Delivery options for large-scale water infrastructure projects
Look out for our previous insight articles (as well as articles still to come) on the following topics:
- understanding water regulation;
- the modern minefield – water approvals for mining projects;
- water theft and offences;
- strategic water sharing options;
- project delivery options for large-scale water infrastructure;
- pricing and competition for water;
- native title and cultural heritage, and water; and
- defects and compliance in construction.
In this article, we offer our insights in relation to a number of the delivery models that are typically considered by proponents of major water projects in Australia.
The water sector is facing unprecedented challenges as climate change, population growth, ageing assets and competing interests increase pressure for limited resources. While a multi-pronged approach is required to address these risks, the upgrade of existing assets and the development of new infrastructure will be a vital part of this strategy.
Given the often substantial capital sums involved and the social and economic importance of the projects themselves, it is critical that governments select the optimum project delivery strategy to maximise value and reduce the risk of cost overruns and delays. There are a variety of contracting models that are typically considered for the delivery of major water projects. Ultimately, the model selected for a particular project will depend on a combination of factors, including the nature of the risks involved and the objectives of the project. Importantly, there is no ‘one size fits all’ solution.
In the ‘construct-only’ model, the project proponent engages a design consultant and separately tenders for contractors to perform the construction works. Pricing is typically on a ‘lump sum’ basis, but less certain components of the construction scope may be subject to a schedule of rates or cost-plus payment model.
This model is generally suited to projects where the works are relatively routine and uncomplicated and the risks are well understood, the timeframe for delivery is not compressed (so that design and tendering of the construction package can occur sequentially) and a high degree of cost certainty is required by the proponent.
Of course, large-scale water infrastructure projects will not necessarily fit this profile, with uncertainties in relation to site conditions and underground works (for example) increasing the ‘interface risk’ between the design and construction scopes. This can leave the proponent to bear the cost of design omissions and changes and increase the risk of delays and also cause uncertainty as to whether defects or performance issues are due to design or construction. There is also limited scope for innovation input from the contractor to drive down costs and the separation of design and construction scopes may result in the development of adversarial relationships between the proponent, contractor and design consultant.
These interface risks may be mitigated to an extent through the adoption of an ‘Early Tenderer Involvement’ (ETI) procurement model, which promotes collaboration between each construction tenderer and the proponent by refining the design during the tender phase in order to reduce the risk of design re-working being required later on. While the ETI process is unlikely to be a complete panacea for design-construction interface risks in complex projects, the construct-only model with ETI procurement has been used on a number of major water projects in Australia, including the current upgrade of Seqwater’s Lake MacDonald Dam on the Sunshine Coast in Queensland.
Design and Construct (D&C)
In the D&C model, the proponent prepares a design brief outlining the functional requirements for the works and a single contractor completes the detailed design and construction.
Like the construct-only model, the D&C model is generally suited to projects where the risks are well understood and a high degree of cost certainty is required by the proponent. There is also scope for an accelerated delivery schedule as construction can begin before full design documentation is completed (although longer tender periods than for construct-only contracts are typically required so the tenderer can assess design risks). Furthermore, unlike the construct-only model, both design and construction risks are transferred to the contractor, meaning that interfaces are able to be better managed and the proponent is provided with a ‘wrapped’ or ‘turnkey’ solution.
On the other hand, and given the substantial risk transfers to the contractor, the D&C model does not always provide the best ‘value for money’ solution as contractors are incentivised to include (sometimes significant) risk premiums in lump sum pricing, particularly in relation to high-risk or novel projects.
As with the construct-only model, the D&C model also has a collaborative variant in the form of an ‘Early Contractor Involvement’ (ECI) procurement process. This process involves the proponent engaging the contractor prior to award of the D&C contract to assist with scoping the project requirements. Again, while this can help mitigate ‘risk pricing’ by the contractor, it is unlikely to be a complete solution for more complex projects where it may remain difficult to efficiently price certain risks (for example, in relation to ground conditions or dewatering/aquatic salvage operations).
