The Alliance PPP delivery model

5 Dec

First published 1 January 2021

1. Introduction

Australian contractors and other non-owner participants are crying out for a risk allocation reset, particularly for inherently risky mega-projects. Consequently, many project owners are turning to more contractor-friendly contracting models such as alliance contracting and other forms of collaborative contracting.

At the same time, high levels of government spending and borrowing, in response to the economic impacts of the pandemic, will place pressure on government balance sheets, causing governments to look for opportunities to leverage private sector investment in public infrastructure. But the risk allocation on privately financed PPP projects has traditionally been at the harsh end of the spectrum for designers, contractors, equipment manufacturers and service providers. And many non-owner participants have paid dearly for under-pricing these risks, under fixed price contracts. The use of fixed price contracts for separate packages/systems also exacerbates the integration risk.

Which raises the question: can a privately finance PPP be combined with an alliance style risk allocation?

Traditionally, the response to this question has been ‘no’, because the obligations the owner participant assumes under an alliance contract are inconsistent with the risk profile it needs to raise limited recourse project finance.

But what if the procuring government agency assumes the role of the owner participant in the alliance, and the project company that raises the limited recourse project finance is treated as a non-owner participant, similar to the designer, contractor and key equipment suppliers? Could this arrangement be made ‘bankable’? The answer is ‘yes’, for the reasons explained below.

But before I explain how to incorporate an alliance style risk allocation into a privately financed PPP, I’ll briefly explain the key features of the two concepts for those who a less familiar with them. Readers who are familiar with these concepts can jump to section 4.

2. Alliance contracting

Alliance contracting is well suited to projects that are very complex or where the scope needs to be developed. The key features of alliance contracting can be summarised as follows:

Single, multi-party contract: The project owner and the other key project participants (i.e those participants whose contribution will be integral to the success of the project) enter into a single, multi-party contract. In Australia (and most other parts of the world), the contract is usually called a “Project Alliance Agreement”. In the USA, it’s called an “Integrated Project Delivery” or “IPD” contract. The non-owner participants in the alliance are typically called “Non-Owner Participants” or “NOPs”.

Early involvement: The project alliance agreement is generally signed at an early stage in the project planning phase, often before the scope of the project is settled and well before an owner would traditionally engage the civil construction contractor. The NOPs are often selected based on non-price criteria, such as credentials, experience and demonstrated ability to collaborate effectively in an alliance contracting framework. The owner and the NOPs then jointly develop the scope of work, programme, cost estimate and other KPIs against which the performance of the alliance will be measured. Sometimes the owner will develop these items with two competing groups of NOPs before selecting the successful group with whom the owner enters into the final Project Alliance Agreement.

Remuneration: The remuneration regime for the NOPs typically involves 3 limbs:

First, the owner reimburses all direct costs incurred by each NOP in undertaking the project, including if such costs exceed the costs included in the agreed target cost.

Second, each NOP will be entitled to an amount on account of profit and contribution to corporate overheads. This limb-2 amount is often called the “Fee”, and is typically a lump sum amount, based on amount of profit and contribution to overheads that the NOP would expect to receive for ‘business-as-usual’ performance the amount of work it is expected to perform.

Third, each NOP will be entitled to a Gainshare Payment from the owner, or will be liable to make a Painshare Payment to the owner, depending on how the alliance performs against the agreed target cost, target completion date and other agreed KPIs. This way the NOPs can generate a better than BAU margin if the alliance performs well. Conversely, the Painshare Payment will eat into the Fee, and hence reduce each NOP’s margin, if the alliance performs poorly.

The maximum Painshare Payment a NOP can be required to pay is typically limited to the amount of its limb-2 Fee. This way the worst-case outcome for a NOP is that it recovers its costs but makes no profit or contribution to its corporate overheads.

Importantly, the KPIs are based on whole of project outcomes, rather than outcomes on the scope performed by the relevant NOP. This encourages each NOP to do what’s best for the entire project, rather than what’s best for its scope of work. This incentivises greater collaboration between the NOPs than occurs under traditional contracting models where the financial performance of each NOP turns on the performance of its scope of work rather than whole of project outcomes. This helps with projects involving the integration of different systems from different suppliers.

