A company built to be too useful to fail
By the end of 2017, Carillion held around 450 contracts with the UK government, spanning hospitals, schools, prisons, roads and defence accommodation. It employed roughly 43,000 people worldwide, 19,000 of them in the UK, with many more working through its supply chain. Six weeks later, on 15 January 2018, it collapsed into compulsory liquidation — the most drastic procedure available in UK insolvency law — with liabilities of nearly £7 billion against cash reserves of around £29 million.
Carillion had been demerged from the construction firm Tarmac two decades earlier and grown since through a long series of acquisitions into a large, diverse network of construction and support-services businesses, without ever settling on a clear identity beyond being a major player in government outsourcing and public-private partnerships. The Institute for Government's later retrospective offered a blunt one-line diagnosis: Carillion was a company that had taken “high-risk, low-price bidding” to its logical extreme (Institute for Government, 2020). The scale of that risk-taking is captured in a single statistic covering the year before the collapse: in 2016/17, none of government's three largest suppliers at the time — Capita, Carillion and Amey — made a profit, despite all three winning more government work that year (Institute for Government, 2020).
Warning signs that predated the crisis by years
The July 2017 profit warning was not the first sign that something was wrong. Concerns about the sustainability of Carillion's debt levels had already been raised as early as 2015, well before the company's headline figures gave any public indication of trouble. The business model that produced those concerns was itself part of the problem: a strategy of growth through debt-funded acquisition, layered onto a portfolio of low-margin contracts, while headline revenue stayed largely flat rather than growing to match the debt taken on to fund expansion.
That combination — rising debt, acquisitions bought on credit, and thin operating margins on the contracts actually being delivered — describes a company with almost no room to absorb a single bad year, let alone the compounding write-downs Carillion was about to announce. A firm whose entire strategy depends on continuing to win new, thinly priced work simply to service debt taken on for past acquisitions is not managing risk so much as postponing it. The warning signs were visible to anyone examining the balance sheet closely enough; what changed in July 2017 was that the problem became too large to keep managing quietly.
A rosy set of accounts, then a record dividend
Nothing in Carillion's own public statements suggested trouble was close. Its 2016 accounts, published on 1 March 2017, presented what a subsequent joint parliamentary inquiry called “a rosy picture.” On the back of those results, the company paid a record dividend of £79 million, £55 million of it paid out on 10 June 2017, and awarded large performance bonuses to senior executives (House of Commons, Business, Energy and Industrial Strategy and Work and Pensions Committees, 2018).
Underneath that confident surface, Carillion's pension fund was already in serious trouble: a deficit of £811 million by 2016, larger than the company's entire £701 million balance-sheet equity at the time (London Business School, 2018). Choosing to pay a record dividend to shareholders rather than direct that cash toward the pension shortfall was, on the analysis of business academics reviewing the collapse afterwards, a decision senior management took with some awareness that it increased the possibility of eventual default (London Business School, 2018).
Four months later, the picture reversed
On 10 July 2017 — just over four months after those rosy accounts were published — Carillion announced a profit warning: a £845 million reduction in the value of its contracts, following a review by its auditors, KPMG. The share price collapsed immediately, and the chief executive and chief financial officer were both replaced (The Pensions Regulator, 2018). The write-down was increased to £1,045 million that September — more than the company's previous seven years of profits combined (House of Commons, Business, Energy and Industrial Strategy and Work and Pensions Committees, 2018).
The government's own contingency planning for a possible Carillion failure began almost immediately after that first profit warning, and accelerated through the autumn, according to the National Audit Office's later investigation (National Audit Office, 2018). And yet, in the period between the July warning and the January collapse, Carillion announced it had won four further central government contracts and three contract extensions, worth around £1.9 billion combined (National Audit Office, 2018) — work continuing to flow to a supplier the government was already quietly preparing to lose.
By the weekend before its collapse, Carillion's market capitalisation had fallen to around £61 million, down from roughly £2 billion in 2016. A last attempt to secure a rescue deal with government and its lending banks failed, and on 15 January 2018 the company's own directors petitioned the High Court for compulsory liquidation (Fortune, 2018; House of Commons Library, 2026).
