Thomas Chacko and Quinlan Windle discuss the options for tax breaks in relation to spending on R&D:

• outline of scheme and changes made in 2024
• what R&D means
• can you claim when R&D is contracted out
• practical points about how to deal with enquiries bearing in mind recent case law

1:25 outline of the scheme pre the Finance Act 2024 - basic deduction and 2 further possible schemes (relief for small and medium enterprises + R&D credit).

2:54 effect of Finance Act 2024

3:23 what is R&D? A common question as well as major issue in many of the cases.

8:03 the effect of HMRC now taking very robust stance on R&D tax credit claims and requiring cases to pass a very high threshold. Issues with highly technical areas eg software. Tax payers must be able to explain highly technical projects to HMRC/tribunal in a way that enables them to understand why the work merits the tax credit.

18:40 restrictions on qualifying expenditure in recent case law and following the Finance Act 2024.

37:09 practical problems in enquiries and litigation. Also the use of ADR.

Citations

Legislation and guidelines

Corporation Tax Act 2009

Finance Act 2024

BEIS Guidelines - latest update March 2023 (originally the DTI Guidelines)

HMRC Guidance - CIRD84250

HMRC CIRD161000

Case law

Hadee Engineering Co Ltd v Revenue & Customs [2020] UKFTT 497 (TC)

BE Studios v Smith & Williamson Ltd [2005] EWHC 1506

Flame Tree Publishing v HMRC [2024] UKFTT 349 (TC)

Get Onbord v HMRC [2024] UKFTT 617 (TCC)

Quinn (London) Ltd v HMRC [2021] UKFTT 437 (TC)

Collins Construction Ltd (TC09332) - 21 October 2024

Stage One Creative Services [2024] UKFTT 1059 (TC)

Welcome to Pump Court Tax Chat podcast series brought to you by the barristers at Pump Court Tax Chambers. We are the largest UK tax set and our podcasts aim to give you insight into recent developments in tax law and procedure. Across the series, we'll take a look at a range of topics covering issues that arise both in disputes and in advising clients more generally.

All of the cases and the legislation mentioned in each podcast are set out on our website in the listing for that particular recording. If you have any questions, please do contact one of our staff team. You can find their details on our website, PumpTax.com.

Welcome to the Pump Court Tax Chat podcast with me, Quinlan Windle.

And me, Thomas Chacko. Today we're talking about beneficial tax treatments available for expenditure on research and development.

I'm going to start by briefly outlining the scheme and discuss the changes made in 2024. We're then going to talk in more detail about what R&D means, when R&D is contracted out, and some practical points about how to deal with enquiries and practise.

There was a relief for small and medium enterprises, which was described as a relief, a detailed and meticulously drafted code, with a series of defined terms and composite expressions and a large number of carefully delineated conditions, all of which have to be satisfied if the relief is to be available. So it's a fairly complicated set of conditions that if you met, gave a very generous relief, and an additional 130% of expenditure on top of the 100% you already got. And that generosity combined with HMRC's tendency to pay first and then check later seems to have resulted in a lot of ambitious claims being waved through, as well as an amount of what appears to be outright fraud.

For small and medium enterprises that didn't qualify for that relief, that didn't meet all the conditions, and for large enterprises, there was also an R&D expenditure credit as an alternative. A company may be able to use the R&D expenditure credits to offset tax liabilities or to claim a payment. The amount of the credit is a percentage of the qualifying R&D expenditure and in outline, the credit is dealt with in seven steps in which it must or may be used to cover tax liabilities of the company or another group company within the remaining credit being payable to the company in certain circumstances.

One of the main questions that has come up in a lot of investigations when people are claiming R&D allowances and in the recent cases is the core question for all of this, which is what is research and development? Because that's the primary test that gives you relief or doesn't give you relief. And that's not set out in legislation. Legislation incorporates guidelines that are called the BEIS Guidelines because of the government department they came from when they were first issued, which are regularly updated and the most recent set was updated in March 2023.

And they define research and development as a project that seeks to achieve an advance in science or technology. And that's said to be an advance in the overall knowledge or capability in a field of science and technology. And you've got to do that by resolving scientific or technological uncertainties.