Governance: Project decisions, such as the final design and construction methodologies to be adopted, the sequencing of tasks, deciding what materials will be used, and whether any works will be subcontracted, typically require the unanimous agreement of all alliance participants. Sometimes the owner may reserve certain decisions for itself, but on the basis that the knock-on effect of such decisions on the target cost and other KPIs will be determined by unanimous agreement.

No blame: Alliance contracts typically include a ‘no blame’ clause under which each participant agrees to waive its right to bring legal claims against another alliance participant, including if the other participant is in breach of its contractual obligations or is negligent. The only real exception to this is typically for wilful (ie deliberate) default by the wrongful participant.

The no-blame clause removes the ability of participants to blame and sue one another when things go wrong. Instead, their desire to minimise the ‘painshare’ (or loss of gainshare) arising from mistakes and poor performance motivates all participants to help the alliance to overcome the problem, regardless of who was is at fault.

The no-blame regime can also encourage participants to accept stretch targets and adopt highly innovative approaches in the pursuit of extraordinary outcomes, without the fear of being sued if things go wrong.

Risk: Under this arrangement all project risks are essentially shared between the owner and the NOPs, until each NOP reaches its painshare cap, at which point the remaining cost and other risk impacts fall exclusively to the project owner. In return for sharing more risk, however, the owner has a contractual framework under which it only pays additional amounts on account of risks that actually occur. Because the direct costs of the NOPs are always paid by the owner, there is no need for the NOPs to include a contingency for risk in the fixed price that they would charge if engaged under a traditional fixed price contract — which the owner pays whether or not the risk occurs.

Alliance contracting is well suited to complex mega-projects that involve the integration of multiple systems.

If you’d like to learn more about alliance contracting, click here.

3. Privately financed PPPs

Private sector investors in new infrastructure projects like to limit their financial exposure to the amount of equity they invest. They do this by establishing a special purpose vehicle that borrows the remaining funds needed to pay for the design and construction of the infrastructure.

The debt finance is borrowed on a ‘limited recourse’ basis, meaning the lenders can only have recourse to the SPV and its assets for repayment of the loan. The lenders cannot have recourse to the equity investors, beyond the amount of their equity in the SPV. The consequence of this is that the lenders are exposed to the project risks faced by the SPV. If the occurrence of such risks affects the SPV’s ability to complete the project and repay the loan, the lenders are exposed.

Lenders of limited recourse debt finance have a relatively inflexible risk/reward equation. They require a high degree of certainty that the debt will be repaid. Accordingly, they have little appetite for their borrower – the SPV – retaining risks that could adversely affect the SPV’s ability to repay the loan. Commensurate with their low risk profile, they only receive a fixed margin by way of interest on the loan and don’t share in profits generated by the SPV if the project is successful.

The upshot of using limited recourse debt is that the ability of the SPV to accept and manage risk is constrained by the amount of equity invested it has to absorb the foreseeable financial consequences of any risks it retains. If the SPV wishes to accept a higher level of risk, it will need more equity.

Equity finance is more expensive than debt finance because it is exposed to greater risk. So more equity will result in a higher weighted average cost of capital for the SPV, which will ultimately translate to a higher project cost. Bidders for PPP projects want to minimise their cost of capital, to maximise their chances of winning. They do this by minimising the SPV’s use of equity, and maximising its use of debt. But doing so also minimises the SPV’s financial capacity to retain and manage risk.

It is for this reason that privately financed SPV’s generally transfer most risks most project risks through to their contracrors, under a fixed price contracts with very few entitlements to extra money.

It’s also the reasons why its very rare to see a privately financed SPV assume the role of an owner participant under an alliance contract.

4. How to incorporate an alliance style risk allocation within a privately financed PPP

But what if:

the SPV is treated like a NOP in the alliance; and
the government procuring agency assumes the role of the owner participant?

Let me expand.

