PPP Consultants
Closing a PPP on time and on budget — the discipline, not the decoration.
The route to Financial Close is decided in the months before bidders see a document. Authorities that give that period its due close faster, at lower cost, and with more bidders at the table.
PPP Consultants · September 2026.
A Public Private Partnership (PPP) is a long-term contract under which a private consortium finances, builds and operates a public asset — a hospital, a road, a water treatment plant — and is paid over twenty years or more for the service the asset delivers. Financial Close is the day the contracts are signed, and the lenders’ money is committed. Nothing is built before that day. Every month of delay before it is a month of service the public does not receive, and a month of bid cost that the private side must recover in its price.
Every procuring authority intends to reach Financial Close on programme. The record shows that most do not, and the reasons are rarely those cited at the time. Delay is blamed on the market, on lenders or on a change of minister. The cause is more often found in the preparation period: in an output specification that could not be measured, a payment mechanism that lenders could not model, a risk matrix that shifted at the final stage, or a scoring method that could not support a decision. This article sets out the five disciplines that decide the outcome, and the cost of omitting them.
Where the time goes.
The United Kingdom ran the largest PPP programme in the world for two decades, and its National Audit Office (NAO), the public spending watchdog, measured what happened. For projects closing between 2004 and 2006 the average tendering period — from the public notice inviting bidders to Financial Close — was 34 months, of which 15 months on average were spent in final negotiation with a single preferred bidder.
One project in three made significant changes to its scope during that final stage, and those changes amounted to 17 per cent of project value. Over the same period the share of projects attracting three or more developed bids fell from 85 per cent to 67 per cent.
Those four figures describe one failure. Fifteen months of negotiation with one bidder mean that the documents issued to the market were not ready to be signed. Scope changes worth a sixth of the project mean that the authority did not know what it wanted when it asked for prices. A falling number of bidders means that the market had learned that bidding was long, expensive and uncertain, and had started to stay away.
None of this was caused in the negotiation room. All of it was caused in the year before.
The pattern is not confined to one country or one programme. The Oxford research on large projects led by Bent Flyvbjerg, covering many hundreds of projects across sectors and continents, finds that nine out of ten have cost overruns, and that overruns of up to 50 per cent in real terms are common. Preparation is the only stage at which that outcome can still be changed cheaply.
Output specifications written to be measured.
An output specification is the document that tells the private partner what the asset must do — not how it must be built — for the life of the contract: how many beds available, what room temperature, what road surface condition, what water quality, and how each will be measured. It is the document to which every other part of the contract refers.
The strongest specifications share three properties.
Every requirement is tied to a quantitative test that can be run by a person with a meter, a checklist or a data feed, so that “available” and “unavailable” are facts, not opinions. The outcomes are aligned with the public benefit the asset exists to deliver, so that the contractor is paid for the outcome the public wants rather than for a proxy. The English is unambiguous and readable by a non-native speaker, because the people who apply the document in year fourteen will not be the people who wrote it.
A specification that fails any of the three tests is renegotiated after signature, at prices set without competition.
Payment mechanisms that align incentive with outcome.
The payment mechanism is the formula by which the authority pays the private partner: a unitary charge for the service, reduced by deductions when the service falls short. It is the contract’s engine: it either makes it commercially sensible to deliver what the specification requires, or it does not.
Three disciplines decide which.
Deductions must be material at the margins where the contractor makes its decisions — large enough that repairing a fault tonight is cheaper than leaving it until Monday, but not so large that a single event threatens the lenders’ repayment and makes the mechanism unfinanceable.
Measurement must be operationally feasible and drawn from data that neither party can manipulate, because a deduction that depends on a disputed reading will be disputed.
The mechanism must remain workable across the life of the contract, through changes in use, in technology and in the people operating it. A mechanism that reads well but cannot be modelled by a lender’s analyst will be rewritten in the negotiation period, and the fifteen months of negotiation begin.