The D&C model has been employed on a number of significant water projects in Australia. Recently, the detailed business cases for the Granite Belt Irrigation Project and the Cloncurry River Dam recommended the adoption of the D&C model for the delivery of those projects.
Alliance contracting is a collaborative contracting model between a project proponent and various ‘non-owner participants’ such as designers, suppliers and construction contractors. While there is no universal form of alliance contracting, alliance contracts generally share the following key features:
- the alignment of interests of, and sharing of risks and opportunities between, the proponent and the non-owner participants;
- a commitment to a ‘no fault, no blame culture’ under which each party agrees that it will have no right to bring legal claims against another participant except in very limited circumstances (e.g. wilful default);
- unanimous decision-making by the proponent and each non-owner participant in respect of most (if not all) of the decisions affecting the project; and
- ‘transparency’ and ‘open book’ reporting between the proponent and non-owner participants.
The alignment of interests and sharing of risks and opportunities between the proponent and non-owner participants is facilitated by a remuneration model under which non-owner participants are reimbursed for their costs (plus margin), subject to a ‘pain-share/gain-share’ regime. For example, poor performance against defined metrics, such as exceeding the target cost for the project, may result in the parties sharing the cost overruns. Similarly, over-performance, such as delivering the project under budget, may result in the parties sharing the cost savings.
Alliance contracting is best suited for complex, high-value, projects where risks are difficult to price at the outset and where there is significant scope for value-adding throughout the delivery phase. The model is also suitable for projects with aggressive delivery timeframes, where there is insufficient time for scope definition, design and construction to occur sequentially. Perhaps unsurprisingly given these features, alliance contracting has historically been utilised on a number of large water infrastructure projects, including the delivery of the Wyaralong Dam and the ‘Stage 3’ upgrade of the Hinze Dam.
It is worth noting however, that the model does not have a consistent reputation for delivering projects on time and on budget and there has been a drift away from the ‘pure’ alliance model in recent years. As such, the model may not be preferred where the proponent requires a high degree of cost and schedule certainty. In addition, alliancing requires greater involvement of the proponent than other delivery models, making it ill-suited for inexperienced or under-resourced project sponsors. Furthermore, tender and alliancing establishment costs tend to be high, meaning the model may not be suitable for lower-value projects.
Public Private Partnership (PPP)
In a PPP, a government proponent engages the private sector to design, construct, operate and maintain infrastructure over the long-term. Generally, the private sector funds the delivery of the asset and recovers its capital costs (plus a return) through service or ‘availability’ payments during the operations and maintenance (O&M) phase. There are a number of variants, including where the government contributes funding to the delivery of the project and where the private sector assumes a level of demand risk based on payment by the public (rather than the government proponent) for use of the asset.
A PPP is similar to a D&C contract in that it offers a high degree of price certainty to the proponent, but goes beyond this by transferring life-cycle risk to the private sector with the aim of encouraging design, construction and operational synergies in order to reduce ‘whole-of-life’ costs. As such, the model is generally suited to major projects with well understood risks and complex capital investment programs which provide opportunities for savings to be generated during the O&M phase.
However, also similar to a D&C contract, the PPP model may not provide optimal value for money where delivery risks are high or poorly understood at the outset. Furthermore, the potential benefits of the model are further reduced where O&M costs are relatively small compared to overall delivery expenditure as there is limited scope for material reductions in whole-of-life costs. The model may also be less attractive to government where there is limited opportunity to fund service or availability payments through user charges. It is also worth noting that tender periods for PPP projects tend to be particularly lengthy given the time required for bidders to adequately assess not only D&C but also O&M risks.
Based on these considerations, PPPs are generally not well suited to delivering large-scale dam infrastructure. On the other hand, the model has been widely employed to deliver water treatment assets across the country.