Who actually paid for it
The cost of Carillion's collapse did not fall on the people who had approved the dividends and bonuses. Around 27,000 members of its defined benefit pension schemes were left facing reduced pensions through the Pension Protection Fund, which absorbed what was, at the time, its largest-ever single hit — a pension liability estimated at around £2.6 billion (House of Commons, Business, Energy and Industrial Strategy and Work and Pensions Committees, 2018). Around 30,000 suppliers and subcontractors were owed roughly £2 billion between them, most of which they were never likely to recover (House of Commons, Business, Energy and Industrial Strategy and Work and Pensions Committees, 2018); Carillion had a reputation as a notoriously slow payer even before its collapse. Estimates of the total job losses caused across the wider supply chain vary by source, with one industry retrospective putting the figure as high as 75,000 — a number worth treating as indicative of scale rather than as an agreed, precise count.
The National Audit Office estimated the direct cost to UK taxpayers of the government's handling of the collapse at £148 million, a figure it described as subject to real uncertainty and likely to take years to settle precisely — and explicitly separate from the wider costs to the economy, to Carillion's own staff, and to its supply chain and creditors (National Audit Office, 2018). A later, dedicated NAO investigation into two hospital PFI contracts Carillion had held — the Royal Liverpool and the Midland Metropolitan — found a pattern the Institute for Government's commentary on that investigation summed up plainly: a willingness to accept low-price bids without adequately testing whether the winning bidder could actually bear the risk it was taking on (Institute for Government, 2020; National Audit Office, 2020).
Parliament's own verdict
The joint parliamentary inquiry into Carillion's collapse did not treat the company's failure as a matter of bad luck. Its report placed responsibility squarely on the company's own board, describing them as having presented themselves to the inquiry “as self-pitying victims of a maelstrom of coincidental and unforeseeable disasters” rather than acknowledging their own role in the outcome (House of Commons, Business, Energy and Industrial Strategy and Work and Pensions Committees, 2018).
A separate committee, the Public Administration and Constitutional Affairs Committee, examined the government's side of the relationship rather than Carillion's own conduct, and reached a formal conclusion in July 2018: government procurement had been driven by price to the point of failing to appreciate differences in quality between contractors, and this had contributed to a decline in the quality of public services delivered (Public Administration and Constitutional Affairs Committee, 2018). That is not journalistic commentary or academic inference. It is the recorded conclusion of a Parliamentary select committee, examining government's own procurement behaviour.
Why “cheapest first” was the rule Carillion exploited
By this point it is worth asking why government was running this kind of price-led process at all. UK public contracting has a long history of being awarded on personal relationship rather than price — the historian Philip Harling's account of “Old Corruption” is the standard scholarly reference point for how entrenched that culture once was, with sinecures and public contracts handed to the well-connected rather than the competent (Harling, 1995; 1996). The formal end of that culture arrived substantially through the founding EU public works procurement directive, in force from 1971 and binding on the UK from its 1973 accession to the EEC, which made competitive, openly advertised tendering compulsory above certain thresholds (Council Directive 71/305/EEC). Choosing the lowest compliant bid became the visible, defensible, hard-to-corrupt alternative to a system with a genuine and long-documented fairness problem.
What that reform removed, along with the favouritism, was a less visible asset the old system also carried: a supplier with an ongoing relationship to protect if a contract went wrong. A system built to stop a procurement officer favouring a friend was never designed to catch the different failure mode Carillion represents — a supplier winning by bidding a price it cannot actually sustain.
Why the same culture survived Carillion
Knowing where the rule came from does not explain why it survived a failure this visible and this expensive. Part of the answer lies in a mechanism the political scientist Christopher Hood has documented in detail: blame avoidance. Hood's research finds that officials and politicians consistently weight the risk of blame for a visible loss far more heavily than the credit available for an equivalent, harder-to-see gain — a pattern he terms “negativity bias” (Hood, 2010). Choosing the cheapest compliant bid is, by this logic, the defensible decision regardless of whether it is the best one: if that supplier later fails, as Carillion did, the official who chose them can point to a transparent, rule-following process. An official who instead chooses a pricier bid, judging it more resilient, has no equivalent shield if that judgement is ever questioned — even when the judgement was sound.