And if you stop there, this would sound like it's a very, very high threshold that what you're talking about is something that pushes the whole field forward. And one of the problems at the moment which people are encountering in enquiries is the revenue very much seem to take that as the sort of, touchstone, for the whole of the guidance, that whatever the rest of it says, you've still got to be pushing the scientific field forwards. But when you go into the detail of the guidance, they clearly don't envisage what you might think of as groundbreaking, entirely novel research. This is not a relief which is only supposed to apply to like Amazon and Google. This is supposed to be quite a wide ranging relief. Indeed, one of the examples they gave of a qualifying project is making a slightly improved DVD player, which almost certainly wouldn't involve some radical new development in the field of optics. And you get this when you read the rest of the guidelines.

So overall knowledge is actually defined, it's defined in paragraph 20, as it's the knowledge or capability which is publicly available or readily deducible from publicly available knowledge. By a competent professional working in the field. And so it's kind of the ordinary scale of what most people working do. But when you go into particular aspects, the threshold seems to get more and more tentative. So you should be making an appreciable improvement to an existing process.

And an appreciable improvement just has to make something better than the original, and should be more than a minor or routine upgrading. That's paragraph 23, something which would be acknowledged by a competent professional working in the field as a genuine and non-trivial improvement. So these words, it's not minor or routine, it's not trivial.

You seem to be getting a very long way from those opening comments about whether or not you've actually advanced a whole field of science. It's also made clear quite early in the guidelines that the research doesn't have to be successful, so that you may not actually achieve what you're trying to do, but you could still be doing research which qualifies. And it's also the case you can be trying to do something which already exists.

So they say, well, if the advance has already been made, but the details are not readily available because it's a trade secret from one of your competitors, for example, you can still do research to find it out. And so that shows that what they mean when they say you're driving forward a field of technology, it's not that you're kind of driving it forward in a fundamental way, because it can be a technique that you know has been in use for several years. You're driving it forward compared to what the kind of public professional baseline is.

And this is actually a change from some of the older guidelines. This is a relief which is now more than 20 years old. And they were originally DTI guidelines, and they used higher language. They used the phrase breaking new ground. And that's important. If you go to some of the even quite recent cases, like Hadee Engineering in 2020, they cite older case law which was using that breaking new ground phrase.

So as Quinlan said, there's been a series of essentially scandals about the revenue very quickly granting claims, often very low value claims, but large numbers of them for taxpayers who would never have qualified. I mean, it's things like plumbers and scaffolders, who are essentially encouraged to put in claims on a very, very thin basis. And there's been a great swing against that.

And what that has meant is the revenue are now being very reluctant to approve claims, even where what seems to be being done is clearly highly scientific, highly technological. And there's a particular issue in software, because it genuinely can be difficult with a big software project to know whether this project, which is taking three years and involves 50 engineers, is just very big and complicated and difficult, but isn't doing anything fundamentally new, or whether the reason it's taking so long is because it is doing something fundamentally new. The guidelines, as I've explained them, do present the revenue of real difficulties, because the revenue obviously do not have great internal panels of people who are at the cutting edge of any given field of technology.

They do now have an internal software panel, they refer software questions to in some circumstances, but it's still the case. It is an enormously diverse area and working out what is and isn't straightforward to a competent professional is very, very hard.

And that's an internal panel, is it? They've not got any sort of arrangements to consult actual sort of professional experts outside of the revenue?

I'm not aware of any such arrangements. There is an internal panel that they say that they make references to for software. I'm not sure what they do in the physical sciences in that area. I haven't seen a reference. A lot of this is just what comes up in enquiry correspondence, or whether they say they've made an external reference. And I've not heard of any cases kind of prior to litigation where they're making reference to an external expert.

The issue of getting external experts is obviously relevant when you come to tribunal, because if you look at the guidelines and the legal test kind of on first blush, you'd think these are cases which would always be determined by expert evidence, because I mean, how else are you going to inform the tribunal sufficiently as to what the current professional standard in any field is? But even though it's very generous relief, the claims are usually not of the scale that you'd litigate using two sets of independent experts.

