In an article first published by Taxation on 2 September 2024, Matthew Greene and Guy Bud consider the lessons on providing the right evidence that can be learnt from a dispute between software company Get Onbord and HMRC.

It is not so long ago that research and development (R&D) disputes were relatively uncommon, certainly in the First-tier Tribunal (FTT). This has changed significantly over the last decade. As the number of taxpayers seeking to rely on R&D relief has grown rapidly, there has been a corresponding uptick in compliance activities to root out non-qualifying projects. HMRC now investigates a substantial proportion of R&D claims.

This more sceptical approach has caused consternation for research-intensive small businesses which are often dependent on receiving these tax credits. Although HMRC has a duty to ensure that the system is not being abused, the investigative process is perceived as costly and uniquely frustrating even for the most apparently clear-cut projects.

Although some of the key legal concepts in the regime can seem vague, the problem is mainly caused by the inherent factual complexity of many R&D projects and the lack of subject-specific technical expertise at HMRC.

The FTT’s recent decision in Get Onbord Limited (TC9238) seems to offer a glimmer of light to those responding to objections and arguments from HMRC which seem misconceived, irrelevant and never-ending.

Background

The case arose from a project by Get Onbord Ltd (GOL) to create an AI-enabled platform, which was intended to run comprehensive know-your-customer (KYC) and anti-money laundering (AML) checks for clients in the financial services sector without the need for human validation at each step. The software purported to offer users a range of functional benefits compared to existing processes including speed and reliability.

The company claimed an R&D tax credit on the basis that the software represented ‘qualifying expenditure on in-house direct research and development’ (CTA 2009, s 1051). HMRC queried whether the project really constituted ‘research and development’ as defined in the relevant legislation and the BEIS guidelines – Guidelines on the meaning of research and development for tax purposes.

The guidelines as in force at the relevant time emphasise that ‘R&D for tax purposes takes place when a project seeks to achieve an advance in science or technology. … An advance in science or technology means an advance in overall knowledge or capability in a field of science or technology (not a company’s own state of knowledge or capability alone).’ It also notes, in particular, that ‘the routine analysis, copying or adaption of an existing product, process, service or material, will not be an advance in science or technology’.

HMRC’s position

Although agreeing that the work done by the taxpayer (GOL) was ‘impressive’, HMRC argued that GOL had not shown that it constituted an advance in science and technology. It reviewed the evidence provided by the taxpayer but was not convinced:

‘AI, algorithms, the data manipulation and database design may have been novel and difficult to achieve, however we have had no evidence that there has been an advance to any technologies in the process. According to … the [2004] guidelines, “the routine analysis or adaptation of an existing product, process, service will not be an advance in science or technology”. We believe that this work was readily deducible for a competent professional skilled in software development and data science.’

This provided little real indication or justification for why HMRC had formed this view on the evidence. The taxpayer considered that HMRC had not completely understood the nature of the project and had not properly engaged with the technical context. For example, HMRC had argued that the project’s use of existing open-source code – a normal thing done throughout the software world – meant that it could not be truly innovative. Clearly more and more frustrated, the company provided further information and complained that ‘despite asking, they were unable to speak to anyone at HMRC who had domain expertise’.

HMRC’s final letter responded that it had ‘taken the time and have carefully gone through your response’. In-house expertise had been sought from HMRC’s chief digital information officers: although they had agreed that the project was not truly innovative, they had provided no reasoning or basis for this view. The taxpayer appealed to the FTT.

Burden of proof

HMRC did not challenge much of the evidence provided on behalf of the taxpayer, but instead simply argued that GOL had failed to discharge the taxpayer’s burden of proving (on the balance of probabilities) that the project did constitute R&D.

It is of course true that the taxpayer bears the burden of showing that their claim is valid on the balance of probabilities. This must logically include demonstrating that the actual activities should be classified as R&D. The problem here was that the taxpayer had advanced a positive case explaining why the activities amounted to R&D, but HMRC had simply refused to agree with GOL’s conclusions and gave minimal explanation or counter-argument.

Faced with this imbalance, the tribunal referred to the case of Wood v Holden [2006] STC 443. Rather than R&D, this concerned corporate residence, an area which also turns on a nuanced assessment of an often complex factual background. The taxpayer in that case had borne the burden of showing that the company’s central management and control was located in the Netherlands and that the company was therefore not resident in the UK. HMRC had sought to pick holes in the taxpayer’s case but had not itself adduced evidence to the contrary.

