As of November 2016, the United Kingdom has just under 7,000 wind turbines. They generate 14.1 gigawatts of electricity, contributing about 11% to the country’s electricity requirements. Of that, 5.1 gigawatts are generated by offshore windfarms. Offshore windfarms also generate litigation and arbitration. In particular, two developments (Robin Rigg and Greater Gabbard) have led to decisions by the English Courts clarifying the scope of a contractor’s obligation to ensure that works are ‘fit for purpose’. We explore these below.
Fitness for purpose - the background
A contractor’s obligation to complete the works can be expressed in a number of ways, ranging from requiring the contractor to act with ‘reasonable skill care’, to complying with ‘good industry practice’ (or even ‘best practice’) all the way to warranting fitness for purpose. Fitness for purpose is seen as the gold standard. If the works prove unfit for their purpose, a contractor might still be liable even though it exercised all reasonable skill and care, complied with the specifications, and followed the applicable industry standards.
The FIDIC Silver Book (“Conditions of Contract for Engineering Procurement and Construction/Turnkey Projects”) contains an express fitness for purpose obligation in Clause 4.1:
“When completed, the Works shall be fit for the purposes for which the Works are intended as defined in the Contract.”
Under English law, quite apart from such an express term, a warranty of fitness for purpose may also be implied where the contractor is carrying out the design. In IBA v EMI Electronics and BICC Construction Ltd (14 BLR 9), the House of Lords confirmed that a warranty of fitness for purpose falls to be implied where the contractor is responsible for the design and holds itself out as having particular skill or expertise - which the employer relies on. The case concerned a television mast that was over 350m tall and of cylindrical design. At the time, it was the highest mast of this design that had been built anywhere in the world. It collapsed after 3 years, as a consequence of ‘vortex shedding’ caused by winds and asymmetrical ice loading. The contractor argued that it should not be held to having warranted fitness for purpose as the design was complex and pushed the boundaries of how such structures could be built. Their Lordships nonetheless held that:
“… in the absence of a clear, contractual indication to the contrary, I see no reason why one who in the course of his business contracts to design, supply, and erect a television aerial mast is not under an obligation to ensure that it is reasonably fit for the purpose for which he knows it is intended to be used. … The critical question of fact is whether he for whom the mast was designed relied upon the skill of the supplier (i.e. his or his sub-contractor's skill) to design and supply a mast fit for the known purpose for which it was required”.
This implied warranty requires that the works be ‘reasonably’ fit for their purpose. In IBA v EMI, the Court of Appeal clarified that the mast had to be able to withstand meteorological conditions that were reasonably likely to occur in the relevant location - so entirely unforeseeable natural disasters such as earthquakes or perhaps a hurricane would not be covered by the warranty. In contrast, FIDIC Clause 4.1 seeks to impose an absolute obligation and makes no reference to the works only being ‘reasonably’ fit for purpose. This is potentially a very onerous provision, although much will depend on how clearly the ‘purpose’ is defined, as is illustrated by the Robin Rigg decision.
Robin Rigg - Overlapping obligations
The Court of Appeal’s judgment in MT Højgaard A/S v E.On Climate and Renewables UK Robin Rigg East Ltd  EWCA Civ 407 demonstrates how an obligation of fitness for purpose can be watered down because the purpose is not expressly defined or stated in any one place in the contract, in circumstances where the contract also imposed ‘reasonable skill and care’ obligations on the contractor alongside ‘fitness for purpose’.
In December 2006, E.On engaged MT Højgaard to design, build and install foundations for 60 offshore wind turbines, part of the ‘Robin Rigg’ development in the Solway Firth in Scotland. Steel monopiles driven into the seabed were to be used as foundations for the turbines. The towers of each turbine were then to be connected to the top of the monopile. That connection involved a steel cylinder, called a transition piece, which is fitted over the top of the pile. The gap between the transition piece and the pile was to be filled with grout. This ‘grouted connection’ anchors the transition piece because of friction between the grout and the steel surfaces either side of it. The turbine tower then slots into the transition piece. Grouted connections can be supplemented with welded ‘shear keys’, which penetrate into the grout and are meant to strengthen the connection. However, shear keys also introduce the risk of stresses near the area of penetration weakening the grout, potentially causing the connection to fail.
