This article was first published on ICLG.com on 9th July 2019
The best way to procure large-scale infrastructure projects has been in the spotlight in recent times. Much of this has been driven by a growing acceptance that current contracting arrangements are not fit for purpose, with concerns that contractors are being asked to take on too much risk and the corresponding commercial benefits for employers can be limited.
In the UK, the collapse of Carillion in January 2018 has been followed by large balance sheet provisions being reported by some UK contractors, who have been quite open that they now regard infrastructure projects as too risky to take on. In international or emerging markets, where infrastructure is often procured under concessions with limited or non-recourse financing, contractors are often expected to assume unquantifiable risks under fixed price turnkey contracts.
The adequacies of current contracting methods from an employer's perspective have also been highlighted by a recent report on Crossrail in which the UK National Audit Office noted that adopting a fixed lump sum price form of contract, which Crossrail has established for some of the remaining work to improve cost certainty, means that Crossrail risks losing commercial levers to ensure that contractors prioritise completion of Crossrail over other projects and opportunities. 1
There is a growing consensus in the UK that the current approach is unsustainable and that different methods of contracting need to be explored. It is possible that the approach for international projects will also be re-examined and in some sectors, like offshore wind, more collaborative contracting practices are already developing.
This chapter looks at how contracting strategies can be made more collaborative. It then considers how English law has approached the interpretation of contractual commitments seen in contracts of this nature.
The Traditional Route to Market
The usual approach to risk management in contract procurement is that risk should be allocated to the contracting party best able to manage the risk. In project finance and other turnkey projects, this means that the vast majority of construction phase risk is devolved onto the contractor who is expected to carry these risks within the cost and programme contingencies it has allowed in agreeing the contract price and programme.
In theory, it creates clear divisions of responsibility between the employer and contractor and in that sense it creates a structure which should deliver certainty of outcome. In seeking to obtain the best possible deal, procuring bodies may try to squeeze the contractor’s contingencies through competitive tendering, except where this is not possible because the supply chain is shallow.
However, it pre-supposes that all risks can be identified and their impacts assessed, that the supply chain is willing to take on the risks allocated, there is sufficient depth of resource and expertise in the supply chain for the employer to select the contractor willing to take on most risk at the lowest price, that the contingency for risk-taking that the contractor charges is considered to be economic and represents value for money and that ultimately the contractor’s balance sheet is sufficiently strong to bear the costs if the risks materialise without going into insolvency.
A method typically seen in traditional contracting to diversify risks has been for contractors to create unincorporated joint ventures under which they collaborate with other contractors for specialist skills and share downside outcomes. However, the contracts will create joint and several liabilities that expose each member of the consortium to the entirety of the risks in the contract. Where a corporate guarantee is provided to the employer as performance security for contractual performance, this can expose the wider group of each contractor to project risk.
Contractors will respond to these risks by requiring clauses which limit liability for contractual breach. For the employer, however, the inclusion of such provisions means that it will be left without a remedy once the limits on liability have been reached.
Despite client resistance to a progressive approach in emerging markets, the contracts published by FIDIC in 2017 contain a number of new provisions which focus on both the employer and the contractor operating in accordance with good project management practice.
A potentially important innovation is a system whereby a contract which complies with FIDIC’s Golden Principles can be given its seal of approval. In particular, the Golden Principles confirm that the rights and obligations of the parties should be generally as defined in the standard form (GP1) and discourage changes being made to the balance of risk and reward (GP3). In effect, FIDIC does not want employers to amend its contracts in a way which significantly amends the standard risk allocation.
Another innovation is the inclusion of an advance warning system requiring notification of events that might increase the contract price or cause delay. There are also provisions for the contractor to make proposals to avoid or mitigate the effects of the causes of increased cost and delay, for both parties to attend management meetings and for the contractor to provide better quality programme information and to update its programme if it does not reflect actual progress.
Many of these arrangements are designed to assist in the management of the contract by bringing greater and earlier transparency to events that cause delay or additional cost. They can be described as collaborative as they encourage the employer and the contractor to work more closely with one another.
However, this is quite a limited aspiration. The changes made by FIDIC do not go so far as to create a genuinely collaborative environment where the employer and contractor work closely together as an integrated team. The contracts continue to operate along normal lines as being either remeasurement or for a lump sum fixed price and the contractor pays damages for late completion or other contractual defaults.
The FIDIC contracts therefore continue to represent quite a traditional approach.