You're normally talking about claims in this kind of mid hundreds of thousands of pounds. And that just isn't the sort of scale at which people comfortably or the revenue comfortably will go and get an independent expert. So until a couple of years ago, these looked like cases where the tribunals made it very, very hard for a taxpayer to win.

Because the question is, is this beyond the sort of current existing standard of a reasonably competent professional? And without calling expert evidence, how are you supposed to show that? And it's true that when you look at look at all those cases, and the most recent one is Flame Tree Publishing, which was just over a year ago, it would also seem that taxpayers have produced very little evidence at all. Often it would be the people who ran the business, but they wouldn't have actually called any evidence from someone who actually was involved in the science. And this was probably related to the fact that the revenue were being quite willing to approve plausible, in some cases quite implausible claims.

Because if you see a field where the revenue always win, one of the reasons for that can be the revenue actually being quite willing to concede strong cases. So it's only the really weak cases that end up in front of the tribunal. And now what we've seen in the last year is because the revenue are fighting stronger cases, the taxpayers are starting to win.

The most obvious case for that, which happened about a year ago, is Get Onbord. So the taxpayer in that case is developing an automated AI enabled process for know your customer verification. And they explained there was no existing technology that could provide this function.

And that there were a series of technical difficulties, such as trying to accurately track individuals of a unique identifier, but without any human involvement, and deriving information from multiple publicly available sources, with the main aim being that it was faster than the existing systems of know your customer. And the revenue maintained that this wasn't good enough. But the revenue defended their position on essentially quite vague ground.

They just said, well, a new function that doesn't yet exist isn't necessarily research and development. And yes, you've improved this relative to your competitors, but it might not be an improvement in the underlying technology. And the fact you wrote loads of unique code can't be enough, because all new large software projects involve loads of unique code.

And the tribunal basically said this isn't good enough. Once you've got a sort of sophisticated set of documents clearly explained, which look like this is presenting a project at the cutting edge of that field, the revenue can't just throw their hands up and say, well, we haven't really heard from independent experts. It's all a bit too difficult.

The taxpayer has not satisfied the burden of proof. At that stage, a tribunal says the revenue have to start explaining what they think was the appropriate professional standard, what in this project could have been done by anybody else at the ordinary standard of that field. It's also the case that they do push back in that case against the revenue's view as to what the threshold was.

They said, well, if you've created a new function and you've used a lot of new material that goes well beyond routine adaptation of existing material, you don't need to also show that every part of that involved new bespoke or novel elements. As they said, if complete novelty was the test, then no software project would ever amount to research and development. And that's clearly wrong.

So in that case, you're seeing both the rejection of the very high threshold the revenue tend to argue for, and an admission that the burden, at least in substance, will shift once you've put up what looks like a serious case that this is proper research. The revenue have to put a counter case up. They can't just say, well, this is all too difficult.

I don't think the tribunal can really say it understands it yet, and therefore the taxpayer loses. It still seems to be the case, and we'll come on the end about how to prepare these cases and how to respond in enquiries. It still seems to be the case that the revenue are putting forward a very high threshold in correspondence and essentially saying, at least in the software field, you need to be doing something extraordinarily radical, but also saying that if they don't really follow what you've explained to them, then you're ought to lose.

So I don't think that the revenue seem internally yet to have taken on what the tribunal has been saying. But that's certainly a very positive development for litigation.

My impression was that the judge was actually quite interested in software development and things, that it helped to have a judge who was curious about it.

I think this is very true. And in a way, it's a general fact about litigation, and particularly litigation where there are facts involved, that if, for the taxpayer, you can get the judge interested in what the client is doing, it tends to make them more sympathetic. I mean, in this field, it's obvious why that's relevant.

If you can get the judge interested in the project, they're more likely to take on board what your witnesses are saying about why it's difficult, why it's novel. And if the revenue can't really say anything about why it isn't novel, then you'll probably win, at least on the research point. But I think this applies to a lot of litigation, that it's always worth having enough material to kind of put real life and flesh on your case and what the witness is doing, what their business is. It just makes the case a little bit more about real people doing real things, which I tend to think helps you, certainly, if you're representing a taxpayer.

