Software R&D tax credit claims – a moving feast?
We’ve known for some time that the UK government is looking to address what they call ‘grey areas’ in software R&D tax credit claims. Some of their concerns seem to run contrary to the spirit of the R&D tax credit scheme.
While HMRC says all R&D reports no matter what the industry should have the same content, they have issued new draft guidelines for software claims. In this article I look in a bit more detail at the apparent direction of travel.
The crux of the issue appears to be that HMRC are concerned that entire or large parts of software application development costs are being included in R&D claims, whether or not all elements of the development involved qualifying R&D, as defined in UK R&D tax credit guidelines. And HMRC can’t really make a judgement about the efficacy of these R&D claims because supporintg reports tend to focus on the functionally being created and not the technological complexities and advances that underly the developments.
It’s also quite possible that HMRC want to narow the boundaries to limit software claims tightly to the parts of software applications that include innovation rather than the development of a complete prototype application. If this is the case it is questionable as to whether they are treating software claims on a par with non software claims - in the world of physical sciences it's generally accepted that the costs involved in building a prototype fall within the scope of R&D.
The R&D guidelines (CIRD 81900) say that:
“The design, construction, and testing of prototypes generally fall within the scope of R&D for tax purposes.” (Item 39)
“The guidelines apply equally to work in any branch or field of science or technology (paragraphs 15 18). This means that work in software engineering, for example, is subject to the same fundamental criteria for being R&D as work in textile science, or nanotechnology, or anything else.” B1
“This equality also applies to the methods used to resolve scientific or technological uncertainty. … As long as the scientific or technological uncertainty cannot readily be resolved by a competent professional working in the field, hardware and software methods are both equally R&D in these circumstances.” B2
One problem is that there is more ambiguity as to what constitutes a prototype in software development - is it the entire software application or is it only the modules of a software application that were technologically ‘innovative’?
If there is only one piece of functionality that requires prototyping because it is the only technologically innovative element within a much larger software development project, most would probably agree that the correct approach would be to include only the cost of developing the ‘innovative’ element.
On the other hand if the building of a new software application involves a number of technologically innovative elements, the entire technological development might reasonably be argued to represent the building of a prototype up to the point that the application works as planned, and therefore it would all fall within the scope of R&D.
Additional complexity arises because software applications are not static. There is commonly a constant stream of incremental improvements being made to existing applications, some of which will involve R&D and some of which will be routine. In this case it seems reasonable that only the ‘innovative’ elements should be included within the scope of R&D claims, but that does not mean that finding the boundary between the routine and the innovative activities is always obvious.
The types of questions HMRC now say they want answered very clearly are:
- What is the existing technology/computer science baseline?
- What was it about technology/computer science that made it uncertain it could be made to do what the company wanted it to do?
- If it could do something else at the end of the R&D project, what was it?
- What was the ‘Non-Readily Deducible Methodology’ you use to resolve the uncertainties?
- When was the Technological Uncertainty identified? When was it resolved? (Not as in date/time but e.g. "it arose when introducing the fourth data source into the main database").
Read HMRC’s Making R&D Easier Guide here.
Read more detailed CIRD 81900 guidelines here.
Are you ready to find out if you qualify or how much your R&D claims could be worth?