MT Højgaard’s scope of work included the design of the grouted connection. MT Højgaard’s design did not feature shear keys. That was because it had carried out calculations set out in a well-known international standard for the design of offshore wind turbines, DNV-OS-J101 (“J101”): the standard said that these calculations specifically aimed at testing the strength of grouted connections. The results of MT Højgaard’s calculations showed that MT Højgaard’s grouted connections would be strong enough without shear keys.
J101 has been published by Det Norske Veritas (“DNV”), and is widely applied and relied on in the offshore construction industry. J101 sets out a design approach that is aimed at achieving a sufficiently high likelihood of the structure in question remaining stable over the course of its service life, whilst recognising that it is impossible to completely eliminate all risk of failure. In relation to the design of steel structures (such as monopiles / transition pieces), J101 states that:
“The design fatigue life for structural components should be based on the specified service life of the structure. If a service life is not specified, 20 years should be used.”
MT Højgaard completed the works in February 2009. A problem then arose at another offshore windfarm in the Netherlands, where grouted connections (also without shear keys) started to fail. This was serious, as the transition pieces started to slide down the monopiles. Later on in 2009, DNV carried out an internal review of J101. It found that the equation proposed for testing the strength of grouted connection contained an error. The value given by DNV for one of the elements in the equation had been overstated by a factor of 10. This meant that the grouted connections at the Robin Rigg windfarm were much less robust than MT Højgaard had assumed. In 2010, grouted connections at Robin Rigg also started to fail, and the transition pieces started to slide.
E.On and MT Højgaard set about repairing the foundations, at a total cost of €26.5 million. A dispute then arose as to who was liable for these costs. MT Højgaard applied to the High Court for declarations that it was not liable, arguing that it had exercised all reasonable skill and care, and had relied on and followed J101. It said that it was not to blame for the error in J101, which no reasonable contractor could have discovered.
The contractor’s obligations - a ‘kitchen sink’ approach
The contract contained a panoply of provisions as regards the standard to be expected of the contractor. The conditions of contract required MT Højgaard to use all reasonable skill and care, but also stated that the works had to be ‘fit for purpose’:
“8.1 GENERAL OBLIGATIONS
The Contractor shall, in accordance with this Agreement, design, manufacture, test, deliver and install and complete the Works:
(i) with due care and diligence expected of appropriately qualified and experienced designers, engineers and constructors (as the case may be)…
(iv) in a professional manner… in accordance with… Good Industry Practice…
(x) so that each item of Plant and the Works as a whole shall be fit for its purpose as determined in accordance with the Specification using Good Industry Practice …”
Good Industry Practice was defined as:
“… those standards, practices, methods and procedures conforming to all Legal Requirements to be performed with the exercise of skill, diligence, prudence and foresight that can ordinarily and reasonably be expected from a fully skilled contractor who is engaged in a similar type of undertaking or task in similar circumstances in a manner consistent with recognised international standards.”
The contract also contained a definition of ‘fitness for purpose:
“... fitness for purpose in accordance with, and as can properly be inferred from, the Employer's Requirements”.
E.ON’s employer’s requirements and the various documents that made up the ‘Specification’ expressly required the contractor to prepare the design in accordance with J101. Importantly, the employer’s requirements also contained statements suggesting that MT Højgaard was warranting that the foundations would have a 20-year life span:
“The design of the foundations shall ensure a lifetime of 20 years in every aspect without planned replacement ...”
“The design of the structures addressed by this Design Basis shall ensure a lifetime of 20 years in every aspect without planned replacement. The choice of structure, materials, corrosion protection system operation and inspection programme shall be made accordingly.”
“The Works elements shall be designed for a minimum site specific 'design life' of twenty (20) years without major retrofits or refurbishments; all elements shall be designed to operate safely and reliably in the environmental conditions that exist on the site for at least this lifetime.”
The reference to ‘design life’ in the last of the passages from the employer’s requirements quoted above echoed the terminology used in J101. In contrast, the other two passages referred to a ‘lifetime’ of 20 years, not a ‘design life’ - a difference that would prove to be material.