I suppose it's partly just as well whether, and I don't know if this is an issue that people have already dealt with in AI, whether some of these cases you benefit from having a judge who actually has some technical background or interest or understanding, that some of the research is so sort of inaccessible. I wonder how the tribunal manages that.

Well, it's a very difficult issue. And I think it's difficult, particularly in software and, I mean, a lot of projects now will at least say that they're using AI, because these are things that the judge will have heard of, but will have no real sense as to what the boundary is between using existing techniques and adapting or developing new ones. Because if you aren't actually in the field, that's not a clear boundary at all.

And this is the problem with legislation in some ways, that it presents this extraordinarily difficult question, both to the revenue and to judges. You can see why the revenue find answering it difficult and that their default answer, if they don't feel that they're on top of the material that the taxpayer has explained to them, they will say no.

I suppose this is the recourse to the burden of proof point you've made.

Yes. And the tribunals seem to not be going down that line completely. I mean, they still could quite easily say no if they really think that the material put in front of them has been confusing.

And this does make these difficult cases, I think, and you'd be very sort of meticulously prepared, both when you're talking to the revenue and when you're talking to the judge, because the core question which the evidence is going to, the people who are explaining it, the witnesses who are actual experts in the field, are not often that familiar with explaining to a complete lay person, which both the revenue and the judge will be, exactly what they're doing and why it's difficult. So you probably do have to spend a lot longer figuring out how you're going to explain this evidence, what kind of examples you need to give, what comparisons you need to make to get across to somebody who probably hasn't read anything about this field of technology until they started their reading into the case the day before it starts, what the current standard is and why this is difficult.

Yeah, to the extent that's even possible.

It's very difficult.

You sort of go away and do seven years at university and come back and you'll be ready to engage with it.

You're listening to the Pump Court Tax Chat podcast from Pump Court Tax Chambers with me, Thomas Chacko.

And me, Quinlan Windle. So far we've been discussing what the research and development allowances are and what research and development means.

And now we're going to go on to some of the restrictions on qualifying expenditure that have come before the courts recently, being whether the expenditure has been subsidised and whether it's contracted out.

Under the Small and Medium Enterprises Scheme, two of the conditions that had to be met were that the expenditure was not incurred by the company in carrying on activities which were contracted out to the company by any person, and that the expenditure was not subsidised. And just a few points worth noting about those conditions. At the time, prior to 2024, what it meant for activities to be contracted out to the company was not defined in the legislation. It is now, and we'll come to that in a few minutes.

Second, when something was subsidised was defined, and it included when state aid had been obtained, the extent that a grant or subsidy had been obtained, and relevant for the cases we're about to look at, to the extent that the company's R&D expenditure was otherwise met directly or indirectly by a person other than the company. And it's worth noting that the not subsidised condition did not feature in the expenditure credits, and now does not feature in the new merge scheme.

But there were three relatively recent cases in the FTT, all of which considered the not subsidised condition, and the second and third also considered whether activities were contracted out to the company. And the cases are interesting because they show how HMRC understood the relief was intended to work, and it informs how the new definition of contracted out is likely to be applied, or at least argued in the future. The three cases are Quinn (London), Collins Construction Ltd, and Stage One Creative Services.

Both Quinn and Collins Construction concerned construction companies that did refurbishment work on big, complex, listed buildings. And they found that sometimes achieving the results that they were contracted to achieve required them to develop new and innovative construction techniques. Stage One was broadly similar, but a bit more unusual.

It was an engineering and construction firm that designed and manufactured installations for live events. The decision gives the example of it designing and manufacturing and installing a complex, bespoke, scenic design for a theatrical show comprising an automated pearl element that was flown through the theatre and opened to reveal a performer within it. And again, designing and manufacturing those sorts of things sometimes required it to develop new and innovative techniques.

