Beaufort Developments (NI) Limited v Gilbert-Ash (NI) Limited and Others: The Demise of Crouch

By Julian Critchlow

Arbitration, in the sense of a private method of dispute resolution, almost certainly pre-dates state administered processes in most cultures. In English construction law its use continues unabated and all the major standard forms of construction contract contain arbitration clauses. That enthusiasm appears not to have been dampened with the coming into force of the mandatory adjudication provisions of the Housing Grants, Construction and Regeneration Act 1996; at least, Amendment 18 to JCT 80, the amended DOM1 conditions, and the amended ICE 6th conditions retain arbitration in their compliant dispute resolution provisions; and the conventional view is that arbitration can deliver a swifter and cheaper resolution of construction disputes, than is generally available in the High Court with all the benefits of privacy and an expert tribunal selected by the parties themselves.

That view has been much challenged. Ian Duncan Wallace has written that "Compulsory arbitration in the full sense.... would not only be disastrously unjust, given the quality of the general class of arbitrators available, but in the most important sense of denial of justice and of the Queen's courts to the litigants in this field [of construction], which must be their right, as that of any other citizen." Whilst there is currently no scheme of compulsory arbitration in existence, other than the limited scheme in the County Courts for claims under £3,000, effective arbitration clauses are not always genuinely consensual because the parties do not necessarily advert to their existence when they agree to a standard form which incorporates such a clause. Moreover they may find themselves subject to arbitration clauses in inconvenient circumstances such as where a multi-party dispute arises and the various parties are faced with the prospect of contending the same issues over and again in different arbitrations. Other criticisms have related to the exercise time and cost of some references. However, whatever the advantages and disadvantages, the decision in Northern Regional Health Authority v Derek Crouch Construction Co. Limited [1984] QB 644; [1984] 2 WLR 676; [1984] 2 All ER 175, C.A. introduced a compelling reason for including arbitration clauses in construction contracts.

The significance to the Crouch case lay in the decision of the Court of Appeal that the Court did not have the same power as an arbitrator under a standard JCT form of contract to "open up, review and revise" architects' certificates. The Court decided that exercise of this function amounted to a modification of the parties' contractual rights rather than a mere determination of what those rights might be: the power could only be conferred on the arbitrator by agreement between the parties and it exceeded the Court's own inherent jurisdiction so the Court could not grant equivalent relief. Browne-Wilkinson LJ said that:

In principle, in an action based on contract the Court can only enforce the agreement between the parties: it has no power to modify that agreement in any way. Therefore, if the parties have agreed on a specified machinery for establishing their obligations, the Court cannot substitute a different machinery.

He went on to say that where an architect's certificate was challenged "the Court's jurisdiction would be limited to deciding whether or not the certificate or opinion was legally invalid because it was given, for example, in bad faith or in excess of his powers. In no circumstances would the Court have power to revise such a certificate or opinion solely on the ground that the Court would have reached a different conclusion, since to do so would interfere with the agreement of the parties." What was probably a significant contributor to the rationale for the decision appears from a remark of the then Master of the Rolls, Lord Donaldson, that the judgment in Crouch might reduce the length of the Official Referee's lists as it would confine more disputes to arbitration with the consequence that there would be less recourse to the Court in construction disputes.

The decision in Crouch as set out above was obiter, i.e. not essential to the main reasoning of the judgment and not binding on other courts. Nevertheless, it had considerable persuasive value in subsequent cases. Some judgments adopted the Crouch reasoning. Others sought to distinguish it. Adopting Crouch, in Finnegan v Sheffield City Council (1988) 43 BLR 124 it was decided (again probably obiter) that the Crouch principle applied even where there was no arbitration clause. The consequence of having no arbitration clause was potentially to make architects' decisions unchallengeable except where the certifying machinery had broken down: in the absence of such a clause there was no arbitrator with express power to overturn certificates, and that power was also outside the inherent jurisdiction of the Court. A similar view was expressed in Oram Builders v Pemberton (1985) 29 BLR, and a parallel analysis was adopted in Balfour Beatty v Docklands Light Railway Limited (1996) CILL 1143 where the employer's representative was designated contract certifier and the contract contained no arbitration clause.

Distinguishing Crouch, it was held in Partington & Son v Thameside (1985) 33 BLR 150 that Crouch was obiter and non-binding. Less overtly, in John Barker Construction Limited v London Portman Hotel Limited the effects of Crouch were modified by the Court's adoption of a liberal meaning to the concept of breakdown of the certifying machinery, it being accepted in Crouch itself that the Court could decide the parties' rights where such a breakdown occurred. Whilst accepting that the Court could not substitute its own decision for that of the architect on the sole ground that it would have reached a different decision, it was held in John Barker that the architect's making of an impressionistic assessment of extensions of time instead of undertaking a logical analysis was sufficient to constitute a breakdown of the certifying machinery, and the Court could substitute its own machinery.