Risk allocation that survives lender due diligence.
Financial Close happens when the lenders’ due diligence — their independent review of every technical, legal and financial risk in the project — accepts the allocation of those risks in the contract. Lenders do not accept the authority’s view of a risk; they accept evidence.
The allocation must therefore do three things.
It must allocate every identified risk explicitly to one party or the other or state the formula by which it is shared; a silent risk is a risk the lenders price at its worst case.
It must show that the party carrying each risk can absorb it — that the contractor’s balance sheet, insurance or subcontracts stand behind the risks assigned to it. Where a risk is shared, the formula must be one that an analyst can enter in a spreadsheet.
Authorities sometimes believe that a harder allocation is a better bargain. The market answers in the price: capital is mobile, and it is offered first to the projects whose risk is clearest. The World Bank’s count of private infrastructure investment in low- and middle-income countries came to US$100.7 billion across 315 projects in 2024, of which US$7.9 billion across 47 projects was in Sub-Saharan Africa.
Every one of those projects competed for the same lenders’ attention, and those with a transparent allocation were financed first and on the best terms.
An evaluation method that supports a clear decision.
The scoring method is the part of preparation that authorities most often leave to the end, and it is the part that most often stops a close. Bids are scored, not read: the evaluation panel does not form a view of the best proposal; it applies the published criteria to each bid and records the result.
That works only if the method was built before the documents were issued. The criteria and their weightings are published with the invitation, so that bidders write to them and no bidder can later claim the rules changed. Each criterion is written so that three assessors working separately would reach the same score, which means a defined scale with a description of what earns each mark.
Every score is recorded with its reasons, so that the award decision can be explained to the losing bidders and, if necessary, to a court. An award that cannot be explained will be challenged, and a challenge stops the clock.
Named accountabilities and a weekly rhythm.
The four disciplines above are documents. The fifth is management. Every task in the preparation programme is assigned to a named individual with a date, and the programme is reviewed weekly against those dates by someone with the authority to move resources.
This appears too obvious to state. In practice, preparation slips because a task belongs to a department rather than a person, because a decision waits for a meeting that is not scheduled, or because the authority’s advisers are working to a fee rather than to the programme. The remedy is administrative and unglamorous, and it is the difference between a preparation period of nine months and one of twenty.
A case from the firm’s work.
Three secondary schools at Clacton, in Essex, were on the verge of failing to close. An affordability gap — the difference between what the bids would cost the council each year and what it could pay — had opened, and the incumbent financial adviser had been unable to close it.
The firm, appointed PPP adviser to Essex County Council, advised the opposite of the usual cut: expand the scheme, placing a large community centre and library at its heart. Costed as additionality, the expansion carried a fresh claim for central government support, and bidders were asked to price the additional build at marginal cost — the cost of adding it to a project already under construction, not the cost of building it alone. The gap closed.
All three schools and the community centre and library were built and are in use today, and the council went on to recommend the firm to other authorities. The case shows the five disciplines from the other side: the affordability test belongs in preparation, and when it is run late the remedy must be found in the structure of the deal, because by then there is no competition left to find it in the price.
What to do differently.
An authority planning a PPP should treat the preparation period as the project, not as the run-up to it. Before the invitation to bid is issued, four documents should exist in a form a lender could sign: the output specification with its tests, the payment mechanism with its model, the risk matrix with its evidence, and the evaluation method with its scales.
Each should have been read by a person who will apply it and by a person who will finance it. The programme to produce them should carry a name against every line.
That is the work that closes a PPP on time and on budget. It is done before the market sees anything; it is invisible in the announcement; and it is the whole difference between the projects that closed in eighteen months with three bidders and those that closed in three years with one.
Sources: National Audit Office, Improving the PFI tendering process, HC 149 Session 2006–07, March 2007. Flyvbjerg, B., What you should know about megaprojects and why, Project Management Journal, 2014. World Bank, Private Participation in Infrastructure Annual Report 2024.