In all three cases, HMRC accepted that the companies were carrying out R&D for the purpose of the legislation, but said that because it was done for a specific project for which the companies were being paid commercially, it was subsidised. And in both Collins and Stage One, that the expenditure was incurred in carrying out activities that were contracted out to the companies by their customers. And HMRC's understanding of how these conditions interacted with legislation was explained in Collins, where they argued that the purpose of the scheme was to provide very generous relief for small and medium enterprises, undertaking standalone or freestanding R&D as principles.

And these conditions were intended to prevent large companies from indirectly accessing the relief by contracting out work to small and medium enterprises, and then effectively having the small and medium enterprises do the work, claim the relief, and share the benefits. And if we look first at Quinn, which only considered subsidised, in a sense, I don't know what you think about this, Tom, I can see HMRC's points, that the language of the condition was very broad. The expenditure just needed to be met directly or indirectly by a person other than the company, and the contract prices were presumably calculated to cover all of Quinn's expenditure and leave it a profit margin.

It does sound broad, I agree with that. And that had this been sort of, had it been in a different context, language like that, meeting something directly or indirectly, the courts will often interpret very widely. And so you feel Quinn is covering its costs of this research because someone else is paying for it. Surely that falls within that.

So, I mean, I think the outcome in these cases makes very good sense, because it does feel to me that all the revenue were implying by their argument, sort of walked into saying that nobody who was running a profit could ever claim his relief because somebody was indirectly funding their research.

Yeah.

But if you'd looked at that phrase on its own, the way courts tend to interpret directly or indirectly, you can see why the revenue sort of lay out a case.

Yeah. And you can see their concern as well, that if you're a big company and you look at a project and think, actually, there's probably going to be quite a lot of research involved in that, you can contract it out to a small company who can effectively do it cheaper than you because they've got the benefit of the relief.

But as you say, once you start saying that using profits or income from commercial work is indirectly meeting your expenditure, it is difficult to see where to stop. And as the taxpayers argued in all three of the cases, a broad reading of that third part of the subsidised condition sat uncomfortably with the other two being state aid and grants or subsidies. It suddenly went narrow, narrow and incredibly broad.

They also concluded that the expenditure was not incurred by the companies in carrying on activities that were contracted out to them. They noted that the companies bore the financial risk. They owned the IP. There wasn't any sense in which the research was carried out on behalf of the customers. And so they said that the companies weren't being paid to carry out the research, they were being paid to carry out these projects. And the research was just a means to an end. And that wasn't what the contracted out condition was aimed at.

Subsection 2 of that section says that a person contracts out research and development if A. The person enters into a contract under which activities are to be undertaken for it. B. The activities undertaken in order to meet the obligations owed to the person under the contract include research and development. And C. It is reasonable to assume having regard to the terms of the contract and any surrounding circumstances that the person intended or contemplated when entering into the contract that research and development of that sort would be undertaken to meet those obligations.

And it's interesting to think about how this would apply to the sort of facts we've just been talking about. Suppose going back to Quinn or Collins Construction, that a building owner or a sort of head contractor looked at a part of a refurbishment and possibly even got some quotes and someone said, well, this building technique is novel, this material is novel or is very unusual. We're going to need to really look into how we can achieve the result you want.

If it then went and got a specialist contractor to carry out that research or to carry out that work knowing that research would be required, would it have contemplated that research and development was involved?

On the face of it, the answer would sound like yes. I mean, if you go back to stage one or for Quinn, if you're in a situation where somebody said, oh, we want to do this display, our normal stage managers say they don't know how to do this and not sure anybody knows how to do this. Let's go to someone like stage one who does all the really novel and innovative stuff.

Then you'd say, well, of course, it's reasonable to assume that you contemplate that they might have to do research into how to do that project. That's why you've gone to this specialist stage manager. And similarly with the restoration work, you go to someone like Quinn because you know that this is too difficult for a more standard contractor.

So it's not obvious that this new test wouldn't catch the cases that have recently succeeded before the tribunal.

But the revenue has published guidance on this provision and their view is pretty clearly that in those circumstances, this wouldn't be contracted out research and development. They say that contemplated does not indicate a minor or fleeting consideration, but a more substantial intention.

