26.In our June 2014 report on tax reliefs we concluded that the exchequer departments did not respond promptly to unexpected increases in the costs of reliefs. We found that HMRC took time to react when it noticed a cost increase as it wanted to ensure that its response was appropriate, but that this increased the amount of public money at risk. In March 2015 we questioned whether entrepreneurs’ relief was value for money given it was costing £2 billion more than forecast. HMRC told us that the cost variation was due in large part to changes in the scope of the relief. It told us that it had no evidence to suggest that there was systematic abuse of entrepreneurs’ relief and no evidence to suggest that Parliament’s intentions were not being achieved.
27.In 2015, HMRC published a qualitative evaluation of entrepreneurs’ relief but it only interviewed 17 claimants as part of the evaluation. In 2017, HMRC published a larger quantitative evaluation of entrepreneurs’ relief which included interviews with 625 claimants. The 2017 evaluation found that at the point they invested, only 8% of claimants had been influenced by the relief. The Chancellor referred to the results of this evaluation when he announced a major reform of entrepreneurs’ relief in the 2020 Budget. Over the period April 2015 to March 2019 entrepreneurs’ relief cost £11 billion. We asked the exchequer departments why it had taken them so long to conclude on the value for money of entrepreneurs’ relief. HMRC responded that it had published information on the cost of entrepreneurs’ relief which had informed the public debate and that it was a matter for ministers to decide which tax reliefs to amend. HM Treasury said that it required careful preparation to change a tax relief. It also told us that the rules covering entrepreneurs’ relief had been changed following a review in 2017.
28.Research and development tax relief for small- and medium-sized enterprise is designed to support companies that work on innovative projects in science and technology. The cost of the research and development relief for small- and medium-sized enterprises increased from £0.8 billion in 2014–15 to £2.2 billion in 2017–18. This increase was faster than any published forecasts suggested. The NAO found that the higher costs of the research and development relief for small- and medium-sized enterprises was in part due to abuse and poor-quality claims. The relief has suffered from abuse by companies with a minimal UK presence. This abuse started after a control was lifted in 2012 which had sought to ensure that companies claiming payments through the research and development scheme had not been arranged solely for this purpose. HMRC identified that companies were exploiting the removal of the control in early 2018, and the government is now planning a further change in legislation to address this. It has proposed introducing a new control capping the payments a company can receive in a year. The cap will be linked to the amount a company pays in PAYE and National Insurance Contributions.
29.The cap was due to be introduced in April 2020, which would have given companies until 2022–23 to make claims under existing rules. HMRC estimated that the proposed changes would have saved the Exchequer around £45 million a year. However, at the 2020 Budget the government announced that it would change the design of its proposed cap to minimise the impact on legitimate business. It also announced that introduction of the cap would be delayed to April 2021. We asked HMRC how much tax will have been lost to abuse before the new arrangements kick in. HMRC estimated that the amount of money claimed through the relief which would have been excluded by the latest proposed cap was £70 million in 2016–17 and £130 million (forecast) in 2020–21. It explained that the estimate covered both claims that HMRC could challenge as abusive, and other claims where the exchequer was subsidising companies not doing research and development in the UK and which the cap would restrict. It also explained that its estimate did not represent the impact on the exchequer as it did not take account of how companies’ behaviour would have changed in response to the cap. HMRC did not provide a separate estimate of the amount of tax lost.
30.The main cause of lost tax on the research and development scheme for small- and medium-sized enterprises is from poor quality claims, which has been an issue since the scheme was introduced. We raised the issue with HMRC of the higher than expected number of claims for the relief and the role of agents in making these claims. HMRC said it was a relief that it had to police quite carefully. It explained it faced two issues. First, some businesses tested the boundary of the relief by arguing that their expenditure was within the scope of the relief. Second, agents produced some claims, but these were not to the standard that HMRC would have liked. HMRC told us that it had confidence in how it administered the relief, and that it placed a great deal of resource and attention into making sure that claims were right. HMRC set out the actions it had taken which included:
50 Committee of Public Accounts, Tax reliefs, Third Report of Session 2014–15 Report, together with formal minutes related to the report, HC 282, June 2014
51 Committee of Public Accounts, The effective management of tax reliefs, Forty-ninth Report of Session 2014–15, HC 892, March 2015
52 IFF Research, , February 2015
53 IFF Research, , May 2017
54 Budget Speech 2020, available at
55 C&AG’s report, Figure 12
56 Q 63
57 HMRC, , March 2020
58 C&AG’s report, paras 2.20–2.27 ad Figure 12; and HM Treasury, Open consultation – Preventing abuse of the R&D tax relief for SMEs: second consultation, March 2020
59 C&AG’s report, para 2.27; HM Treasury,
60 Q 44
61 Letter from Jim Harra, Chief Executive and First Permanent Secretary, HMRC, 26th June 2020
62 Q 34, C&AG’s report, para 2.27
63 Letter from Jim Harra, Chief Executive and First Permanent Secretary, HMRC, 26th June 2020
Published: 20 July 2020