Target Cost Contracts
Target cost contracts have been regarded in some markets as being much more collaborative than traditional methods of procurement.
They have been used on a large number of infrastructure projects in the UK over recent years including the London 2012 Olympics, Thames Tideway Tunnel, Crossrail and Hinkley Point C. Target cost contracting has also been promulgated for High Speed 2. The NEC family of contracts contains two variants and the ICC published a model form in 2018. FIDIC has not yet followed suit.
Target cost contracts usually operate by comparing the actual outturn cost of works (ignoring certain categories of disallowed cost) against an agreed target cost and dividing between the employer and the contractor in accordance with pre-agreed share ranges the benefit and burden of any cost savings or overruns.
The underlying philosophy is that the contractor will be incentivised to beat the target cost as its return will be boosted by its share of any achieved saving. The financial gains to be made by both the employer and the contractor if the target is beaten will encourage them to co-operate for the benefit of the project rather than concentrate on protecting self-interest.
In its simplest form, the share range is equalised (50/50) between the employer and the contractor. In practice, such a simple arrangement may not always be achievable for a variety of reasons. For example, on large and complex projects, the contractor may not be willing to accept that its share of cost overrun can continue without some sort of tapering off or an absolute upper limit. On occasion, the contractor’s share of cost overrun may be combined with an aggregate cap on the contractor’s liability for default as the contractor will wish to know that its overall exposure under the contract is ring-fenced so as to preserve its balance sheet should the contractor encounter significant problems.
Setting the target at the right level is crucial for the effectiveness of a target cost contract. The employer may be wary of setting the target price too high as this insulates the contractor from the risk of cost overrun and potentially rewards the contractor for cost savings that have not been genuinely earned. The contractor may be wary of setting the target price too low as it makes the achievement of savings much harder and increases the risk of generating a cost overrun.
Other risks may materialise if the target cost is set too low. The contractor might be encouraged to look for claims that will result in an upward adjustment to the target cost. This can therefore inadvertently create an adversarial environment which is not too dissimilar to that more often associated with lump sum fixed price contracts, placing a heavy focus on claims and claims management. Using a competitive tender process to drive down the target cost to the lowest achievable number may not therefore be in the best interests of either the employer or contractor.
Target cost contracting does transfer less risk to the contractor than would be the case under a lump sum fixed price contract. For example, it usually provides for the employer and the contractor to share under the gain/pain mechanism the risk of making good defects before completion and extra costs due to sub- contractor default.
However, the contractor still retains substantial risks under target cost contracting, against which the contractor will wish to provide a pricing and programme contingency to help manage its share of retained risk. As an example, the insolvency of a sub-contractor and re-procurement of a replacement sub-contractor can cause significant delay and additional cost. The contractor may also face potential liability for delay liquidated damages for late completion.
Where it is running late, the contractor may wish to manage its risk for late completion by accelerating works and increasing resources to recover as much of the programme as is possible. It is likely that the extra cost of these measures will be treated as outturn cost and will be shared with the employer under the gain/pain share mechanism.
Even if the target cost has been accurately assessed and fixed at the outset of the contract, a claims culture can still therefore be prevalent in target cost contracts. It is to be expected that the contractor will look to protect its commercial interests first and foremost and will regard any additional return by achieving a saving as a secondary consideration.
It is often crucial for an effective target cost model that the pain/gain mechanism agreed by the employer and contractor is flowed down into the supply chain. By doing so, the position of sub-contractors can be aligned with the main contractor. If this is not done, the benefits for the employer of target cost contracting may be impaired if it means the sub-contractors are operating on a traditional fixed price basis, as this is likely to drive the position taken by the contractor at main contract level.
So, whilst target cost contracting is often regarded as a collaborative contract, it needs to be recognised that there can be tensions within the contracting structure due to the fact that the interests of the employer and the contractor are not truly aligned. This will become most apparent if it becomes clear that the outturn cost could exceed the target or there are significant delays or quality issues.
For large or complex projects, or where there is an emerging scope of work or the project involves potentially high-impact risks, and where success depends to a greater or lesser extent on a genuinely collaborative approach, this tension may not drive the right behaviours between the contracting parties.
Incentive-based contracts are generally set up on a cost reimbursable basis where the contractor has limited downside exposure. The contractor is paid the actual cost incurred to complete but only profit, or a proportion of profit, is put at risk. It is paid all the costs to complete, except for narrowly defined disallowed cost, and certain liabilities are curtailed or removed completely.
