Technical Newsdesk: March 2026

Technical Newsdesk: March 2026
20 February 2026

At the time of writing this introduction (in mid-February), many of our members will be breathing a sigh of relief after working hard to meet the self-assessment deadline for clients. HMRC have published their annual press release giving an overview of filing rates, including the following statistics:

  • 12,029,168 self-assessment returns were expected.
  • A total of 11,489,825 returns were received by 31 January (including voluntary returns).
  • 10,957,825 of the expected returns were received by 31 January (91.09% of returns, following adjustments).
  • An estimated 1 million missed the deadline.
  • 97.25% of returns were filed online, following adjustments.

What do these statistics tell us? Over the last ten years, the number of returns filed online has steadily increased. In 2018, HMRC reported that 92.5% of returns for 2016-17 were completed online. By 2025, this had reached a high of 97.36% (for 2023-24 returns). This year, there has been a small decrease in the percentage of returns filed online – but at 97.25% it still shows that the vast majority of people prefer to file online. It is also interesting that this figure has been achieved without mandation, suggesting that people want to file online and that it benefits them compared to completing a paper tax return. This contrasts with HMRC’s decision to mandate people into Making Tax Digital, a system they say will also provide benefits to those in scope.

Another other statistic worth highlighting is the rate of expected returns received by 31 January. We saw a slight dip in this figure in January 2024 (2022-23 returns) to 90.48%, which equated to nearly 180,000 fewer returns filed on time compared to January 2023. It remained at a similar level in January 2025 (2023-24 returns) at 90.53% but appears to have returned close to its 2023 levels (91.35%) this year.

My predecessor, Richard Wild, reviewed the 2024 figures and acknowledged that there was no way of knowing what caused the fall from 2023 to 2024, but speculated that it could be linked to average call waiting times. In early 2024, HMRC’s average call waiting time was over 22 minutes and helpline restrictions were in place during the self-assessment peak. By November 2025 (the latest call handling statistics available at the time of writing), the position had improved, with the average speed of answering around 13 minutes. The percentage of ‘telephony adviser attempts handled’ also improved, from around 53% in April 2024 to around 85% in November 2025 – which suggests that call waiting times may indeed have contributing to the earlier fall in returns.

Some of you may have seen the confusion around the service HMRC offered on 31 January. Initially, HMRC said they would not open the self-assessment helpline on that day, instead intending to support people entirely via web chat and a call-back process for those needing extra support. However, a few days before the deadline, HMRC reversed this decision and confirmed that they would provide a limited telephone service. The figures show that HMRC advisers handled 5,409 webchats and 10,483 calls to the main helplines on 31 January – demonstrating that the phone service was needed.

What these statistics do not tell us anything about, however, is the accuracy of the returns received. The number of people having to interact with HMRC continues to increase due to the freezing of allowances and thresholds. HMRC have an ambition, by 2030, to become a digital-first organisation, with at least 90% of interactions with customers and intermediaries taking place digitally. One concern I have is about the quality of those interactions. In trying to reach this 90% target, we are seeing the deployment of various initiatives by HMRC that push people away from speaking to a HMRC adviser in favour of self-serving digitally. For example, callers to the helplines may be ‘deflected’ – having their call ended and receiving a text message with a link to GOV.UK guidance instead.

There are two possible consequences of this that cause me concern. First, those who need extra support due to a vulnerability may be unable to reach it (HMRC’s extra support teams are excellent when an individual gets through to them). Second, people may give up and do what they think is right – which could affect the accuracy of what they submit. It can also mean that they do not take the required action or take the wrong action if they have misunderstood guidance – something which may have been prevented in a conversation with a helpline adviser. It is important that HMRC analyse the statistics properly, do not jump to conclusions or make assumptions, and carefully assess the impacts on compliance of their digital initiatives.

Finally, it would be remiss of me not to mention the Finance Bill. This month’s Technical Newsdesk highlights several Finance Bill submissions made by CIOT’s technical team, CIOT’s Low Incomes Tax Reform Group, and ATT. The teams have worked incredibly hard to produce these detailed briefings, which have been quoted numerous times in the Finance Bill debates in Parliament. Well done to all staff and volunteers involved in that incredible effort. A full commentary on the measures, including mentions of CIOT and ATT, can be found on the CIOT’s External Relations team blog.


Victoria Todd Head of Technical [email protected]