E.ON’s case against MT Højgaard primarily rested on the first two of the provisions set out above (which referred to ‘lifetime’). E.ON argued that it did not matter that MT Højgaard had used all reasonable skill and care, and had followed J101: MT Højgaard had warranted that the foundations would in fact be good for 20 years. The judge at first instance agreed. He found that MT Højgaard had assumed an absolute obligation, reading Clause 8 of the contract together with the passages from the employer’s requirements that spoke of achieving a 20 year ‘lifetime’ in mandatory terms.
MT Højgaard appealed. Jackson LJ in the Court of Appeal noted that the task of the Court was to distil the extent of the obligations that the contractor had undertaken from contractual documents that contained a considerable amount of loose wording, and featured multiple apparently overlapping provisions (suggesting that ‘less’ may very well be ‘more’).
Do absolute obligations trump reasonable skill and care?
E.ON argued that an absolute obligation, such as a warranty of ‘fitness for purpose’, should take precedence over an obligation of reasonable skill and care, where both appeared in a contract. The Court of Appeal was referred to passages in Hudson's Building and Engineering Contracts which stated that where a contractor was required to carry out works in accordance with plans or specifications, whilst also promising that the works when completed would be able to perform a certain duty (or ‘be fit for purpose’), then the contractor would be liable if the works proved incapable of performing that duty, even though the plans or specifications had been complied with.
While there are no English cases in support of that proposition, the Court of Appeal was presented with Commonwealth authority on the point, including the decision of the Court of Appeal for British Columbia in Greater Vancouver Water District v North American Pipe & Steel Ltd and Moody International Ltd  BCCA 337. In that case, the contractor had used a specified coating for pipes. This turned out to be unsuitable. The contractor was nonetheless liable, because it had also warranted that the pipes would be fit for purpose, and would not suffer from any defects due to their design. In coming to that conclusion, the Canadian Court made the following statement of principle:
“The general rule is that defects caused by an owner's specification are not the responsibility of the contractor, unless the contractor expressly guarantees that the construction would be fit for a specific purpose, or a warranty can be implied by the owner's actual reliance on the contractor's skill and judgment.”
The Court of Appeal goes back to first principles
Rather than applying any kind of general principle giving priority to absolute obligations, Jackson LJ instead approached the matter by going back to first principles of contractual interpretation under English law. Contracts are of course interpreted by asking what a reasonable person having all the background knowledge which would have been available to the parties would have understood the contractual language to mean. English law recognises that parties do not easily make mistakes in formal documents, but it also accepts that where it is obvious that something has gone wrong with the language, a Court is not required to find that the parties in fact intended a result that no reasonable person would have understood them to have had in mind.
Jackson LJ noted that an overly literal interpretation of complex documents that are bound to contain ambiguities and discrepancies might frustrate the commercial purpose of the agreement, and that he was entitled to adopt the more commercially sensible construction where a clause could be read in more than one way. He described the approach that the Court would take as follows:
“In essence, a court seeking to construe the contract between E.ON and MTH must postulate a reasonable person (X) having all the knowledge available to those two parties. The court must consider what X would have understood clause 8.1 of the conditions and [the passages in the employer’s requirements referring to a 20-year lifetime] to mean. This is an iterative process, which involves checking each of the rival meanings against the other contractual provisions and investigating its commercial consequences. The court must accept that there are likely to be ambiguities and inconsistencies within the documents. It must not allow itself to be led astray by those ambiguities and inconsistencies. Approaching matters in that way the court must determine whether [the contract] required MTH not only to comply with J101, but also to achieve a result, namely foundations with a service life of 20 years.”