The basic premise is that successful delivery of the project is most likely to be achieved where there is a close alignment between the commercial interests of the employer and the contractor, so that they work together for the benefit of the project, rather than against one another because they have different commercial objectives.
This can be achieved if the contract incentivises successful outcomes in return for which the contractor can improve its profit, rather than by the employer relying primarily on damages or other contractual penalties for non-performance (as is the case with a traditional or, to some extent, a target cost contract). For example, this means that the contractor should not face the risk of delay damages being imposed in the event of late completion of the works, since the imposition of time risk might drive an approach which is inconsistent with the collaborative behaviours that the remainder of the incentive contract is designed to encourage.
The incentives can be structured in a variety of ways and there is no one-size-fits-all approach. They can focus on near- and longer-term objectives, usually through a KPI process, and can also operate by reference to strategic project outcomes, such as beating the employer’s project budget and achieving commercial operation by the employer’s target date.
The contractor does not therefore need to include the cost and programme contingency in its proposals, or at least to the same extent, by comparison to other contracting models. As the employer holds more of both the risk and the contingency, there is less risk of adversarial circumstances arising.
An incentive contract can still contain elements of fixed prices where this is appropriate, although perhaps not in a conventional sense. For example, it might stipulate a fixed price or cap on the cost of the contractor’s management resource.
Payment of a proportion of the contractor’s own monies (whether for its fee or management resource) might be subject to a monthly assessment against agreed KPIs. These might measure achievement for matters such as project reporting, design deliverables, expenditure forecast accuracy, quality assurance and programme.
Some flexibility might be introduced into the KPI mechanism to encourage good performance. For example, if the contractor misses the target for a management KPI, it may still recover lost payments if it achieves the target on subsequent occasions.
At the end of the project, the contractor could earn a share of the saving where the outturn cost is lower than a pre-determined incentive target and if commercial operation is achieved within an agreed timescale, perhaps with a sliding scale for short delays to avoid a cliff edge. The employer bears actual costs to complete in excess of the incentive target and the contractor loses its share as there is no saving.
In order to avoid any such incentive payment being regarded as too remote on long-running projects, the contract might provide for the contractor to be paid an interim share of any forecast saving calculated at pre- agreed intervals, if at that time the programme shows that commercial operation will be achieved within the agreed timescales.
Whilst incentive mechanisms may focus on the risks of cost overrun and timely delivery, the approach on quality of work may vary. The contractor may retain responsibility for quality including achieving compliance with the contract specification and drawings. However, the KPI mechanism might also reward matters such as identifying and fixing defects prior to delivery to site and general compliance with an agreed quality assurance protocol.
Even though it is important that the incentive payments should not be too remote for the contractor to achieve, a number of project risks might not be within the direct control of the employer or the contractor. The incentives should in these cases encourage the contractor to respond to the occurrence of risk events in a positive way in order to reduce their cost and programme effects should the risks materialise.
As in target cost contracting, it is important for incentive contracts to see that the incentive payment arrangements at main contract level are cascaded down to the sub-contractors engaged below the main contractor on a fair and proportionate basis so that the wider supply chain is working to a common set of goals.
Alliance contracting was conceived in the North Sea oil and gas industry as part of an effort to achieve enhanced project outcomes by eliminating the adversarial atmosphere inherent in traditional contracting. Since then, alliance contracting has been used to some extent in the UK but more significantly in Australia.
Alliance contracting is a more holistic approach to collaborative contracting. Whereas target cost contracts and incentive-based contracts still rely on the employer entering into contract with a main contractor, alliance contracts by contrast create a horizontal multi-party relationship between the employer and the wider supply chain. The underlying philosophy is a “one team” approach where there is collective rather than individual accountability.
The alliance team (including the employer) works together to achieve common goals and shares risks and responsibilities as opposed to risks and responsibilities being allocated between the parties as happens in a traditional contract. However, some risks may be retained by the employer, particularly where it is the only entity that can influence or manage the risk. The commercial interests of the alliance team are aligned for the benefit of the project through a common pain/gain share regime in which the entire alliance team participates.
Alliances are typically formed of three components: the alliance board; the alliance manager; and the alliance delivery team. The alliance board is comprised of representatives of each alliance participant and is responsible for setting the strategy of the alliance, allocating work and appointing an alliance manager. The alliance manager manages the day-to-day activities of the alliance delivery team and implements the decisions of the alliance board. The alliance delivery team are the individual participants of the alliance and are responsible for delivering the work of the alliance. The employer is responsible for setting the alliance objectives such as defining the scope and setting performance requirements.
