The R&D Tax Incentive and Software Development: A better fit than the alternatives.

March 4, 2021 Ian Ross-Gowan

Since the announcement in the October 2020 Budget that the R&D Tax Incentive (RDTI) cuts were not going through, the attention has turned to the thorny issue of the eligibility of software R&D. Many commentators are pushing for the introduction of a separate tax concession program for software R&D in certain circumstances.

MJA Director, Ian Ross-Gowan, reflects on the wisdom of this push:

Submissions before the Selection Committee on Financial Technology  have been considering whether to tinker with the RDTI or split it into two programs, one for software R&D and one for the rest.

The options are:

1.    Make it clearer when business expenditure on software development meets the requirements to be eligible R&D under the RDTI, or
2.    Create a new program that targets just early-stage companies doing software development as their core business.

However, this is a choice between two solutions in search of a problem that has already been fixed.

The assumption behind the first point is that it is unclear when software development is eligible for the RDTI and this has uniquely exposed software businesses to financial risks on audits by the ATO and AusIndustry. This assumption is based on how the RDTI was being managed in the period following the 2016 Review of the R&D Tax Incentive  by Dr. Finkel and Messrs. Ferris and Fraser. This led to a push by the administrators to make the program more of a scientific research program and less of a business experimental development program despite the Review recommending no changes to the legislative definition of R&D.

Under this mindset, software development by businesses struggled to be seen as eligible R&D. However, this narrowing of the concept of what was considered R&D was also being applied to experimental development of new and improved products and processes by manufacturers, miners, farmers and engineers.

The solution in the second point seeks to consider software as being different from, say, physical or process engineering. It also seeks to limit the RDTI to being only for startup businesses for whom software development is their core business. This flies in the face of the program’s objective of encouraging more Australian companies to do more R&D for the benefit of the Australian economy. Once a software development company is no longer able to be considered a “startup”, they will be excluded from the unique program and, if the industry has a unique program, they may also be excluded from the general RDTI. If a startup company is not able to demonstrate that software development is its core business, they will never have been encouraged to do software development requiring R&D in Australia from the get-go. Rather than being a way to look to the future, this is a return to the pre-1985 R&D Tax Concession’s Section 73A R&D program for small startups only.

All this to solve a problem that arguably no longer exists.

The court response to the narrowing of the definition of R&D to scientific research was to reject this narrowing as not consistent with the current legislation. The impact of this narrowing was not limited to software developers and this was demonstrated by the Full Federal Court decision in Moreton Resources Limited vs. Innovation and Science Australia  in July 2019. This decision re-affirmed that the RDTI is for the development of new knowledge in the form of new, and especially improved, products and processes. The distinguishing features of R&D activities from non-R&D activities are the same whether it is to overcome uncertainties in software solutions or physical solutions. They are both determined by whether the development activities include the experimental resolution of uncertainties in creating or integrating new or improved capabilities or functionalities in these new solutions. Outside of the RDTI, neither endeavour regularly uses the term “hypothesis” but both seek to answer the technical issues that can be expressed as one.

Following this case, the Australian Small Business and Family Enterprise Ombudsman completed the Review of the R&D Tax Incentive  in December 2019. This review focused strongly on the attitudes and approaches of the regulators  with respect to eligible software development. It addressed how the development of software is included directly and specifically in the definition of R&D in the Income Tax Assessment Act 1997  and how development methodologies such as Agile or Rapid can meet the requirements of the legislation. The response to the court cases and the Ombudsman’s review has seen recent assessments and audits resulting in more certainty for businesses doing eligible R&D activities as part of software and physical developments.

The solution to divide the program is not justified. Clear, relevant and concise guidance material on when software development is eligible R&D and what development costs are allowable R&D expenditure is a better solution than tinkering with legislation or creating a limited program for a small part of the software development industry in Australia. To that end, consolidating the three guidance documents regarding software eligibility on the www.business.gov.au website into one coherent and consistent piece would be an excellent start.

Should you wish to discuss this matter further, please do not hesitate to contact Ian Ross-Gowan on 02 9810 7211 or email ian.ross-gowan@mjassociates.com.au