The Court of Appeal’s starting point for the analysis where the passages in the employer’s requirements stating that the contractor’s design “… shall ensure a lifetime of 20 years …”. Looked at in isolation, that was a warranty, or an absolute obligation. However, the approach to design set out in J101, which was the principal standard that MT Højgaard had to follow, was aimed at building structures that had a sufficiently high probability of remaining in service for 20 years - and that was how the term ‘design life’ would reasonably be understood. Jackson LJ recognised that it was quite impossible for anyone to predict with certainty whether an offshore structure, continuously exposed to wind and waves, would in fact survive for 20 years. A contractor might design a structure that could withstand waves of an order of magnitude expected to occur only once in 50 years, only to then find the structure destroyed by a ‘once in a 100 years’ wave in the first month of its service life. In that situation, provided J101 had been followed, the contractor would have done enough to ensure a sufficient likelihood of the structure surviving for 20 years - it would have a ‘design life’ of 20 years. The Court of Appeal considered that J101 was simply not aimed at producing structures with a guaranteed life of 20 years.
The ‘purpose’ was limited by the accepted industry standard
That was the first blow to E.ON’s case. Turning to the conditions of contract, and Clause 8 specifically, Jackson LJ also noted that the majority of provisions imposed obligations of reasonable skill and care, and not absolute obligations or freestanding warranties. The fitness for purpose clause said that the ‘purpose’ was to be “… determined in accordance with the Specification using Good Industry Practice …”, and the definition of fitness for purpose referred to the ‘purpose’ as something to be “inferred” from the employer’s requirements. The Court of Appeal felt that all this ultimately went back to J101 and the approach to ‘design life‘, which was the central feature of the technical specifications in the employer’s requirements. The references to a lifetime of 20 years in the employer’s requirements therefore had to be read in the context of J101, which was concerned only with a sufficient probability of the structure surviving for 20 years.
In conclusion, Jackson LJ found that a reasonable person in the position of E.ON and MT Højgaard would know that J101 was the generally applicable standard in the construction of offshore wind farms, and that J101 required structures that could be expected to survive for 20 years, although this could not be guaranteed. He found that there was an inconsistency between the two statements in the employer’s requirements, that “… the design … shall ensure a lifetime of 20 years …”, and all the other provisions. The Court of Appeal was, however, not to be led astray by that inconsistency. Limiting the contractor’s obligations to following J101 was also commercially sensible. If the contractor had been required to give an absolute guarantee of a lifetime of 20 years, this should have been flagged clearly in the tender documents, so that MT Højgaard could have proposed a suitable allowance for this. That had not been done. As MT Højgaard had not been negligent in designing to J101 (the mistake was only subsequently discovered by DNV itself), it was not liable to E.ON for the remedial costs.
The Court of Appeal’s judgment illustrates how overly complex provisions imposing a range of obligations can lead to the most onerous clauses being watered down. If the contractor really had been giving an absolute, unqualified warranty of quality or fitness for purpose, then why would there be any need to include a (lesser) reasonable skill and care obligation in the contract, too?
Greater Gabbard - Apparent defects that might not preclude the intended use
Fluor Ltd v Shanghai Zhenhua Heavy Industries Ltd  EWHC 2062 (TCC) also concerned defects in monopiles and transition pieces for an offshore windfarm. Fitness for purpose was one of the many issues that the Technology and Construction Court had to consider.
Fluor was engaged as the EPC contractor for the foundations and infrastructure for the Greater Gabbard windfarm, a 140-turbine site in the North Sea near Suffolk. Fluor in turn subcontracted fabrication of the monopiles and transition pieces to Shanghai Zhenhua. Delivery was to take place in the Netherlands. From there, Fluor would transport the monopiles and transition pieces offshore to the installation site.
Fabricating the foundations required an enormous amount of welding. When the first shipments arrived in the Netherlands, Fluor noticed extensive cracking in the welds. The employer issued a number of non-conformance reports under the main EPC contract. Fluor then implemented an extensive programme of testing and repairing welds, having come to the view that the foundations were still suitable for installation once the cracking had been repaired. Subsequently, Fluor sought to recover the additional costs of this remedial programme from the employer in arbitration proceedings, but was ultimately unsuccessful. Fluor then pursued Shanghai Zhenhua for damages in excess of US$ 400 million (the subcontract price for the foundations had been €234 million).
The subcontract between Fluor and Shanghai Zhenhua contained a warranty of fitness for purpose, stating that the foundations to be supplied:
“… will conform with the specifications, drawings and other descriptions supplied or advised by [Fluor] and will be new, of good quality, fit and sufficient for the purposes for which they are intended as evidenced in this Purchase Order … of good materials, design and workmanship, free from defects, and will fulfil satisfactorily the operating conditions specified herein.”
