New AusIndustry Guidance on Software R&D – A Year Well Spent?

April 19, 2022 Kris Gale

The Department of Industry, Science, Energy and Resources recently released its latest guidance product, ‘Software-related activities and the Research and Development (R&D) Tax Incentive’.

The guidance was released in draft in April 2021 and public consultation closed in June. Since that time, AusIndustry and the ATO, assisted by the Technology Council of Australia, set up a working group involving a number of software companies. The final product apparently included input form that working group.

A Positive Result

In our MJA Update dated 10 June, 2021, we made the following conclusion about the April 2021 draft:

“The draft is a marked improvement on previous versions released in the market in the past half dozen years and includes a case study, something that has been absent from guidance material for a long time.”

We are happy to report that our overall conclusion hasn’t changed. The key improvement in the guidance is the explicit recognition that software R&D can make use of existing methodologies “like Agile, Waterfall, Rapid and similar”(page 9).

This is a very welcome development as use of existing methodologies was recurrently put forward by AusIndustry as a reason to disallow claims in assessments in years gone past.

Along with the acknowledgement of the fact that software testing can equate to experimental development, software R&D claims are now on a much sounder footing and can be made with much greater confidence.

So, a good news story then?

Well one question that remains to be answered is what value was added in the twelve months between the release of the draft and the final product?

Was This Worth The Wait?

At the recent RDTI Roundtable in March, the supplied Program Update promised that the final version of the product would contain the following:

“Key updates to the new software guide include:

• additional software-specific R&D examples
• revised case study to provide additional detail and illustrate the self-assessment process using the customer portal registration form
• updated guidance regarding the software exclusion to align it with all other exclusions listed in subsection 355-25(2) of the ITAA 1997 (NB: core R&D activities subject to the software activity exclusion may still be assessed as supporting R&D activities)
• providing links to supporting resources.”

Overall, it has to be said that the guidance did not arrive in the advertised form.

Whilst the software exclusion commentary was upgraded and additional links to resources provided, there were no additional software-specific R&D examples provided. None at all.

Further, the revised case study was stripped of much of its detail and a somewhat impenetrable diagram entitled ‘Diagram: Self-assessment tool – Is my R&D eligible?’ was added.

Finally, the addition of the illustration of the self-assessment process using an approximation of the customer portal registration form has proven to be highly alarming as we will explain below.

In summary, the final product is a diminished version of the draft that received extensive public consultation back in mid-2021. It has removed significant components of the original text and added some material that sits somewhat uncomfortably in the restructured framework of the document. It very much feels like an abbreviated version that has involved some ‘cut & paste’ for unclear reasons.

So, the twelve months wait doesn’t appear justified, and we would be surprised if the working group was heavily involved in this final version of the product.

We expect that the contribution of the working group will be more clearly reflected in the upcoming modules referred to in the aforementioned Program Update:

• record-keeping in software-development
• sector-specific case studies
• expanded reference to prominent classes of software-related R&D activities
• video content designed to support applicant registrations.

Important Homework

We believe two aspects of the guidance require immediate attention:

1. Competent Professional

There has been a recent trend to rely on the concept of the ‘competent professional’, taken from patent law, to help illuminate the eligibility of R&D activities in the RDTI legislation.

Yet the RDTI legislation is clearly based on a taxpayer test, rather than a competent professional version.

This appears to be explicitly recognised in the discussion on page 7 regarding ‘Outcome cannot be known or determined in advance’,  where the following statement is made:

“We expect you to search widely for an existing way to achieve your activity’s goal before you start it.”

This is an excellent expression of the obligation that a taxpayer must prove it has discharged before commencing its R&D activities. The use of the word ‘widely’ sensibly replaced the word ‘worldwide’ from the draft.

Yet, immediately thereafter, the guidance goes on to make the following statement:

“For an activity to be a core R&D activity, it must not be an excluded activity…, and a competent professional in the relevant field cannot know or determine the outcome before the activity starts. “

 “A competent professional is a hypothetical person who has access to the knowledge, skills and qualifications needed to demonstrate expertise in an appropriate field. They have access to all of the current knowledge available in the relevant field from anywhere in the world, provided that the information or experience is publicly available or reasonably accessible.” (page 8).

The statement is a misstatement of the law and stands in contradiction of the statement immediately above it on page 7. The test is a taxpayer one, not one regarding a competent professional, and the conflict should be resolved by removal of this discussion from the guidance immediately.

1. Registration Example

At first blush, the market would have been pleased about the inclusion of a worked example of a registration portal application based on the Far Side case study (pages 27-29) as there is an oft-expressed desire for this type of guidance.

Unfortunately, the worked example does not reflect our understanding of how an application should be prepared and, in fact, it is our contention that a taxpayer application that followed the precedent set out in the example would be at a very high risk of failing any form of assessment or audit activity.

We submit that the example should be removed immediately and a working group be formed under the auspices of the RDTI Roundtable to produce suitable output in this regard.

The concerns include the following:

• The sample form is not an exact facsimile of the current registration portal application. There is no clear reason as to why this is the case.

• The sample includes sections that are not part of the current form. For example, the sample includes a box entitled ‘A brief description of the activity’ under the Core Activity section that is not required by the portal. Incidentally, the sample sets a 1,000 character limit and proceeds to prove a response with more than 1,500 characters.

• Certain responses in the sample are grossly inadequate. By way of illustration, in the ‘What was the hypothesis? section, a 200 character response is provided in a box that sets a maximum of 4,000 characters:

“Our hypothesis was that we could build a new mathematical and statistical framework for 3D patch generation to achieve a smooth and continuous graphic visualisation of an entire planet surface from a limited selection of known 3D patches.”

The sample is not an example of economic expression of the information required. It is an example of a technical objective and is not an example of a hypothesis. It does not provide a provable or disprovable proposition. It does not state the variable and independent variables in play. It is highly misleading and would not pass muster in an audit situation.

• The text is scant on quantitative detail that is the hallmark of successful applications and includes statements around unclaimed activities that would not typically feature in an application.

A Better Way Forward

If we are correct regarding the immediate concerns detailed above, the question must be asked is how is it possible that the guidance was released to the market, particularly in light of the working group that has been involved in the revision process for several months.

The answer lies in better engagement with the right stakeholders. And these stakeholders can be found in the current membership of the RDTI Roundtable. These are the industry associations and advisers who have the knowledge and detail as to how the RDTI works at a practical level. In fact, MJA first requested a working group regarding software R&D under the RDTI’s Terms Of Reference at the Roundtable’s first meeting in March 2019. This request has been consistently ignored. Yet when the regulators do set up a working group on software R&D, only one Roundtable member was included.

The result is the published guidance that is a significantly diminished product in comparison to the original draft.

It’s time for the right mix of key stakeholders to come together, not to continue to be kept apart.

Should you wish to discuss this matter please do not hesitate to contact Kris Gale on 02 9810 7211 or email kris.gale@mjassociates.com.au