Just nine Ministry of Defence fraud cases were referred to the Crown Prosecution Service over five calendar years, according to new Freedom of Information disclosures obtained by Action on Armed Violence.
Between 2020 and 2024, eight of those referrals resulted in prosecutions and seven secured convictions. The MoD said one case was still to be heard at court.
The figures do not represent every fraud investigation carried out across Defence. Some CPS referrals are “managed by the Ministry of Defence Police”, a statutory police force whose officers have the powers and privileges of civilian constables operating within the civilian criminal justice system. Separate service-police investigations can proceed through the military justice system and the Service Prosecuting Authority, which prosecutes cases before the service courts.
The nine referrals therefore should not be read as meaning that only nine fraud cases were investigated during the five-year period. Instead, they represent one specific outcome: cases reaching the CPS through the MoD’s civilian policing route.
Even with that qualification, the figures provide a rare indication of how few Defence fraud cases reached civilian prosecutors during a period in which auditors and MPs have raised serious concerns about the department’s exposure to fraud and its ability to investigate it.
The National Audit Office reported in January that, since 2020-21, the MoD’s annual calculations of its “potential fraud risk exposure” had produced a lowest minimum estimate of £1.1bn and a highest maximum of £1.5bn, with most of the estimated exposure arising from procurement.
The report noted that the MoD ”receives hundreds of allegations of suspected fraud or economic crime each year, but relatively few result in detection, disruption and recovery.”
The £1.5bn figure, however, is not an estimate of how much is actually stolen from the MoD each year.
The NAO said the calculation relies partly on external fraud benchmarks rather than robust data on confirmed Defence losses, which does not exist. The true loss could be considerably higher.
The NAO investigation said that applying the methodology to around £40bn of procurement expenditure could produce a maximum unmitigated fraud exposure of roughly £2bn.
Billions at risk
The sums passing through the department are substantial.According to the MoD’s latest trade, industry and contracts statistics, the department paid £40.6bn to UK and foreign-owned organisations in 2024-25, excluding Foreign Military Sales expenditure. That was £3bn more than the previous year. 45% of MoD Core Department payments were made through non-competitive sourcing, excluding Foreign Military Sales.
During 2024-25 the MoD considered around 1,700 allegations of fraud and economic crime, including cases carried over from previous years. Around 60% related to personnel issues such as pay, expenses and allowances, but those cases accounted for only around 2% of the estimated value of alleged losses.
Procurement, however, represented about 18% of allegations but more than 95% of the estimated value.
This points to a particular concern: while personnel-related allegations make up the majority of cases, the largest financial exposure appears to sit within procurement.
Fragmented Investigations
The NAO also identified serious weaknesses in the way suspected fraud cases are recorded and followed through the department.During 2024-25, the MoD’s Confidential Hotline recorded 603 cases as having been referred to defence police. Of these, 147 were recorded as going to the Ministry of Defence Police and 456 to the service police.
Separate police datasets, however, recorded 363 cases as having been investigated during the year.
The NAO was readily able to match only 75 cases between the hotline and police systems.
Auditors identified several possible reasons for the mismatch, including cases being reported directly to the police, investigations treated as intelligence rather than crimes, police failing to update Fraud Defence when they closed their cases, and inconsistent case-reference numbers. The result, they found, was a fragmented system in which the department could not easily establish what had happened to individual fraud allegations or assess their eventual outcomes.
The number of cases recorded as producing criminal or service-justice consequences was also small.
Of 1,032 outcomes recorded by the Confidential Hotline during 2024-25, 826 ended with no issue identified or insufficient evidence to proceed. A further 184 led to formal or informal action. Only 18 outcomes, around 2%, involved criminal or service-justice action.
The number of cases recorded as producing criminal or service-justice consequences was also small.
Of 1,032 outcomes recorded by the Confidential Hotline during 2024-25, 826 ended with no issue identified or insufficient evidence to proceed. A further 184 led to formal or informal action. Only 18 outcomes, around 2%, involved criminal or service-justice action.
Limited investigative capacity
There are also questions over investigative capacity.Simon Dobinson, then acting chief constable of the Ministry of Defence Police, told the Public Accounts Committee in March that the force had the equivalent of only around seven staff providing a dedicated economic-crime investigative capability at any one time.