Decisions are made by the alliance board on a ‘best for project’ and unanimous basis and in accordance with the alliance charter. The common payment arrangements are a key enabler for this approach. Accordingly, participants are expected to set aside self-interest for the interests of the project and are encouraged to reach mutually acceptable solutions. Unanimity in decision-making dilutes the employer’s control but the employer may reserve power to direct the alliance on key issues (including regulatory requirements, scope changes, suspension and termination). In the absence of a deadlock breaking mechanism, the requirement for unanimity in decision making exposes the alliance to the risk of deadlock where consensus cannot be reached.
Alliance contracts also seek to foster a no-blame culture by extinguishing all liability under the contract but exceptions will apply (e.g. in relation to liability that cannot be excluded by law, wilful default and insolvency). One significant practical issue which this creates is that the employer foregoes any rights against participants for damage or loss suffered as a result of defective work after final account. Any other position regarding defects would undermine the collective philosophy of alliancing.
The pain/gain share compensation regime operates by assessing actual outturn cost against a fixed target outturn cost for the whole project to determine if there is a cost overrun or saving. As with target cost contracts, the supply chain participants’ share of cost overrun is likely to be capped at a certain level. In addition to sharing in cost savings and overruns, alliance contracts may also feature rewards for KPIs which are linked to programme and performance.
Budget and programme targets are agreed by all participants of the alliance and so early involvement needs to be sufficient for all parties to have confidence in them. It is crucial that these are set at the right levels.
Alliance development agreements have been used on some projects and provide the employer with an opportunity to road test collaboration in practice and to fix an agreed target outturn cost. Inevitably, the collective sharing of risk and reward means that participants are exposed to the risk of poor performance by other participants. This accentuates a need to ensure that participants have confidence in each other’s capabilities.
Another key feature of alliance contracting which fosters integration and collectivity is that participants are required to commit to open and transparent procedures with mutual access and audit rights extending beyond paperwork to acting in a wholly transparent fashion.
There is sometimes a misperception that alliance contracting disproportionately shifts the balance of risk against the employer as it bears the ultimate liability for cost overruns and late completion. However, this perception is unjustified where there is a structured and robust pain/gain share regime which exposes the supply chain to tangible risks in the event of poor or late performance. For example, the financial implications to the employer of late completion could be treated as part of the outturn costs so that they are taken into account in the calculation of the gain/pain share.
An employer considering an alliance contract should be aware that its success is dependent on actual implementation of effective team working and collectivism (e.g. demonstrable implementation of the desired behaviours set out in the alliance charter). Disruption to the efficient operation of the alliance can be caused if key individuals are not retained and are subject to replacements. The employer also needs to recognise that setting up the alliance will require a significant investment in time and human resources and expertise, on its part, and this commitment will need to be deployed for the duration of the project. It is not a soft option for employers.
Alliance contracting is therefore generally considered to be most appropriate for projects which have the following characteristics:
- project risks which cannot be defined clearly or accurately;
- the cost of transferring unknown or unquantifiable risks is prohibitive (e.g. large-scale projects involving legacy systems; offshore oil and gas projects);
- the employer has valuable knowledge, skills and capacities or a significant role to play (e.g. water distribution networks, rail projects in a live rail environment); and/or
- there is a need for flexibility to accommodate changing circumstances (e.g. interfaces with stakeholders and other work packages; system integration issues).
Legal Developments in the Construction of Collaborative or “Relational” Contracts
It is perhaps not a coincidence that, as the general demand for more collaborative, and less adversarial, contractual relationships has grown, there has also been growing judicial interest in what are often referred to as “relational” contracts, which are typically defined by similar characteristics to those found in some of the collaborative contracting methods described in the preceding paragraphs. Although relational contract theory is not a novel concept, the English courts are increasingly recognising the collaborative or “relational” nature of certain contractual arrangements, and suggesting a willingness to adopt a slightly different kind of approach to the construction of these contracts compared to more traditional forms of commercial contracts.
What are “relational” contracts?
In the recent decision of Bates v Post Office, 2 Fraser J helpfully provided the following list of “special characteristics”, which would be relevant to determining whether a contract between commercial parties was a relational one:
- There must be no specific express terms in the contract that prevent a duty of good faith being implied into the contract.