And then they go even further. Intended or contemplated, they say, goes beyond mere awareness that R&D will take place and requires a specific appreciation of what R&D will be done and therefore the ability to understand and specify that. And going on, with this specification as the sort of R&D that is envisaged, the words carry a greater weight than mere belief or even knowledge that R&D will be required.

So going back to our examples, if your production company theatre decides that it wants to put on some show, wants some aspect or some installation on the show, goes to its usual stage manager. The stage manager says, there's nothing like that. You are going to need something novel and innovative. It is going to move forward the sort of things that we have in the industry. And they go to a company like Stage 1 and say, we understand this is the case. We're not really interested in how you achieve it.

We know you're going to have to do research. Please go away and do it and bring us this finished product. In that case, they have not intended or contemplated that Stage 1 will do work as the revenues guidance seems to understand it.

Yes. So it does sound that the revenue don't think that they're reversing the outcome of some of these cases. But it's not obvious that that's the natural reading of the test. How do questions like who will own the IP and so on come into this? I think one of the issues here, and it's interesting, they've kept contracted out as a restriction, but they've dumped the requirement that you're not subsidised. And this is in the context where they're merging the two regimes. So it's no longer an issue, really, that a large company that can't enjoy the enhanced deductions might fund its research via smaller companies that can.

Is the point of this restriction just that both companies can't claim the relief at once so that you're either doing the research and contracting out the work, or you are doing the research to provide someone with services, but you're then not a subcontractor, but you can't both be claiming it as your own research so you pay someone else to do it and doing that research and claiming the relief?

Yes. The idea, essentially, is that it determines which of the two companies is able to claim the relief.

It certainly sounds like there's quite a lot of room for argument on the text. That's really what this test is talking about?

Yes, and it then creates problems in negotiations about contracts as to you possibly have to broach who claims the relief, you have to talk about how much you're expecting to spend on R&D because if it's the sort of head company, the customer, then you have to sort of apportion the contract out into the parts that are on R&D and the parts that are on manufacturing and construction, installation, all the sort of basic things.

So I think in some of those cases who owns the IP was quite an issue, right?

Yes, indeed it's taken into account as a factor in HMRC's guidance. They say the circumstances that you might consider include ownership of the IP, financial risk in undertaking the work, autonomy in how the activity is executed, means by which the R&D is ultimately to be exploited, decision-making process and the nature of the parties. So again, these are all factors that you can see are a sensible basis for deciding which of the companies should be entitled to the credit.

They don't obviously to me seem to be sensible factors to take into account in determining whether R&D was contemplated when the contract was entered into. And it probably means that at some point there will be a discussion about how purposefully this new provision can be interpreted, whether it can be interpreted to achieve what HMRC say it's intended to achieve or whether the fact that you knew that R&D was going to be required but didn't really engage in what that was or how that would factor in in detail.

I mean it's true and this probably goes into this point that while this is the new law and we've talked about the revenue guidance in the recent cases showing the revenues old approach was wrong, this section was drafted before those cases came out except Quinn.

Yes

And Quinn, the revenue maintained, was wrong. So what may be happening here is that they have a section that was drafted on the assumption that their view of subsidy and contracting out in Quinn in Collins Construction was correct, but their guidance is based on the view that the tribunals are correct. But it doesn't necessarily match what the way the section was drafted.

And the other interesting thing about this section is that it says broadly it's assessed at the time that you enter into the contract. Paragraph C, that the person intended or contemplated when entering into the contracts that research and development of that sort would be undertaken. And so there's an issue if you enter into a contract not really expecting much in the way of research and development.

As the contractor is carrying out their work, they realise that there is an issue that's going to involve quite a lot of research and development. They go back and renegotiate the contract, vary the terms, add some extra costs to cover this research and development. Does that renegotiation remake the point at which you enter into the contract? Is it a variation? How does that work?