4.1 Typical service payment PPP scheme

The basic scheme under modern Australian service-payment PPP contracts for major projects is presently as follows:

The Government Agency typically commits to fund a portion of the capital cost of the project, as the infrastructure is designed and built.
The SPV finances the balance of the capital cost, and progressively pays the fixed D&C price to its D&C contractor as the D&C works progress.
The D&C contractor, in turn, progressively pays the fixed prices it has agreed with its Design Consultant and two System Suppliers as their respective deliverables are produced.
The Government Agency agrees to pay a quarterly service payment (QSP) to the SPV, from the time construction is completed and the infrastructure facility becomes operational, until the end of the PPP contract term.
The QSP is used by the SPV to:
- pay its O&M contractor;
- service and repay its debt financiers; and
- provide a return to its equity investors.
The QSP is reduced (abated) if the operational performance of the project does not meet all agreed operation phase KPIs, which results in a reduction to the return that the PPP company can pay to its equity investors.
Very poor operational performance can result in the Government Agency terminating the PPP contract for default, if the SPV and/or its debt financiers are unable to remedy the default.

4.2 The Alliance PPP model

The basic PPP scheme described above could be adjusted as follows to incorporate an alliance style risk allocation:

The Government Agency, the SPV, the Design Consultant, the two System Suppliers, the Civil Contractor and the O&M Contractor enter into a single contract.
The Government Agency pays the actual direct costs incurred by the Design Consultant, the two System Suppliers, the Civil Contractor and the SPV (together, the non-owner participants) during the D&C phase, monthly in arrears.
The Government Agency also pays these parties (other than the SPV) their limb-2 fee in accordance with the alliance contract.
The SPV progressively draws down its debt facility to reimburse the Government Agency for the direct costs and limb-2 fee, up to the maximum cumulative amount permitted under the loan’s monthly cumulative drawn down schedule.
The Government Agency funds any payments to the non-owner participants in excess of this maximum cumulative amount.
At the end of the D&C phase, the performance of the alliance against the D&C phase KPIs is finally assessed, and any gainshare payments to, or painshare payments by, the non-owner participants are calculated and paid.
For the non-owner participants other than the SPV, the final gainshare or painshare payment is paid shortly after the D&C phase performance of the alliance is finally assessed (following the expiry of the agreed defects correction period for the Civil Contractor and Systems Suppliers).
In the case of the SPV, the gainshare or painshare payment is paid over the remaining term of the PPP contract by way of an increase or decrease to the QSP in accordance with the pre-agreed formula. This way, the SPV’s equity investors share in the D&C phase ‘gain’ or ‘pain’ by way of an increase or decrease to their equity return (but not total loss of their equity investment) over the term of the PPP.
The (adjusted) QSP would continue to be subject to abatement for poor operational performance, to motivate the SPV and its O&M contractor to optimise the operational performance of the facility.
Likewise, the Government Agency would continue to have the ability to terminate the contract for very poor operation phase performance, if the SPV and/or its debt financiers are unable to remedy the situation.

5. Further optimising

No doubt the Alliance PPP model outlined above could be further optimised. For example:

Drawdown of the debt facility could be timed to avoid the need for the Government Agency to fund the D&C phase payment pending reimbursement by the SPV (up to the maximum cumulative drawdown amount).
The Designer, Civil Contractor and Systems Suppliers could be further incentivised to optimise whole of life outcomes by tying a portion of their final gainshare (or painshare) payment to the facility’s longer term operational performance.
Perhaps the SPV could be paid a fixed price for the managing its contractors and suppliers during the D&C phase, although there is much to be said for better aligning the SPV’s interests with the Government Agency’s by tying a significant portion of the SPV’s D&C phase profit margin to the alliance’s performance during the D&C phase.

I’m keen to hear the views of others on this model, and how it could be optimised.

6. Conclusion

The Australian mega-project market is crying out for contracting models that:

address the concerns of the sector regarding risk allocation;
facilitate the high levels of collaboration needed to overcome the package integration risks associated with complex, transformational mega projects;
overcome the inherently adversarial nature of traditional fixed price contracting and the high level of disputation it causes;
facilitate the adoption of collaboration dependant technologies and practices that would significantly enhance construction sector productivity, such as BIM, Critical Chain Project Management (shared float), and shared cost contingency arrangements; and
can leverage private sector capital.

The alliance PPP model proposed in this article is surely worthy of consideration.

[Updated on 15 January 2021]

Learn more