Carillion paid out £554 million.
It never priced the execution environment that distribution was landing into.

NAVETRA does not replace Carillion's audit, accounting, or pensions oversight. It prices what those systems already hold. Carillion collected and reported the cash, debt, pension and supplier-financing data behind the distribution decision. What its board never had, and what it increasingly needed, was that data expressed as one dollar figure it could challenge before signing the next dividend, instead of reading it as a liquidation balance afterward.

What this casebook is, and what it is not

This is not a legal reconstruction or a substitute for the Parliamentary inquiry, the regulatory findings, or the audit-related proceedings. It is a capital-allocation read built entirely from the public record.

NAVETRA was never engaged by Carillion. Nothing here attributes any Carillion outcome to a NAVETRA-led decision, and no Operating Profit at Risk figure is assigned to Carillion. Inventing one would be the exact overclaim this casebook exists to refuse. The accounting and audit failures, and the later enforcement action against former executives, belong to the assurance seat and are not what NAVETRA prices. The claim is narrow and deliberate: the distribution data existed, Carillion reported it, and the decision to keep paying it out was never priced into one board-readable range before the cash ran out.

The decision being priced

Carillion was created in 1999 by demerger from Tarmac and expanded through acquisitions, among them Mowlem, Alfred McAlpine, and Eaga, into scale, public-sector reach, and a balance sheet shaped by goodwill, thin contract margins, a growing pension deficit, and reverse-factoring that bought time while obscuring the true debt position.

The decision a board owns here is not the accounting treatment. It is the distribution call: whether to keep paying, and increasing, dividends while the cash, pension and supplier-financing data described an execution environment that could not carry them. That decision recurred every year. Each row below is a decision point, not a complete causal account.

2009–2016. £554M paid in dividends, a large share of operating cash generation across the period. The cash the distribution consumed was the same cash the long-cycle contract book and the pension deficit had first claim on.

2012–2016. Reverse-factoring and creditor growth increased balance-sheet pressure while making the true debt position harder to read. The exposure was in the working-capital data; it was carried as a financing technique, not priced as a distribution constraint.

2013 onward. Pension and financial-risk concerns were raised by trustees and analysts. The pension claim on future cash was quantified and external; it was not weighed as a hard limit on the distribution decision.

2015. Board material referenced external analysis linking pension exposure and financing structure to a more stressed debt picture than the headline figures. The read existed; it did not change the distribution.

2016. A record dividend was paid while goodwill and reporting assumptions were carried at levels Parliament later criticised. The distribution decision moved in the opposite direction to the environment data.

Jul 2017 → Jan 2018. A major profit warning, then compulsory liquidation with £29M against nearly £7B. The decision converted to a lagging number.

The split that protects the read

A casebook that claimed a priced read would have prevented Carillion's failure would be dismissed by any director who has run a long-cycle contract book, and rightly so. The discipline is to separate the two halves and only claim the endogenous one.

The reverse-factoring point, read as a distribution constraint

Reverse factoring is not improper. The capital point is narrower: it moved a real obligation into a form the distribution decision did not treat as debt. Parliament and external analysts later argued it left the true debt burden understated. Priced as what it was, a claim on the same cash the dividend consumed, it tightens the distribution constraint at the point the board signs, not after the working capital unwinds.

The artifact

NAVETRA assigns Carillion no Operating Profit at Risk figure here. What the artifact shows instead is structure: which client-facing domains carried the endogenous exposure on the distribution decision, expressed as the actuarially weighted, sector-validated range a board reads on one page before it signs the dividend, not the liquidation balance it reads after.

Connect it to the data Carillion already collected

Every input above was already inside Carillion. The cash generation sat in treasury. The pension claim was quantified and raised by trustees. The reverse-factoring obligation sat in working-capital reporting. The contract underperformance sat in programme data. The board received papers and external analysis. Carillion collected all of it and reported most of it.

What Carillion did not have was the dollar layer that data represented set against the distribution decision before it converted, expressed as one actuarially weighted, sector-validated range aligned to ISO 31000 and the company's existing enterprise-risk framework. Not a new metric to adopt. The price tag on the data already on the table. That alignment is the difference between a board chair finding the analysis persuasive and a board chair able to defend the distribution decision to creditors and trustees.

How much was external, and how much was organisational

Not every part of Carillion's failure was preventable. Thin-margin, long-cycle contracting carries structural fragility, and sector conditions were genuinely hard. Treating the whole collapse as an isolated board failure would be inaccurate, and this casebook does not.

But treating it as sector misfortune would be equally inaccurate. The distribution decision, paying and increasing dividends while the environment data worsened, was endogenous and Carillion's own reporting supported the read. The split is analytical, not accounting-based, and it can be debated. The harder point survives the debate: a meaningful share of this loss was carried as a confidence narrative when it could have been read as a number.