The ‘managing contractor’ model is a form of ‘collaborative’ contracting which has characteristics of both D&C and project management contracts. In the managing contractor model, the proponent engages the managing contractor, who subcontracts all (or the majority) of the design and construction works packages and typically only ‘self performs’ project management services.
Like in a D&C contract, the managing contractor remains directly responsible to the proponent for the quality or ‘performance’ of the asset. However, unlike a D&C contract, the proponent retains a much higher degree of control over the delivery of the project, particularly in relation to appointment of subcontractors. The trade-off is that the managing contractor typically only undertakes a ‘best endeavours’ obligation to deliver the project on-time and on-budget.
This reduced responsibility for project delivery is reflected in the remuneration regime. That is, the managing contractor is generally paid a lump sum for carrying out its management services and is also reimbursed for its costs (plus margin) in relation to each of the subcontracted works packages. An incentive payment structure is sometimes introduced to encourage delivery by the target completion date and under the target cost.
A key benefit of the managing contractor model is that it allows the proponent to retain flexibility while sharing some of the delivery risks with an expert project manager. Like other collaborative contracting models, it is best suited to high-risk projects requiring complex project management where innovation is likely to be required during the delivery phase. However, it also suffers from some of the ‘downsides’ of collaborative contracts, such as reduced time and cost certainty. Furthermore, if works packages are subcontracted on a lump sum basis, ‘value for money’ concerns may arise, particularly in relation to more complex packages.
Of course, the models discussed above do not constitute an exhaustive list. There are number of other options, including hybrid structures that seek to combine elements of different models to suit the requirements of particular projects.
For example, a proponent who requires cost certainty but is keen to limit the upfront pricing of risk may seek to include a ‘risk pool’ mechanism in a traditional construct-only or D&C contract. In this case, the contract works would generally be priced on a lump sum basis, except for certain ‘higher risk’ scope items (for example, in relation to ground conditions or the quantity of certain materials). The ‘risk pool’ scope items are instead remunerated according to a ‘gain-share’ model (similar to the model discussed for alliance contracting above) where a target cost is established and the contractor shares in any ‘under-spend’ in completing the scope item. A ‘pain-share’ variant also exists, where a contractor’s entitlement to be paid for the risk pool item is not capped, but the contractor is required to absorb a portion of the ‘over-spend’ above a target cost.
At the other end of the spectrum, the ‘delivery partner’ model may be considered as an alternative to a ‘full blown’ alliance contracting structure. Under this model, a proponent engages one or more ‘delivery partners’ to assist with project planning, programming and design and construction management. The delivery partner model shares many of the features of an alliance contract, including the remuneration of delivery partners on the basis of a pain-share/gain-share model based on overall project performance. However, unlike a pure alliance, delivery partners are generally precluded from self-performing design and construction work, which is contracted directly by the proponent (meaning the proponent retains ‘hard’ contractual rights against package contractors in relation to the performance of their individual scopes).
While the delivery partner model is relatively ‘young’, it has been adopted in Australia on a number of high profile projects, including the delivery of the Western Sydney Airport by the federal government and the Woolgoolga to Ballina Pacific Highway upgrade by New South Wales Roads and Maritime Services. In the water context, the model is being used by Sydney Water to deliver the Lower-South-Creek-Treatment project, a $500 million program comprising the staggered upgrade and renewal of various wastewater assets in Sydney’s north-west.
The selection of a delivery model is rarely straightforward, but is a critical step in ensuring the objectives of a project are met. Our experts have experience with all project delivery models and are available to assist you with the selection of a ‘fit for purpose’ delivery model for your next project.
If you require assistance with any of the issues raised in this article, please contact one of the authors below.
This publication covers legal and technical issues in a general way. It is not designed to express opinions on specific cases. It is intended for information purposes only and should not be regarded as legal advice. Further advice should be obtained before taking action on any issue dealt with in this publication.