Because in some of those earlier cases, you actually have provision in the original contract to revisit the consideration if the expected costs went up. And that would seem to say that was well, when you enter into the contract is the earlier time. And you sort of then walk into the difficult questions in cases like Cobalt as to what the real contract is when you have a head contract with a lot of stuff left to be dealt with later. So that does seem to be another area in this legislation where the test as drafted counter up some surprisingly difficult questions under general law.

Yeah. And one final point just as we're working through that paragraph C, perhaps the way you get to the revenues interpretation of how this provision works is the two words or the three words of that sort. That if it's research and development that's intended or contemplated of that sort, maybe by interpreting that quite strictly that you have to have a very clear idea what sort of research and development it is.

That's how you get to the revenues. You've actually really got to have discussed what you're going to do and what it's going to involve.

And those are circumstances where questions about risk and questions about who owns the IP start to be very relevant. Because in a case like Stage One or Quinn, even though the person who wants the work done may have a pretty good idea of what they want the research into, they don't really care how it is that you get the performer flying across the stage or in the restoration case, they don't really care how it is you hold this church tower steady while you rebuild the adjoining building. If when they get this contract, Stage One realised that they've solved a very similar problem already, or a few years later, a similar problem comes up, the client through whom they first solved it isn't going to say, well, really, you should be paying us licence fees. That was our research, because they just wanted the end product.

Whereas in cases where the client wants the technique, because that's what they're paying for, you would probably expect them to be buying the rights to use it as well as paying for it.

I suppose that perhaps fits in with one of the other examples that's used in the guidance about paying a company to construct a new machine that produces your widgets faster, because it may very well be that then to hang on to the commercial benefit of having this improved widget-making machine, the client wants the IP to stop the company going out to the marketplace and saying, we've made this advance, we can now sell this to everyone else and everyone can come and get in on the advantage. One very last point in the guidance that I think perhaps will be a problem in practise is that these sorts of discussions, this sort of working out who claims the R&D, if it's the client finding out what proportion of the contract price is attributable to R&D, are all going to be quite commercially sensitive.

The guidance says, perhaps optimistically, the revenue accepts that for contractual and other reasons, it may not have been possible for the parties to share, that is between themselves, information including about their tax positions and will not reject or question claims simply because this has not been done. It understands that the parties may be reluctant to share financially or commercially sensitive information and won't reject claims on that basis, although how that fits in with the points we've already talked about relating to HMRC's enthusiasm to resort to the burden of proof isn't necessarily an easy one for companies to solve in the moment.

Well, no, because it throws up an unusual problem at the heart of 1133 sub 2C, this question about contemplation, which is the person claiming the relief is not the person whose contemplation you're asking about, it's their client.

Yep.

And they may very reasonably say, we have no idea what our client had in contemplation about the work we're going to do apart from what they asked us to do.

Perhaps it means if you're in the position of Stage One, you actually have to ask the client what they thought, expect that a proposed contract will involve.

And with that, shall we move on to look at some practical problems about enquiries and litigation?

Yes. So as I mentioned earlier, it sounds like in this area of contracting out, acting out and subsidies, the revenue have moderated their position in light of the recent decisions, though, as Quinlan’s been saying, what their guidance actually means and how it relates to the new law isn't actually that straightforward. In the area of what the threshold is and how much material you have to show them to persuade them that you're doing proper research, the revenue still seem to be quite reluctant to grant claims or accept that something is research and development without entering into extensive correspondence.

This seems to be a bit less true in the kind of traditional physical sciences or sort of biological research, possibly because in those areas, the kind of interaction between research and the patent system and sort of what looks like research is a bit more well established. But in software in particular, I've seen a lot of very long running enquiries where the revenue will take and digest long, complex reports and then essentially come back saying, well, we're not satisfied. We don't think this was novel.

So this can be a real problem because a lot of these businesses are relying on the cash flow advantage of the credit to fund their research and to keep the business going. And in many cases, they were, until a few years ago, being granted the credit reasonably promptly. And then suddenly, the revenue open an enquiry and it proceeds extremely slowly.

One of the reasons of this, of course, being this is actually difficult information. The revenue take a long time to digest the sort of material that's being sent to them. And though they do now have an internal panel they seem to use for software, there are simply a lot of claims being made.