There was no doubt that the ‘purpose’ for which the foundations were being supplied was for onward transportation to Greater Gabbard, to be installed offshore.
Fluor and Shanghai Zhenhua had also signed a waiver letter, following discussions during the testing and repair process. It appears that in the context of signing that letter, Fluor had accepted that the foundations were suitable for installation once the cracking had been addressed, and that they could in fact support wind turbines for 25 years, the agreed service life.
In Court, Fluor argued that when the foundations were delivered in the Netherlands, they were nonetheless not fit for their purpose. The cracking in the welds that was apparent on inspection meant that further investigation was required. Fluor said it could not send Shanghai Zhenhua’s foundations out to the Greater Gabbard windfarm in the condition they were in on delivery, as should have been the case. Shanghai Zhenhua’s case was that the test of fitness for purpose was objective: all that mattered was whether the foundations were fit for the purpose of being installed offshore, in order to support wind turbines for a service life of 25 years. That was something that Fluor did not ultimately disagree with, and so Shanghai Zhenhua argued that Fluor’s fitness for purpose claim should fail.
Need for investigations and potential repairs renders goods unfit for their purpose
Edwards-Stuart J noted that there was no English law authority directly on this point. This was a case where there was only one potential use for the foundations, so that if they were not suitable for that use, they would not be goods of ‘merchantable quality’. The judge referred to a decision of the Privy Council, Grant v Australian Knitting Mills Ltd  AC 85, in turn citing the High Court of Australia as to when goods would be ‘of merchantable quality’:
“The condition that goods are of merchantable quality requires that they should be in such an actual state that a buyer fully acquainted with the facts and, therefore, knowing what hidden defects exist and not being limited to their apparent condition would buy them without abatement of the price obtainable for such goods if in reasonably sound order and condition and without special terms.”
Fluor knew of the condition of the goods - they suffered from cracking in the welds - but, without carrying out lengthy further investigations, it did not know whether that meant they could not be used as intended. The experts for both parties agreed that the only reasonable course of action for Fluor, when presented with the foundations in the Netherlands, was to establish the extent of the problem, and repair what needed to be repaired, before transporting the foundations to the windfarm.
The judge concluded that a buyer in Fluor’s position would not have bought the goods without imposing a ‘special term’ of the kind that the Privy Council had mentioned in Grant v Australian Knitting Mills, namely that the foundations would pass the relevant investigations and tests. Where a buyer would have imposed such a special term had he known of the true condition of the goods (but where such a special term had not in fact been included in the actual contract), the goods would not be fit for their purpose, or ‘merchantable’. Applying that reasoning, Edwards-Stuart J therefore found that the foundations were not fit for purpose. The judgment only concerned liability, so we do not yet know what damages Fluor may recover as a result of that breach of contract by Shanghai Zhenhua. In particular, the waiver letter signed by the parties would seem to limit Fluor’s entitlement to recovering the costs of the investigations and repairs very significantly, by excluding the costs of such works carried out over a lengthy period of time.
As a matter of principle, however, the decision provides welcome clarification. A buyer in Fluor’s position should not have to take the risk of installing goods that appear to be suffering from defects, even though they might still be usable and ‘fit for purpose’ once repaired. What about a situation where the goods do not appear to be defective on delivery, and problems - or the need for repairs - only become apparent after installation? The passage that Edwards-Stuart J cited refers to a buyer who is “… fully acquainted with the facts …”, and who knows what “… hidden defects exist …”. This suggests that it may not matter if the buyer could not in fact have discovered whatever defects would have required further investigations upon delivery.
These recent decisions illustrate that while fitness for purpose is in principle the most onerous obligation that can be imposed on a contractor, it is still necessary to pay particular attention to defining the purpose, and to ensuring that other (perhaps apparently innocuous) clauses referring to reasonable skill and care do not undermine that standard. Of course, if the contractor is expected to give an absolute, unqualified warranty, not only should the purpose be expressly stated, but it may also be advisable to flag this in the tender documents (as Jackson LJ commented).