Dobinson said investigations had recovered just over £850,000 during the previous year.
The new FoI disclosures also expose gaps in how Defence counter-fraud investigations connect with other government enforcement bodies.
Asked specifically about arrangements for handing over procurement-fraud cases to HM Revenue and Customs, the MoD said that “no formal MOU with HMRC exists for referral of tax related offence”.
It added that no cases recorded on the MoD’s Confidential Hotline had been referred to HMRC during the period covered by the request, from 2020-21 to 2024-25.
The response also said that “HMRC are not recorded in MOD counter fraud strategy”.
Similarly, when AOAV asked whether any cases had been referred to the Serious Fraud Office during the five calendar years from 2020 to 2024, the Ministry of Defence Police said it held no information.
A low return on counter-fraud spending
Permanent Secretary Jeremy Pocklington told the Public Accounts Committee’s March hearing that the department operated a “zero-tolerance policy on fraud”. Its central Fraud Defence team contained around 24 people, he said, with approximately 50 specialists working across the wider department.
The National Audit Office concluded that it was “highly likely” the amount of fraud the MoD investigates, recovers and prevents is considerably less than the loss it incurs.
The January NAO report also critiqued the cost-efficiency of the system. As it noted, “between 2021-22 and 2024-25, the MoD… spent an average of £5.7 million a year on counter‑fraud work and prevented and recovered an average of £2.8 million, of which half was fraud and half was error.”That represents an average return of only 48p for every £1 spent on counter-fraud activity., in contrast to the government’s expected return of £3 for every £1.
In May, the Public Accounts Committee concluded the MoD’s £1.5bn annual potential fraud exposure had not been addressed with the required “focus or leadership”.
The government promises tougher action
The government has since responded to the Committee’s recommendations. Its formal response was published on 4 August.
The MoD says it is tightening its counter-fraud regime. Senior responsibility has been strengthened under the Director General Finance. The department is developing a commercial fraud “playbook”, a new case-management system and a more integrated model bringing together Fraud Defence and Defence policing. It also says it plans to make wider use of data analytics and AI to spot suspicious transactions and says it intends to raise the return on counter-fraud spending to at least £3 recovered or prevented for every £1 spent by 2028.
Questions also remain over another mechanism intended to protect public money from dishonest or unsuitable suppliers: debarment.
The empty debarment list
The Procurement Act 2023 introduced a new central debarment regime, which came into effect on 24 February 2025.
Under the system, the Debarment Review Service can investigate whether suppliers meet exclusion grounds. The ultimate decision to debarment is made by a minister.
In their FoI reply,The MoD made it clear the department does not itself debar suppliers.
When asked how many suppliers had been debarred, the MoD said none had been added to the list.
More than 18 months after the regime came into force, the government’s published central debarment list remains blank.
A parliamentary question by Patricia Ferguson Labour (MP for Glasgow West) tabled on 14 July asked the Cabinet Office how many active cases were being considered by the service.
As of early September it was still recorded by Parliament as awaiting an answer, despite having been due for response on 16 July.
A system struggling to keep up
For AOAV, the concern is not just that public money may be being lost, It is that it reveals a system through which increasingly vast sums are being channelled into defence without much in the way of accountability and process.
At a time of expanding military budgets, lucrative contracts and substantial profits across the defence industry, the MoD itself acknowledges potential fraud exposure running into the billions. Auditors have found fragmented investigations, weak recovery rates and limited criminal outcomes.
The danger is one of accumulation. The more money that flows through a system marked by inadequate scrutiny, the greater the opportunity for abuse — and the greater the risk that waste and weak controls become normalised.
Every pound lost through fraud, waste or misconduct is also a pound that cannot be spent elsewhere, including on diplomacy, conflict prevention, humanitarian aid and other non-violent means of addressing insecurity.
Defence is enjoying a rare high moment of political and financial favour in Whitehall on a scale not seen for years. However, the mechanisms for ensuring that money is honestly spent, properly investigated, and meaningfully accounted for do not yet appear to have kept pace.
Did you find this story interesting? Please support AOAV's work and donate.
Donate