An attempt was made by the legislature to curtail the influence of Crouch by the enactment of Section 100 of the Courts and Legal Services Act 1990 which inserted Section 43A into the Supreme Court Act 1981 and which enables the Court to exercise the same powers as an arbitrator under the parties' arbitration clause where the parties agree that it may do so. In fact, since that provision requires the existence of an underlying arbitration clause, and since, when a dispute arises, it is in any event unlikely that all parties will consent to extending the Court's jurisdiction, its influence has been limited.

However, the Crouch issue appears now to have been finally resolved by Beaufort Developments (NI) Limited v Gilbert-Ash (NI) Limited and Others, a House of Lords judgment handed down 20th May 1998. The House of Lords unequivocally and unanimously decided that Crouch was wrong. The Court held that there was, indeed, binding authority contrary to Crouch being Robins v Goddard [1905] 1 KB 294, and the Court of Appeal in Beaufort should have followed it. Lord Hoffman's judgment was forthright. He stated:

In my opinion, therefore, the dicta on this point in [Crouch] were both obiter and wrong.... it seems to me that cases since Crouch show that the decision has caused such uncertainty and even injustice that its dicta should be disapproved.

The essence of the judgment in the Beaufort is as follows. No private arbitrator has any authority to determine a dispute other than as conferred on him by the parties or by statute. However, an arbitrator appointed under a broadly drafted arbitration clause which enabled him to decide all disputes between the parties relating to the contract in question would be impliedly entitled to open up, review and revise architects' certificates. Therefore, the express power given to the arbitrator to open up review and revise was simply the draftsman's "desire to be certain that every conceivable point has been covered." Exercising that power did not amount to a modification of the parties' rights under the contract, as was held in Crouch, but was a determination as to what the parties rights were. The Court possessed that same power as part of its inherent jurisdiction. The arbitration agreement was neither giving the arbitrator a power that it did not possess, nor impliedly denying the Court its exercise of an existing power.

It should be noted that the judgment predicated that the architect's certificate that was sought to be challenged was interim rather than final, i.e. it had what Lord Hoffman called "provisional validity". The judgment made it clear that if the certificate were expressed in the contract to be final and binding, a different regime would apply. In such a case, a certificate could only be challenged to the extent expressly provided by the contract. Such a power might be contained in the arbitration agreement itself and exercisable by the arbitrator. But the Court would not have inherent jurisdiction to do so. There was, however, no suggestion in this case that the architect's certificates were anything other than interim. Therefore, the Court was able to exercise its inherent power to examine and determine the validity of the architect's certificates. In doing so it was merely deciding the parties' legal rights.

A noteworthy ancillary view expressed by Lord Hoffman was that making architects' interim certificates conclusive "could easily cause injustice", partly because the architect might have to give his decision when the knowledge of the state of the work or the effect of external causes is incomplete, but also because although he is "a professional man [he] can hardly be called independent". Accordingly, he considered that "one should require very clear words before construing a contract as giving an architect such a power."

The Beaufort decision has less impact now than it would have had if it had been made shortly after Crouch itself. Partly this is because, as observed earlier, Section 43A of the Supreme Court Act 1981 has allowed the parties to agree that the Court should have the same powers as an arbitrator although, as also noted, such consent has rarely been forthcoming in practice. Additionally, there is the effect of the Arbitration Act 1996. Section 9 of that Act, linked with the decision not to proceed with Section 86, makes it exceedingly difficult to bring proceedings in Court, in the absence of consent, where a valid arbitration agreement covers the dispute. The Court will be bound to grant any application to stay to arbitration and will not, therefore, have the opportunity to review certificates itself. Moreover, any appeal from the arbitrator's award will be confined to important points of law - it will not be possible to open up the facts underlying the certificate.

Nevertheless, Beaufort is important. In the first place, Crouch has long been regarded by many as wrongly decided and to reflect a manipulation of principle by the Court with a view to reducing the burden of the lists. It was always difficult to see how the arbitrator, in determining the accuracy of a certificate, was modifying the parties' rights under the contract. The parties' rights were to have a certificate issued in accordance with the underlying contract terms. If a party were dissatisfied with the certifier's decision this plainly gave rise to a dispute as to the parties' legal rights. Indeed, if it did not, it could not, in truth, be challenged before an arbitrator (properly so-called) any more than before the Court, because a dispute as to legal rights is a fundamental requirement, in English law, of an arbitration's coming into being.

Furthermore, not only will Beaufort reduce injustice of the sort observed in some of the post-Crouch cases, but it will enable the parties to construction contracts consciously to reject arbitration as the forum for dispute resolution right from when they formulate their contract and to invoke the Court's jurisdiction instead. Accordingly, draftsmen will doubtless keenly compare the rival merits of judge and arbitrator, and arbitration will need to demonstrate its superiority if it is to be the preferred choice.

Back to the previous page