In a passage later approved by the Court of Appeal, the High Court in Wood observed with apparent frustration:

‘I accept that the onus was on [the taxpayer] to show that [the company] was not resident in the United Kingdom, but rather was resident in the Netherlands. … Surely at that point they can say: “We have done enough to raise a case that [the Company] was not resident in the United Kingdom. What more can the Special Commissioners expect from us? The burden must now pass to the Revenue to produce some material to show that, despite what appears from everything which we have produced, [the company] was actually resident in the United Kingdom”.’

Contrary to HMRC’s contention, the FTT held that GOL had provided evidence that its activities constituted R&D. Happily, it noted that the use of existing code would not be fatal to a claim as this would effectively prevent any software development from qualifying for R&D relief. Considering the nature of what GOL had set out to achieve and what had been done to achieve this, it was clear that the project had been attempting to realise a novel result that was clearly uncertain.

It noted that ‘creating a new function, solving a real-world problem in a new and creative way using technology, is at least an indication that there has been an appreciable technological advance’.

As a result of GOL’s evidence, the FTT considered that there must be a similar ‘shifting of the evidential burden’ in the case. GOL had provided evidence that the software was more than a routine adaptation. HMRC needed to go beyond assertion and provide its own evidence to show that this was not the case.

Practical implications

The FTT was keen to emphasise that it was not seeking to create any new principles of law and suggested that Wood v Holden involved the same evidential assessment. It noted:

‘We have reached our decision simply by asking ourselves whether, on the basis of the evidence we have read and heard, we think it more likely than not that [Get Onbord’s arguments] are correct. [Our] answer to that question is “yes”. To the extent that the shifting of the evidential burden might suggest a lower state of satisfaction (and we are not sure that it does), that forms no part of our analysis.’

This somewhat non-committal position should not disguise an important point. In tax disputes, it is typically the taxpayer that has to do most of the heavy lifting when it comes to evidence. This is unsurprising in a self-assessment tax system like ours, but it can lead to situations (like in the GOL case) where HMRC receives a significant volume of information, documents and analysis but simply decides that it is not ‘enough’.

This approach misses a fundamental point – the taxpayer is not required to provide 100% proof or even to prove its case beyond reasonable doubt. If they can readily do so, they should, but in many cases this is not realistic. Evidence can depend on human recollections, subjective intentions or a nuanced assessment of a whole range of interlocking factors. In other cases the taxpayer needs to prove a negative – for example that tax avoidance was not one of the main objects of a transaction or, in an R&D situation, that the relevant technology did not already exist somewhere in the world.

In these types of cases taxpayers need HMRC to take a reasonable stance and not hold them to an impossibly high standard of irrefutable proof.

Strongest position

What can taxpayers do to put themselves in the strongest possible position?

Firstly, it should go without saying that the taxpayer ought to gather enough evidence to establish its case. In other words, if the evidence assembled is all that is available, would it be enough to enable a tribunal judge to decide in the taxpayer’s favour?

It is worth keeping in mind the distinction between ‘evidence’ and after-the-event explanations. A contemporaneous chain of emails which explains why a transaction was undertaken is evidence; a letter sent to HMRC during an enquiry the following year which confidently asserts the rationale behind a transaction is not. If a dispute proceeds to trial, oral evidence from witnesses may be essential. This is more likely to be necessary where the contemporaneous documentary evidence is weak or contains gaps or other inconsistencies which need explaining by someone who was there at the time.

A second factor arises as a result of the way information is gathered in tax enquiries. Taxpayers often gather information and evidence in response to specific requests from HMRC. This means that the evidence in HMRC’s possession can reflect the nature of their requests rather than the overall picture. Taxpayers should consider whether it would be helpful to go beyond the confines of HMRC’s information request where necessary and proactively set out their own position together with supporting evidence. That may well involve more work upfront to begin with, but can save a significant amount of time and resources in the long run.

Going back to the GOL case, the tribunal suggested that the parties in an R&D dispute should ‘put their scientific cards face up on the table’. It noted: ‘Ideally, [Get Onbord] would have produced a single document in which it marshalled all its scientific/technological evidence, including evidence from a competent professional (which is clearly highly desirable, whether or not it is strictly necessary), and HMRC would then have replied to that document with details of its own scientific analysis and evidence.’

This is a sensible exhortation. Taxpayers should do what they reasonably can to prove their case – but HMRC, for its part, needs to be mindful of the relevant standard of proof.

You can find further information regarding our expertise, experience and team on our Tax Litigation and Resolution page.

If you require assistance from our team, please contact us.

Subscribe – In order to receive our news straight to your inbox, subscribe here. Our newsletters are sent no more than once a month.