Paraguay just doubled the room its PPPs may occupy.
Under Ley 7452/2025 — which repealed the 2013 PPP law and was brought into force by decree in February 2026 — the cap on accumulated firm and quantifiable contingent commitments, in present value and net of fiscal assets, moves from 2% to 4% of GDP. The annual Treasury payment cap doubles too, to 0.8% of GDP.
The design is better than the headline suggests: contingencies count inside the ceiling, and the Ministry of Economy and Finance now holds functions that were scattered across agencies. A ceiling with one gatekeeper is a real constraint.
The State may now fund 70% of a “private” partnership.
Vietnam’s amended PPP law (Law 57/2024/QH15) lifts the state capital share from 50% to 70% of total investment where land clearance and resettlement are heavy, where the project sits in a disadvantaged region, or where technology transfer needs the support to pencil.
At 70% public capital the label stops doing any work. What is left to the private partner is delivery and operation — which can be exactly right. But the value-for-money case has to be re-run at the new ratio, not inherited from the old one.
£50bn bought. £136bn still owed.
NISTA’s PFI and PF2 summary data, published March 2026, counts 665 operational PFI contracts across hospitals, schools, roads, waste and IT — a combined capital investment of roughly £50 billion.
This is what a ceiling looks like from the far end. The commitments were affordable, one contract at a time, on the budget arithmetic of the year each was signed. The stock is what nobody voted on.
Roughly £2.70 of future charge for every £1 of capital delivered — financing, maintenance and service bundled into a payment that runs for decades after the ribbon is cut. Each contract was defensible on its own. The programme was never appraised as a programme.
Private money is back above $100bn.
The World Bank’s PPI database puts private participation in infrastructure at USD 100.7 billion — up 16% on the year and 20% on the five-year average, and the first time the threshold has been cleared since the onset of COVID.
Rising volume is not the risk. The risk is that capacity to appraise grows slower than capacity to commit — and the gap between them is filled by contracts nobody has priced at portfolio level.
A record year is also a correlation year. The deals signed into one buoyant window share a vintage — the same rate curve, the same traffic forecasts, the same currency assumptions — so they sour together, not one at a time.
Raise the ceiling. Then count what fills it.
A doubled cap, a 70% public stake, a record year of private capital — and, from the country that started earliest, £136bn of bills still to pay on £50bn of assets. Room is easy to create. Capacity to pay is the thing that has to be measured:
Every ceiling above was set by someone who had counted the commitments. None of them had counted the capacity.