So what can be done about this? I think it's important to bear in mind that you're dealing with difficult and complex information that the person receiving the letter and the report is not going to be familiar with. It's true that if you have professional representatives drafting your correspondence, that will help because they're also not going to be specialists in the actual technical field. But they will have had a lot more access to the ultimate clients and their, what the revenue calls the competent professionals, the software engineers themselves, or the scientists themselves, than the revenue inspector who's reading a letter will have.

So these take a lot of careful drafting to really get stuff into layman's terms without making it sound easy. Because of course, the whole point is it isn't easy, otherwise you're not entitled to the relief. So this is quite labour intensive, unfortunately, in terms of enquiry correspondence.

If it appears to get stuck, the answer is to try to force things to move to a point where the revenue have to explain what they disagree with you about. And it's more difficult just to say, we just don't think we've had a sufficient explanation. And you can do that in correspondence, partly citing the recent cases, the comments and get on board and saying, well, look, tribunal’s has been saying you need to say what it is that you think is wrong here.

But ultimately, that means getting into a situation where they have to explain themselves to a third party. And that starting tribunal process or going to ADR, which you can do within the, during the enquiry before you've appealed. Because the advantage there is not so much the ADR itself necessarily, as the fact they will have to then put out a position statement.

And the process internally for ADR to get started kind of means that the option of just saying, we're still not sure, so no, has gone away. And you'll get a more detailed explanation of what they're not sure about and which bits of your work they think were currently achievable.

Are you aware of technically qualified or specialist mediators being involved in this sort of ADR?

I haven't heard of that. But I mean, I just thought you could ask for it, I mean.

It seems like quite a good idea that if you, in some ways, you've got a better chance that way than a tribunal of getting an independent third party involved with some background in the area, who can come in and say, well, I do actually have some understanding of this. And here's what I think the strengths and weaknesses of the case are.

Yes. No, I think that might be a fruitful road to go down. Just to sort of clarify what I said earlier, when I say not the ADR itself, I mean, the ADR itself will also be very useful.

I mean, well before you get to that, when you get the sort of proper documents from the revenue, what you've done is you've prompted them to sit down and reflect and say, well, this is our position. That itself is quite useful for moving discussion forward. And ultimately, the point is to appeal.

If the revenue aren't giving you a closure notice and aren't actually to coming to a decision, then you might need to ask for one, which itself, if you ask for one, there will still be months of delay before you're in front of a judge to determine your application or closure notice. But before that, the revenue will have had to sort of produce a formal document saying what they think about your application. And that itself gets things onto a firmer footing if they seem to have got, if you seem to have ended up going around in circles, essentially.

The trouble is none of this, this is all faster than letting the enquiry drift on, but none of it is fast compared to the timeframe that the business may expect. And so there would need to be quite a lot of careful sort of client management in that sense, explaining to them what timeframes you're talking about. One option is to apply for ADR and you can apply in the course of the enquiry.

And what you hope for from that isn't just the ADR itself in terms of the mediation, but the fact that in the run up to that, the revenue will produce a statement of their position. And this means that it can stop an enquiry which appears to be going around in circles with sort of lots of very similar letters being sent back and forth, because they will say for you and for the mediator, what it is they don't think they understand, what it is they think that is already being done in that part of the market and so can't be outside the competence of an ordinary competent professional.

I'm Thomas Chacko. Thanks for listening to this episode of the Pump Court Tax Chat podcast. It was great to delve deeper into research and development with my colleague Quinlan Windle today.

Please remember to like, subscribe and tune into the next episode.

Thank you very much for listening to Pump Court Tax Chat. We do hope you found it informative. If you would like us to cover any other topics or have any comments, please do get in touch with one of our staff team.

You can find their contact details on PumpTax.com. Please note that this content is provided free of charge for information purposes only. It does not constitute legal advice and should not be relied on as such. No responsibility for the accuracy and/or correctness of the information and commentary or for any consequences of relying on it is assumed or accepted by any member of Pump Court Tax Chambers or by PCTC as a whole.