HMRC’s analysis of non-compliance in R&D tax is out!
Share this article
This week HMRC published their Annual Report and Accounts: 2022 to 2023. Included within this is HMRC’s official estimate of the level of error and fraud within R&D tax relief. They’ve increased this to 16.7%, or £1.13 billion—a huge increase from the previous figures of 3.6%/£336m for 2020-21.
Clearly, HMRC and the government are not happy about this, and HMRC have produced a document detailing how they came to these figures and what they plan to do about it, which makes for interesting reading!
We’ve written at length about the changes HMRC is making to tackle fraud, as well as their recent approach to enquiries, but what’s interesting about this article is both its use of statistics and a noticeable change in HMRC’s tone and attitude in relation to R&D tax.
First of all, let’s look at the positives, the biggest of which is that change in tone. The document is much softer in its language than we’ve seen recently from HMRC, and could even been seen as conciliatory. HMRC acknowledges that the vast majority of the error and fraud in the scheme is unintentional and tends to come from technical misunderstandings, which is far more reflective of the experience of those working in the sector.
We’re also pleased to see HMRC acknowledge that there is a place for ‘reputable, professional’ agents in R&D tax relief, even going so far as describing them as playing a vital role in the regime! Their commitment to identifying and driving out those agents that seek to abuse the reliefs is also very welcome.
Having said all of that, however, we can’t help but be slightly disappointed in HMRC’s presentation of the results from the mandatory random enquiry programme (MREP). This programme audited 500 SME claims, looking at levels of compliance, and found that 24% of SME claims were either partially or wholly non-compliant. While this seems fair, the comparison with the level of fraud and error in the RDEC scheme (3.6%) and the conclusion that SME claims are massively more likely to be non-compliant seems unfair given that RDEC claims were not included in the MREP.
Alongside this, HMRC conclude that the smaller the claim, that more likely it is to be non-compliant, a conclusion that is borne out by the data. However, claims with less than £100,000 of expenditure (53% of all claims) only accounts for 19% of the cost of non-compliance, suggesting to us that measures to prevent these small claims being submitted will not reduce the cost of error and fraud as significantly as, for example, improved compliance checks across the board.
Overall, we welcome HMRC’s commitment to improved compliance, and look forward to seeing how the recent legislative changes along with better education for claimant companies affects the level of observed error and fraud in R&D tax relief.