The Public Accounts Committee has warned that the Ministry of Defence is failing to address a £1.5 billion annual fraud risk with requisite focus or leadership, calling for a radical change of culture and the appointment of a senior two-star officer to lead counter-fraud efforts.
The PAC report found that over the four years from 2021-22 to 2024-25, the MoD recovered an average of just 48p for every pound spent on counter-fraud activity, well below the government’s expectation of a £3 return for every pound spent. The department spent an average of £5.7 million per year on counter-fraud activity over this period.
While the MoD highlighted an improvement to £1.34 for every pound spent in 2024-25, the committee noted this mostly relates to a one-off exercise to identify commercial leakage, and the department could not convincingly explain how it would ensure the improvement would continue. The MoD has said it does not expect to reach the government’s expected £3 return until 2028. The committee is calling for a clear plan by November 2026 setting out how the department will achieve that level at minimum.
Sir Geoffrey Clifton-Brown MP, Chair of the Public Accounts Committee, said: “The MoD is far behind the curve in preventing the loss of precious public funds which could be spent on keeping our nation safe. This is on top of the £1.6bn that our inquiry into financial audit found had been wasted by the MoD through cancelled projects. Incremental change will not suffice. There must be a radical change of culture within the MoD if the flow of funds lost to fraudulent activity is to be stemmed. The apparent normalisation of fraud in the procurement process is symptomatic of a wider issue; there is no overarching strategy within the MoD of how to tackle fraud and economic crime. The MoD must embrace the new technologies helping other Departments to tackle fraud and error if it is to replicate the levels of success that a proper safeguarding of the public purse demands.”
The committee found that the MoD’s £1.5 billion annual fraud exposure figure is itself unreliable, described to the inquiry as an “academic construct” by the department itself. The figure is derived from external benchmarks rather than a robust assessment of the MoD’s own fraud risks and controls. The department was unable to say when it would have a reliable estimate, pointing instead to a counter-fraud strategy due in September 2026. Work to develop a more reliable estimate is expected to take around a year. The PAC further noted that the MoD’s numerous fraud risk registers do not quantify potential fraud losses, limiting their usefulness as a management tool, and that the department has not made sufficiently active use of them to direct resources to areas of highest risk.
Procurement represents the area of highest risk, making up most of the department’s estimated £1.5 billion fraud exposure and most of the value of the fraud allegations it receives. In 2024-25, the MoD stopped around £400 million of contract payments it judged to be invalid, submitted by suppliers despite the department having open-book access to their financial data. The committee said this suggests suppliers may regularly and repeatedly claim more than they are entitled to, and that the MoD does not do enough to deter such behaviour.
A 2024 exercise examining commercial leakage in digital procurement contracts found £17.5 million of potential overpayments, but the department has not extended this type of exercise across its wider procurement activity. The committee also expressed concern about the risks created when staff move between the department and its suppliers, questioning whether the MoD’s response to such commercial threats is strong enough to deter future fraud.
The report also criticised the MoD’s fragmented approach to counter-fraud, with responsibility split across several counter-fraud and police teams that operate separately and lack consistent communication or joined-up case management. The department uses multiple case management systems, meaning it has no clear overall view of case progress or outcomes. This makes it harder to verify that allegations have been properly investigated and more difficult to give feedback to those who report suspected fraud. The committee noted that this fragmented structure may create gaps that fraudsters can exploit. The MoD said it plans to improve its understanding of how investigations are carried out by sampling cases and that there may be an opportunity to introduce a shared case management system, with further proposals to be set out in its September 2026 counter-fraud strategy.
Previous reviews of defence policing have repeatedly identified siloed working, weak ownership and a lack of trust between the different parts of the department responsible for tackling fraud and economic crime. While the MoD considers some of these cultural issues to be historic and points to examples of improved collaboration, the NAO reported that this remains inconsistent. The department still lacks a single senior individual with primary responsibility for bringing together the various functions involved in responding to fraud. The PAC is recommending the appointment of a two-star individual predominantly focused on supporting the Permanent Secretary on fraud and economic crime, with a clear mandate to reduce losses and change the culture of the whole department.
On technology, the committee found the MoD has yet to capitalise on data analytics to prevent fraud occurring in the first place, with most analytical work to date focused on finding fraud after it has happened. The department intends to move checks earlier in the process to prevent incorrect payments, and plans to use technology to strengthen its control environment as part of the Defence Reform programme, but the committee said very little detail has been provided on specific techniques or timelines. The PAC is calling for the department to write to the committee within six months with an assessment of areas where data analytics could be applied cost-effectively to detect and prevent fraud, along with a plan and timelines for doing so.
The committee’s six recommendations also include developing a more robust fraud loss estimate in partnership with the Public Sector Fraud Authority and NHS Counter Fraud Authority, publishing increasingly robust estimates in annual reports from 2026-27, and developing a procurement counter-fraud playbook setting out how the MoD will investigate and take criminal or service justice action against suppliers, individuals or collusive networks who abuse contract arrangements for personal gain.
It talks about potential losses due to fraud and potential overpayments, but what about actual losses and actual overpayments? There’s evidence of contract overpayments being stopped, but what are the level of contract overpayments that aren’t stopped? How do we know there’s a big problem? Maybe the small return of attempts to recover funds lost to fraud shows these massive potential fraud losses aren’t actually real in the first place.
£1.5bn exposure to fraud (total volume of vulnerable assets *before* internal controls).
£253m alleged fraud (aggregated potential money value of investigations by MOD counter-fraud teams)
£2.3m detected fraud (with a similar amount prevented)
£0.3m recovered
Less than 2% by value of investigated fraud turns out to be substantiated.
“Cabinet Office and HM Treasury expect that public bodies should save £3 for every £1 spent on counter-fraud work.” says the report. Maybe the problem lies here. The NAO wants MOD to “maximise its return on investment”. However, MOD already has a lot of security and a lot of associated vetting. It even has its own police force. Perhaps the reason it recovers less per pound spent than other government departments is because easily detectable fraud has already been weeded out by dealing with vetted staff and suppliers?
Does anyone else wonder if the NAO are wrongly focused? Improved leadership and systems aren’t going to detect more fraud if it doesn’t actually exist. Perhaps the way to bring down the estimated losses, which is what the NAO is recommending, is to estimate less loss.
Hand it over to the MOD police not some untraine two star , or simply get the normal police to do it, clearly officer run over sight is not working, like most things in Armed Forces.
This should be run by a NAO and they should be looking into all governmental fraud from MoD to NHS with companies like Capita interwoven into most government departments like DWP, NHS, and the MoD.
Companies that straddle several departments seem to be given a free rain to do as they please with out much oversight.
Then we have the procurement fiasco and again it is not just the MoD who seems unable to procure anything at a realistic cost.
It seems to me that there is a whole culture within government that gets rich from being inept from all of the UK’s many political institutions and departments that are supposedly there to serve the people.
The size of the problem is large. MoD’s fraud detection and rmonies recovery system must be totally inept and very costly to run – more details needed.
Companies that defraud the MoD should be named and shamed and forbidden to tender for MoD contracts ever again, or at least for a very long time.
I wonder if this £1.5bn risk includes scenarios of MoD empolyees defrauding the MoD? This must surely be small beer?