Since 7 October 2024, UK banks and payment firms have been required to reimburse most victims of authorised push payment (APP) fraud on Faster Payments. Eighteen months of regulator data and an independent review now show what the rules have changed, and where they have not yet delivered.
What the reimbursement requirement actually covers
APP fraud happens when a person is tricked into sending money themselves, for example to a fake supplier or to someone impersonating their bank. Before the requirement, whether a victim got money back depended largely on a voluntary code and on which bank they used.
According to the Payment Systems Regulator (PSR), the current rules apply to Faster Payments APP scams. They cover individuals, plus small businesses and charities with annual income under £1 million, and they sit under a regulatory instrument known as Specific Direction 20. Payments between two accounts at the same firm, sometimes called “on us” payments, are excluded. The PSR dashboard also records that a £100 excess applies and that the maximum reimbursement is £85,000.
Sending and receiving firms now share the cost of reimbursing victims, which gives the receiving side a financial reason to police the accounts that scammers use. Claims can still be rejected, for example where a case is a civil dispute rather than a scam, or where the customer showed insufficient caution.
What does the PSR’s own data show about money returned?
The PSR’s reimbursement dashboard, updated on 30 July 2026, covers the 18 months from 7 October 2024 to 31 March 2026. Its headline figures:
| Measure | Figure |
|---|---|
| Share of money lost to APP scams that was reimbursed (18 months) | 88% (£316m) |
| Claims reported (approx.) | 438,300 |
| Of which in scope of the rules (approx.) | 301,500 |
| Reimbursed share of claim value, January to March 2026 | 89% (£72.6m) |
| Claims closed within five business days | 82% |
| Claims closed within 35 business days | 98% |
The PSR describes the first-quarter 2026 reimbursement share as the highest since the policy began. A quick check on the table shows that about 69% of reported claims were in scope (301,500 divided by 438,300), which is a reminder that the rules do not cover everything labelled a scam.
The dashboard measures how quickly claims are closed, not when money lands in an account, so the five-day figure should be read with that in mind.
What did the independent review find?
On 1 July 2026 the PSR published an independent review by Frontier Economics. Its estimates, as reported by the regulator:
- APP fraud losses are about £73 million a year lower because of the policy.
- The number of APP scams has fallen by nearly 35,000 because of the policy.
- Reimbursement rates across all claims rose from 54% to 65%.
- For claims within the scope of the rules, firms now reimburse 97%.
- After allowing for higher costs to payment firms, there is a short-term net benefit of £17 million to £29 million, which Frontier describes as a conservative assessment.
- There was no evidence of the market exits or reckless consumer behaviour that some had predicted.
The largest improvements were at firms that had the highest APP fraud levels before the policy began. David Geale, the PSR’s managing director, said: “The evidence is clear – APP reimbursement is working.” He added that the regulator was “not complacent”.
Why do the 88%, 65% and 97% figures differ?
They are not contradictory; they measure different things. The 97% applies to claims in scope of the rules. The 65% covers all reported claims, including those outside the policy. The 88% is a share of money lost rather than of claims. The PSR’s dashboard also warns that its figures are not directly comparable with UK Finance data, which counts by the date a claim is closed and covers all APP fraud, while Frontier dates losses by the scam transaction.
For readers, the practical lesson is to check which denominator a headline uses before comparing it with another. A rate of 97% of in-scope claims says little about the many scams that fall outside the rules.
A worked example of how the excess applies
Illustrative arithmetic: suppose a person loses £1,500 to a fake invoice scam paid by Faster Payments and the claim is accepted as in scope. With the £100 excess, the reimbursement would be £1,500 minus £100, or £1,400. At the other end, for a very large loss, the £85,000 cap is the ceiling on what the rules require firms to repay.
The figures show why it matters whether the payment type is in scope. A card payment, an “on us” transfer or a payment made through a different route may be handled under different rules, which is why the route of the payment, not just the scam, shapes the outcome. How those routes work is covered in how modern UK bank transfers work.
Has the rule changed fraud behaviour?
The PSR’s one-year review, published on 8 October 2025, reported 126,000 claims between October 2024 and June 2025, about 15% fewer than in the same period a year earlier. Almost 60% of APP fraud victims in its survey had been hit by purchase fraud, while delivery, impersonation and charity fraud had increased.
Its survey of 1,448 UK adults, which included a boost sample of 436 APP fraud victims, also found that half of reimbursed victims said they trust their bank more, compared with about one in three of those not reimbursed. The regulator noted that comparisons with earlier years are not fully possible because of changes in methodology.
Where are the gaps?
Three stand out from the published material.
- Awareness. In the PSR’s survey, 49% of fraud victims did not try to access reimbursement and 71% said they were unaware of the policy.
- Consistency. The PSR says implementation is uneven, so outcomes can vary by bank, and it intends to consult on improving consistency before the end of the year.
- Where scams start. The PSR has called for “a step-change in the approach taken by tech firms and telcos”, and says it will publish new data at the end of the year on which platforms fraudsters use to target victims.
Rejections for insufficient customer caution remain a minority. The dashboard puts them at about 1,700 claims, or 2%, in the first quarter of 2026, and 3% over 18 months.
How does this fit the wider payments picture?
Reimbursement is one layer of protection alongside prevention. Faster payments are hard to reverse once sent, so safeguards such as fraud warnings matter, as do the data-sharing tools covered in what open banking means for UK consumers. The convenience that has driven the shift to instant digital payments, discussed in how digital payments change consumer behaviour, also raises the stakes when something goes wrong.
Reimbursement is also separate from deposit protection, which deals with a firm failing rather than fraud, as explained in how the £120,000 deposit protection limit works.
What the evidence leaves open
The data supports the view that mandatory reimbursement has raised the share of victims who get money back and pushed firms to invest in stopping scams. It does not show that fraud is solved. Losses still run into hundreds of millions of pounds across the period, most victims in the survey were unaware of their rights, and the origin of many scams on social media and telecoms sits largely outside the payments rules.
The next test is consistency between firms and whether pressure on technology platforms reduces scams at source. Anyone who thinks they have been a victim should contact their payment provider promptly, since the rules depend on a claim being made.
This is general information, not personalised financial advice. Rules, limits and figures change, so check current details with your provider and the regulator.
Sources: Payment Systems Regulator