Public scrutiny and political incentive both pull in the same direction. A minister announcing efficiency savings on a contract has a straightforward, popular story to tell; a minister explaining why a department paid more than the lowest bid, in order to reduce a risk that has not yet materialised, has a much harder one, and a press ready to call the extra spending “waste” regardless of the reasoning behind it. None of this required government to have learned nothing from Carillion's own collapse. It only required the incentives that made “cheapest first” the safe institutional choice in 2017 to still be in place afterwards.
Did the lesson stick?
Government's own answer, at least on paper, was yes. In February 2019, a little over a year after Carillion's collapse, the Cabinet Office published the Outsourcing Playbook — eleven new policies intended to ensure that departments “get right more projects from the start,” engage a healthier and more diverse marketplace of suppliers, and allocate risk more sensibly between government and contractor (Cabinet Office, 2019). The Playbook was explicitly framed as the government's response to Carillion, and to longstanding concerns about the financial fragility of the wider outsourcing sector.
Just over a year after that, the Institute for Government checked whether the new guidance was actually being followed. Its March 2020 assessment found that it was not, consistently: some government departments had not even updated their own internal policies on procuring outsourced services to reflect the Playbook's requirements (Institute for Government, 2020). A formal set of rules written specifically to prevent a repeat of Carillion existed, in other words, alongside continuing evidence that parts of government had not yet adopted it.
None of this proves that lowest-price procurement is always the wrong approach, or that every underbid contract ends in collapse. What Carillion's case does establish, on the government's own evidence and Parliament's own conclusion, is that a procurement culture built around price first and resilience second produced a failure serious enough to cost billions of pounds across taxpayers, pension savers and small suppliers. And the blame-avoidance logic that made “cheapest first” the safe institutional choice before Carillion collapsed did not disappear once the company did. A written Playbook that goes unfollowed is not evidence that the lesson wasn't understood. It is evidence that understanding a lesson and having a reason to act differently next time are not the same thing.
References
Cabinet Office (2019) The Outsourcing Playbook: Central Government Guidance on Outsourcing Decisions and Contracting. London: Cabinet Office.
Council Directive 71/305/EEC of 26 July 1971 concerning the co-ordination of procedures for the award of public works contracts, OJ L 185, 16.8.1971.
Fortune (2018) What You Need to Know About the Collapse of Carillion, a U.K. Construction Giant, 15 January 2018.
Harling, P. (1995) ‘Rethinking “Old Corruption”’, Past & Present, No. 147, pp. 127–158.
Harling, P. (1996) The Waning of ‘Old Corruption’: The Politics of Economical Reform in Britain, 1779–1846. Oxford: Clarendon Press.
Hood, C. (2010) The Blame Game: Spin, Bureaucracy, and Self-Preservation in Government. Princeton: Princeton University Press.
House of Commons Library (2026) The Collapse of Carillion [research briefing, CBP-8206]. London: House of Commons Library.
House of Commons, Business, Energy and Industrial Strategy Committee and Work and Pensions Committee (2018) Carillion, Second Joint Report of Session 2017–19. House of Commons.
Institute for Government (2020) Carillion: Two Years On. London: Institute for Government.
London Business School (2018) Two Lessons from the Failure of Carillion. London: London Business School.
National Audit Office (2018) Investigation into the Government's Handling of the Collapse of Carillion. London: NAO.
National Audit Office (2020) Investigation into the Rescue of Carillion's PFI Hospital Contracts. London: NAO.
The Pensions Regulator (2018) The Carillion Group: Regulatory Intervention Report. London: The Pensions Regulator.
Public Administration and Constitutional Affairs Committee (2018) After Carillion: Public Sector Outsourcing and Contracting, HC 748, Seventh Report of Session 2017–19. House of Commons.
Topics: #Carillion #PublicProcurement #Outsourcing #PFI #GovernmentContracting #PensionProtectionFund #NAO #PACAC #BlameAvoidance #OldCorruption