The number of companies reportedly making R&D Tax Incentive claims in Information and Communication Technology (ICT) is second only to those in the ‘Engineering’ field of research. One would expect that this number will only increase as uptake by SMEs and the tech Startup community continues to grow.
Recognising this, AusIndustry has issued Guidance Material for ICT that outlines, through commentary and case study, the government’s views on what will and won’t qualify for claim. ‘Algorithms’, ‘cloud computing’, ‘GUIs’ and ‘software modules’ all get a look over. Software development is also always a hot topic in AusIndustry forums and blog posts on the topic abound from every adviser imaginable, including the ‘Big 4’ through to smaller independents.
There’s lots of confusion.
Despite all this guidance, discussion and opinion there is still a lot of confusion as to exactly what software development activities will qualify for claim. To date, we’ve tended to resist putting our two cents worth out there and have preferred to offer tailored advice on a client by client basis. However, in our travels we continue to be asked for our views on software development and R&D tax, so perhaps the time is right to offer up some thoughts.
‘Guidance’ material is exactly that…guidance.
We acknowledge that the guidance material published by AusIndustry should be factored into any discussion on the eligibility of particular software development activities. But we would qualify this by noting that it is, as the name suggests, “guidance” only that is intended to help address general questions of uncertainty and difficulty. It cannot be regarded as definitive.
Not all software is created equal.
It seems much of the difficulty with ICT/software development is that it is an area of rapid transformation. Change is embraced and adapting to new technologies is the norm. It seems that innovative methods and techniques are introduced at such a rapid rate that what requires “R&D” one day is almost a routine activity the next. There is a danger when government (and advisers) buy into this popular perception too readily, for it can lead to setting themselves up as pseudo tech experts.
This danger is realised when they then become prescriptive about what qualifies and what doesn’t based purely on false assumptions. The number of times I’ve heard another adviser – with no background or expertise in the technology – pronounce that “the development of a phone app can’t be R&D because there are so many phone apps out there nowadays and so many tools available to help build them”. Now, the development of that phone app at the end of the day might well not benchmark strongly for claim but the rationale can never be based on the fact that “there are so many of them out there now”. The same approach in respect of the cloud or mobility is equally ill conceived.
Then again, not all software is innovative.
On the other hand, another view might be that all software development is potentially eligible because the agile iterative software development methodology that is now an industry standard is inherently “experimental”. This approach is also flawed because it doesn’t pay sufficient regard to whether there were any precedents for the development and the extent to which any new knowledge was generated.
So, where does all this leave us?
Clearly, there is great potential to claim software development activities under R&D tax and the stats prove that a lot of such activity is indeed claimed. Guidance material from government can help provide a framework when qualifying such activities. Nevertheless, each development should be benchmarked against the legislative definition and treated on its merits, free from any assumptions that certain software development activities in particular areas cannot be “R&D” because they are now “old hat”.
Claimants (and their advisers) should always work from first principles and assess what activities have been conducted and the extent to which they can legitimately be described as “experimental” (remembering evidence of the journey from “hypothesis” to “test” to “analysis of results” to the drawing of a logical conclusion). And not forgetting the need to articulate a compelling argument that some “new knowledge” has been obtained from the exercise.
In this last respect, determining what is “new knowledge” and what is a relatively transparent extension of the “prior art” can be another contentious area. But that’s a topic perhaps best left for another post.
