, whereas in only 12.5% of cases did the defendant receive the award, with the remaining 27.5% involving both parties receiving part of the award.

In those cases where the main contractor was the claimant, they were less successful against the employer than a sub-contractor claimant against the main contractor defendant. The main contractor claimants received the entire award in only 54% of arbitrations against employer defendants and were unsuccessful in just under a quarter of cases. In contrast, sub-contractor claimants received the entire award in their arbitrations against main contractors in 71% of cases and the remaining 29% of awards were shared, rather than lost.

However, this is not entirely surprising. A sub-contractor's claim will contain fewer and simpler issues for resolution than that of a main contractor's inevitably more complex claim against an employer respondent. The sub-contractor will therefore be far more likely to succeed.