Nationwide refunded 85% of fraud losses – how does your bank compare?

The worst banks for reimbursing bank transfer fraud victims and preventing criminals from abusing their systems have been revealed – how does yours stack up?

The Payment Systems Regulator (PSR) has published the third set of industry-authorised push payment (APP) fraud data from the payments industry. 

The latest figures cover 2024 up to the start of the mandatory reimbursement scheme, which took effect from 7 October 2024, following years of Which? campaigning for fairer outcomes for APP fraud victims. 

Read on to find out which banks were most and least likely to refund scam victims, and why Which? believes this data must continue to be published. 

•  Select Which? as a preferred source on Google so you can see more independent news written by expert Which? journalists.

Which banks are worst at refunding fraud victims?

Banks and other payment firms report authorised push payment (APP) fraud data – also known as bank transfer scams – to the PSR. 

Nationwide topped the 2024 reimbursement table, reimbursing 85% of APP fraud losses to customers, followed by TSB (80%) and Barclays (77%).

Victims who bank with AIB (31%), Monzo (25%) and Danske Bank (23%) were least likely to be refunded. These same three banks reimbursed the fewest cases overall, for example, Monzo only fully reimbursed 18% of claims.

However, customers of AIB and Danske Bank were also the least likely to fall victim to APP fraud in the first place. For every million customer transactions AIB sent, only 30 were reported as APP scams payments, compared with 131 at Starling and 117 at Monzo. 

Which payment firms receive the most fraud?

Reimbursing innocent victims is only half the story – payment firms must stop criminals from abusing their systems to move stolen funds, including mule accounts. 

The PSR’s league table lists many names that won't be familiar to the general public because they provide services behind the scenes or offer business accounts.

PrePay Technologies (also known as PrePay Solutions) was the worst in 2024, when for every £1m received into customer accounts, £3,132 was linked to APP scams. It provides services for Monese, a multi-currency banking alternative. 

More familiar brands include Revolut, which is yet to obtain a full banking licence in the UK; TSB; and Wise, which provides international transfer services. All three received disproportionately higher levels of APP fraud money than many high street banks.

Make money make sense

Make every penny count with expert, impartial advice for just £49 a year

Join Which? Money

Why victims need the reimbursement scheme

These figures make it clear that APP fraud victims were facing wildly inconsistent outcomes before the introduction of a mandatory scheme in October 2024.

Payment firms also needed stronger incentives to prevent fraudsters from using their accounts, something the scheme aims to fix by forcing sending and receiving firms to share the cost of reimbursing customers equally.

The PSR is yet to publish firm-specific data on APP fraud reimbursement under the scheme. However, aggregate data shows that 88% of the money lost to APP scams was returned to victims, up from 60% before the mandatory scheme was introduced. Since 7 October 2024, the PSR says: 

• 82% of claims were closed within five business days, meaning consumers are generally receiving timely responses
• consumers reported around 269,000 claims, of which 188,000 were in scope for reimbursement
• only 3% of claims were rejected due to the consumer not taking enough care over the transaction or their claim.

Gaps in fraud protection remain

The scheme is leading to better outcomes overall. However, some victims are still left out of pocket because the scheme doesn't cover bank transfers to international accounts, civil disputes against businesses or other payment methods such as plastic cards and cryptocurrency.  

We think it's important that firm-specific data continues to be published to help give confidence in the scheme and shine a spotlight on poor performers. The public should also see data on claims considered out of scope to ensure that payment firms aren't unfairly dismissing valid claims – we've previously raised concerns that banks are unfairly telling victims of complex investment fraud and rogue builders to pursue scammers in court. 

Which? made a super-complaint to the PSR in 2016 calling for better safeguards and fairer outcomes for APP fraud victims. We have also led the campaign for scams to be included in the Online Safety Act, so that tech giants will also be held accountable for scams originating on their platforms.

• Find out more: why your APP fraud claim could be rejected

What the banks told us in response

Monzo said that since 2024, before the introduction of new industry-wide rules for reimbursement, it has 'continued to invest in building cutting-edge technology and hiring industry experts to tackle financial crime head-on and keep customers’ money safe. And it’s working – we prevented 2.9 times the value of fraud in 2025, compared to the year before'.

PrePay Technologies told us it has 'implemented significant improvements, including major enhancements to its technology, monitoring and controls, which have contributed to a reduction in APP fraud levels' since the period covered by this 2024 data. It has also 'undertaken a rationalisation of its business portfolio, including a decision to gradually exit part of its "banking as a service" and "business to customer" activities.'

AIB NI pointed out that it has the lowest level of APP fraud per million transactions: 'This is in part due to the level of protection and support we have in place...Of the low level of APP fraud cases identified in the report’s period, each was investigated promptly and dealt with on a case-by-case basis. Anyone, at any time, can be the target of financial fraud. Scams have become increasingly sophisticated, and we continue to focus on preventive measures to help protect customers.'

Danske Bank is yet to respond to our request for comment.