When the PFI ends, can anyone prove what condition the lights are in?

The first contracts expire in 2028. More than 30 English councils are behind them in the queue, and most have not checked what they are getting back.

Between 2000 and 2013, more than 30 English local authorities signed 25 year Private Finance Initiative deals to replace and maintain their street lighting. Those contracts are now running out. The first expire this decade, the rest through the 2030s, and every one ends the same way: the network comes back to the council, along with whatever condition it is in.

Handback is not a date in a diary. It is a process with a financial test at the end of it, and the evidence that decides the outcome has to be gathered years before the lights change hands.

What the contract actually requires

Street lighting PFI agreements do not let a concessionaire hand over a tired network and walk away. Most contain three obligations that bite in the final years.

The first is residual life. Columns, lanterns and cabling have to come back with a defined minimum life remaining. If they do not, the cost of putting that right is not the council’s to absorb.

The second is the asset record. The contract requires an accurate inventory: GIS mapping, column age and material, cable routing, central management system data. Not an approximation, a record that matches what is on the street.

The third is transfer. Operational data, software licences and asset management systems have to move across in a form the authority can use on day one.

Behind those sits the enforcement machinery. Retention of the monthly unitary charge into a joint account. Deduction of the value of works not completed. And, where the asset record cannot be relied on, the authority’s right to commission independent testing across the network and charge it back.

The test nobody has evidence for

The residual life conversation is where most handbacks will get difficult, and it is the one both sides are least equipped for.

To argue that a column does not meet the required standard, an authority needs evidence. To argue that it does, so does the contractor. In most cases the only formal survey in the programme is the joint condition survey commissioned near the end, by which point both parties have taken a position and the conversation is adversarial by design.

Walsall Council, whose contract expires in April 2028 and which has published its expiry plan, names this in its own risk register. It records that discussions about post-PFI residual life may become a point of contention, and says early discussions should take place about assets that appear not to be in acceptable condition. Its plan does not set out how those early discussions get evidenced.

That gap is not unique to Walsall. It is structural. The sector has built a process that produces the evidence at exactly the moment it becomes contested.

Six years for one network

Walsall is unusually well prepared, and its plan shows what the timescale really looks like. Expiry planning began in January 2023 for an April 2028 handback, with a further year of certification and account closure running to April 2029. Six years, for one network.

The arithmetic is unforgiving. A joint condition survey takes time to scope, procure and complete. Whatever it finds has to be remediated, inspected and agreed. Retention arrangements have to be set up and settled. Authorities that have not started have less room than the expiry date suggests.

There is a further difficulty no single council can solve. More than 30 networks reach the same stage within roughly the same decade, and they will all need the same things. The specialist market is not large. Authorities that instruct late will compete for capacity with authorities that instructed early, in a window that cannot be extended because the expiry date is fixed in the contract.

Which brings us to how the survey actually gets done

Inspecting the top of a 10 metre column means putting a person up there. That means a mobile elevating work platform, and on anything but a quiet residential street it means protecting people working next to live traffic. A typical gang is three vehicles and three people: the platform vehicle, a support vehicle carrying temporary traffic management, and on faster roads an 18 tonne impact protection vehicle.

Every column takes the same cycle. Drive to it, set out signs and cones, deploy outriggers, check the ground will take them, raise the operative, inspect, lower, retract, collect the cones, move on. Plus a permit for every piece of road space occupied.

Birmingham City Council‘s own works record puts that rate at 14.5 columns a day.

The alternative is to fly the column instead of climbing it

Working with Kier on Birmingham’s 94,000 column network, Highways Drone Services has surveyed 13,248 columns from the footway using drones. No lane closures, no impact protection vehicle, nobody at height.

The rate was 75.1 columns a day. Five times the traditional method. Across the programme that removed 737 days of site occupation from the public highway, and zero accidents, injuries or near misses were recorded across the trial and the rollout that followed.

On a 10,000 column pre-handback survey that is the difference between 690 working shifts, a little over three calendar years with a single crew, and 133 shifts, around six months. Which is the difference between finishing in time to enforce remediation against the contractor and finishing too late to do anything with what you found.

What the survey actually found is the interesting part

The productivity numbers get the attention. The data is the part that matters for handback.

Across those 13,248 columns the survey recorded 3,574 OSID node numbers, 900 timers and 424 photocells. It also found 520 columns with missing or damaged identification stickers, and 355 where tree growth meant no access at all.

Read that in handback terms. On a sample of roughly 14 per cent of one city’s network, nearly 4 per cent of assets could not be reliably identified from their own labelling, and a further 2.7 per cent were effectively invisible to a ground inspection.

That is what an unverified asset register looks like when someone finally goes and checks. Not catastrophic, not unusual, and entirely invisible until it is captured. It is also exactly the sort of discrepancy that turns a residual life discussion into a dispute, because neither party can evidence a column nobody can positively identify.

Why the licensing matters more than the aircraft

Highways Drone Services holds CAA operator registration and the relevant pilot authorisations, along with NHSS 12D traffic management certification, Highways Passport, CSCS and £10m public and employers liability cover. It runs a Chapter 8 compliant vehicle, equipped and Starlink connected, so data leaves site the same day.

Most drone firms hold the aviation side. Most survey firms hold the highways side. This work needs both, which is why it is not a job to hand to a general drone operator.

What it does not replace

Drone survey does not do ultrasonic column wall thickness testing, structural load testing under ILP GN22, or electrical testing to BS 7671. Those need physical contact, and all three are done at ground level anyway.

What aerial survey does is tell you which assets warrant that testing, instead of assuming an even spread across the network. On most networks that targeting is where the money is saved.

Three questions worth answering this year

When was the asset register last checked against the street, rather than updated from records?

Is there a dated, independent condition baseline that predates the formal survey? If not, the first evidence either party has will arrive in the middle of a negotiation.

Has anything been budgeted? Handback surveys are a capital-scale line item, and they fall in years already written into medium term financial plans.

Start with one area

Pick a single ward or maintenance area. HDS surveys it, reconciles it against your existing register and shows you the discrepancy rate. You then decide on the full network from a real number rather than an estimate.

The Birmingham work began as a controlled 10 day trial. It is now business as usual on the city’s maintenance contract.

highwaysdroneservices.co.uk | 01442 968223 | hello@highwaysdroneservices.co.uk