- The contract will be a long-term one, with the mutual intention of the parties being that there will be a long-term relationship.
- The parties must intend that their respective roles be performed with integrity, and with fidelity to their bargain.
- The parties will be committed to collaborating with one another in the performance of the contract.
- The spirits and objectives of their venture may not be capable of being expressed exhaustively in a written contract.
- They will each repose trust and confidence in one another, but of a different kind to that involved in fiduciary relationships.
- The contract in question will involve a high degree of communication, co-operation and predictable performance based on mutual trust and confidence, and expectation of loyalty.
- There may be a degree of significant investment by one party (or both) in the venture. This significant investment may be, in some cases, more accurately described as substantial financial commitment.
- Exclusivity of the relationship may also be present.
Many of these characteristics are features of collaborative contracts, particularly when used for long-term projects. However, whilst previous authorities have tended to focus on the long-term nature of relational contracts, Fraser J suggested that the above list was not to be exhaustive, and no single one of the above characteristics would be determinative (with the exception of the first). This could mean, therefore, that collaborative contracts may be deemed “relational” regardless of whether or not they are entered into on a long-term basis, although long-term contracts remain more likely to fall into this category, since they will generally satisfy more of the above criteria.
What then, if any, special rules are applicable to the construction of, and implication of terms in relation to, relational contracts, and how do they affect collaborative contracts?
Interpretation of express terms
It has long been recognised that the courts may need to adopt a flexible approach to the interpretation of relational contracts, in particular, long-term contracts that are necessarily drafted broadly and flexibly so that the parties can adjust their bargain in accordance with changing conditions.3
This principle was reaffirmed in the recent case of Amey v Birmingham,4 which involved a 25-year PFI contract comprising around 5,190 pages. Several years into the contract, the contractor modified its performance based on a new interpretation of the contract, which effectively reduced its workload. In rejecting the contractor’s interpretation, the Court of Appeal noted that:
“Any relational contract of this character is likely to be of massive length, containing many infelicities and oddities. Both parties should adopt a reasonable approach in accordance with what is obviously the long- term purpose of the contract. They should not be latching onto the infelicities and oddities, in order to disrupt the project and maximise their own gain.”
This hints at a move away from the literal and natural meaning of the words used in a contract if that would involve giving effect to “infelicities and oddities” in a relational contract. The authorities on the interpretation of relational contracts are still relatively few and new. It remains to be seen whether and to what extent the courts will interpret relational contracts in a different, perhaps more purposive manner.
Implication of good faith duty
It is widely accepted that the duty to act in good faith is not an overriding principle that is implied into any commercial contract. However, where the contractual relationship between parties is of a relational nature, the English courts recognise that the parties will be subject to an implied obligation of good faith.
Fraser J summarised this obligation in Bates v Post Office as meaning that: “the parties must refrain from conduct which in the relevant context would be regarded as commercially unacceptable by reasonable and honest people”. Further, there is more to a duty of good faith than a requirement to act honestly.
It is increasingly the case that construction contracts provide expressly for parties to act in good faith or stipulate a similar formula. For example, the NEC contracts have for many years required the parties to act “ in a spirit of mutual trust and cooperation”. Even traditionally adversarial contracts such as the JCT have begun to include provisions for collaborative working “in good faith and in a spirit of trust and respect”.
Whilst the precise meaning of good faith will be determined by the relevant contractual context,5 in practice, parties to collaborative contracts should expect to be bound by some form of good faith obligation, whether expressly or impliedly. Nevertheless, a general obligation of good faith is unlikely to qualify or limit any specific rights that the parties have expressly agreed, and parties can generally expect to exercise such rights freely (e.g. a right to terminate for convenience).6
While collaborative contracting is by no means new, it continues to evolve as parties look to move away from traditional, adversarial forms of contract. Of course, no form of contract is a panacea, and construction is an inherently risky activity, but incentivising and driving the right behaviours – both in the structure of the contract and in its terms – can reduce the scope for surprises and disputes.
- See section 3.9 of “Completing Crossrail”, a report by the National Audit Office dated 3 May 2019.
- Bates v Post Office Limited  EWHC 606 (QB).
- Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd  EWCA Civ 396.
- Amey Birmingham Highways Ltd v Birmingham City Council  EWCA Civ 264.
- Compass Group UK and Ireland Ltd (t/a Medirest) v Mid- Essex Hospital Services NHS Trust  EWCA 200.
- See endnote 5.