Value for money analysis compares the whole-life cost of delivering a project as a public-private partnership against delivering it conventionally, through a public sector comparator. The framework is sound. Its reputation is mixed because the result is highly sensitive to a small number of inputs that are difficult to observe and easy to select favourably.
The four inputs that determine the answer
The discount rate
P3 structures typically shift cost from early construction years to later availability payments. A higher discount rate therefore mechanically favours the P3. This makes the choice of rate the most consequential single assumption in the analysis, and it should be justified explicitly rather than inherited. Applying different discount rates to the two options requires a strong argument; applying the same rate to both, with sensitivity analysis across a stated range, is the defensible default.
Risk quantification
The public sector comparator is adjusted upward for risks the public sector would retain and the private partner would assume — construction cost overrun, schedule delay, latent defects, lifecycle cost variance. These adjustments frequently constitute the entire measured advantage. They should be built from a documented distribution of outcomes on comparable projects, not from a percentage uplift applied by convention.
Two disciplines improve credibility considerably: quantify each risk separately with its own probability and impact rather than applying a blanket contingency, and show the analysis with and without risk adjustment so a reader can see how much of the conclusion depends on it.
Competitive neutrality
The comparator must be adjusted for advantages the public sector enjoys that are not real economic savings — most obviously tax exemptions and, in some analyses, self-insurance. Omitting these overstates the public option. Overstating them understates it. The adjustment should be itemised.
Whole-life cost boundary
A P3 concession typically includes maintenance and lifecycle renewal to a defined condition standard. The comparator must include the same scope, funded at the level required to meet that standard — not at the level the entity has historically budgeted. Comparing a fully funded P3 against a chronically underfunded public maintenance programme compares two different service levels and answers the wrong question.
The disclosure that makes the analysis credible
Publish the sensitivity results, not just the base case: how the conclusion changes across a plausible range of discount rates, construction cost variance, and lifecycle cost assumptions. If the P3 advantage disappears within the plausible range of any single input, that is the most important finding, and it should be stated rather than buried.
Risk transfer versus risk sharing
Risk is only genuinely transferred to the extent the private partner can bear it and the public entity can credibly decline to intervene. Demand risk on an essential facility, force majeure on a critical asset, and catastrophic events all tend to return to the public sector regardless of contractual allocation, because the alternative is service failure. Analyses that credit full transfer of risks the entity would in practice absorb overstate the case.
The useful test is a question: if this risk materialised at its worst plausible severity, would the entity actually allow the partner to bear the consequence, including the possibility of the partner's insolvency? Where the honest answer is no, the risk is shared and should be priced as shared.
What value for money analysis does not decide
It does not decide whether a project should proceed — that is a separate benefit-cost question, and a project with poor economics does not become worthwhile because P3 delivery is marginally cheaper. It does not measure affordability, which concerns whether the entity can accommodate the payment stream within its budget and debt capacity across the concession term. And it does not address the governance question of committing successors to a payment obligation for twenty-five or thirty years.
Practical governance
Three arrangements improve outcomes materially: the analysis is prepared or independently reviewed by a party with no financial interest in the transaction proceeding; the assumptions register is published; and the analysis is refreshed at each major gate, including after bids are received, when actual pricing replaces estimates. The third is the most often skipped and the most informative, because the received bids are the first genuine market evidence in the process.
This publication is general information and is not legal, accounting, audit or financial advice. See our Disclaimer. Found an error? Write to [email protected] — we correct in place and note what changed.