Carillion Counted the Profit Before It Arrived. It Never Arrived.
S U M M E R S C H O O L · L E S S O N 5 O F 7
Lesson 5: the biggest corporate failure in British history — and why every business that runs on projects is running a small version of the same risk.
Welcome back to Summer School. This one is for everybody whose business runs on projects — agencies, consultancies, developers, builders, anyone who invoices for work that takes months to deliver. Which, among my readers, is most of you. Fair warning: of all seven lessons this summer, this is the one most likely to describe something happening in your accounts right now.
In January 2018, Carillion — Britain’s second-largest construction group, 43,000 employees, 450 government contracts covering everything from HS2 to hospital maintenance to school meals — went into compulsory liquidation. Not administration, where a rescue is attempted. Liquidation. It was the largest of its kind in British corporate history. The company had £29 million of cash against liabilities of nearly £7 billion, including around £2 billion owed to 30,000 suppliers and subcontractors, many of them exactly the size of business you run.
Six months earlier, it had been a FTSE 250 stalwart worth over a billion pounds, reporting healthy profits, paying a dividend that had risen every single year. So this week’s question: how does a company report profits for years and then discover, more or less overnight, that they were never really there?
Profit on a long contract is a prediction
Here’s the mechanic at the heart of it, and it applies to a £50k website build exactly as it applies to a £350 million hospital.
On a long project, you don’t wait until the end to record your profit — you recognise it as you go, in proportion to how complete the work is. Perfectly standard, perfectly sensible. But notice what that means: the profit you book in month three depends entirely on your estimate of how the project finishes. What the final costs will be. Whether the client will pay for the overruns. Whether that claim you’re negotiating will land. Profit on unfinished work isn’t a fact. It’s a forecast wearing a fact’s clothing.
Carillion, by the parliamentary inquiry’s account, ran those forecasts on sunshine. Troubled contracts were carried at optimistic values year after year. Claims were counted before they were agreed. And because next year always contained the recovery, the gap between the story and the reality compounded quietly — until July 2017, when a new finance director’s balance sheet review forced the truth out all at once: an £845 million write-down on contracts, most of it three UK hospital and road projects and the Middle East. By that September the total hit reached £1.2 billion — enough to wipe out the previous eight years of reported profits combined.
Eight years of profits — celebrated, bonused, dividended — handed back in one announcement. The money hadn’t vanished. It had never existed. The share price fell 70% in two days, and six months later the company was gone.
The two lie detectors that were flashing all along
You couldn’t see the optimism from outside by reading the profit line. You could see it in two other places, and both of them exist in your business too.
The first is cash conversion. If profits are real, they turn into cash — not instantly, but reliably. Carillion’s didn’t. In the five and a half years before the collapse, it paid out £333 million more in dividends than its operations generated in cash. For over half a decade, the reward for profitability exceeded the cash the profits produced. Longtime readers will recognise the Wilko lesson wearing a hard hat — profit is an opinion, cash is a fact — but here the gap wasn’t a dividend policy problem. The gap was the tell that the profits themselves were inflated.
The second is what the borrowing was dressed as. Carillion ran an “early payment facility” — suppliers waited up to 120 days, and a bank paid them sooner at a discount, with Carillion owing the bank. Around £500 million of what was functionally borrowing sat in the accounts as ordinary trade creditors rather than debt. Two weeks ago we watched Wilko die when supplier credit was withdrawn; Carillion is the same instrument from the other side — a giant quietly financing itself out of the pockets of businesses your size, and calling it something else.
And the safety nets? The auditors had signed the accounts for nineteen years; the regulator later called the failures in those audits exceptional and issued a record £21 million fine. If you took one thing from last week’s Patisserie Valerie lesson, it holds here at four hundred times the scale.
The small-business version of this disease
Scale it down and the symptoms are mundane, which is what makes them easy to live with. Work-in-progress that’s been sitting on the balance sheet for months because nobody wants to admit the project’s gone sour. Profit booked on jobs where the scope changed and the paperwork didn’t. A management-accounts margin that’s somehow always healthier than the margin projects actually finish at. Drawings and dividends set against the booked number rather than the banked one. None of it fraudulent. All of it Carillion, in miniature.
So, four disciplines for any project business, all cheap.
One: age your WIP monthly, and treat old WIP as guilty until proven innocent. Work that’s been “about to be billed” for ninety days usually isn’t an admin backlog. It’s optimism, hiding.
Two: track your cash conversion — over a rolling year, what percentage of reported profit actually arrived as cash? There’s no single right number, but a persistent, unexplained gap is your accounts telling you the profit line is running ahead of reality. It’s the simplest honesty check a set of management accounts can face, and most have never faced it.
Three: hold a five-minute post-mortem on every completed project — final margin against the margin you recognised along the way. One project finishing below its booked margin is a project problem. Most projects finishing below it is a recognition problem, and it means your profits are systematically overstated. Better to find that out from a spreadsheet than the way Carillion’s shareholders did.
Four: only count what’s agreed. Scope changes, overruns you expect the client to cover, claims under negotiation — until there’s a signature, that’s hope, and hope doesn’t belong in the P&L.
This corner of finance — WIP, utilisation, revenue recognition on projects — is genuinely specialist, and it’s one of the areas my practice goes deepest on, precisely because it’s where growing project businesses most often mislead themselves entirely by accident. The £845 million question fits any size of company: if someone independent reviewed your contracts tomorrow, would your booked profit survive the meeting?
Next Wednesday — Lesson 6: The Body Shop. A beloved British brand, bought and then broken within months. What the sale of a business does and doesn’t fix, and why the first hundred days after a deal decide everything. Subscribe and it’ll find you.
Sophie Wright is the founder of WrightCFO, a top-five UK fractional CFO practice working with founders scaling between £1M and £10M. The team includes CFOs who specialise in exactly this territory — WIP, utilisation and revenue recognition for project businesses. If this lesson landed a little close to home, book a 30-minute discovery call: tell Sophie about your business, and she’ll match you with the right specialist from the practice.
