H2O Exchange Pty Ltd and Innovation and Science Australia (Taxation) [2019] AATA 4195 (14 October 2019)

October 31st, 2019


A recent case heard in the AAT has affirmed a previous decision by Innovation and Science Australia (AusIndustry) finding that Activities registered under the R&D Tax Incentive were not in accordance with the legislative requirements.

The case: H2O Exchange Pty Ltd and Innovation and Science Australia (Taxation) [2019] AATA 4195 from 14 October 2019 refers to a company that registered R&D activities for the FY14 and FY15 periods.


The project registered related to the development of an online platform for the cross-border (inter-state) trading of water entitlements and allocations, that could react to changes (legislative, trade induced) in real-time.

The core/experimental activities registered were:

Activity 1.1 – Development of core trading platform and date base, CRM and various APIs

Activity 1.2 – Integration of matrix rules

These activities persisted between the two applications.

AAT Decision

Relying on the evidence of subject matter experts, the AAT upheld AusIndustry’s finding that the outcome of the activities were not unknown, and could have been readily deduced on the basis of available knowledge, specifically noting that

  • In regards to Activity 1.1 – “the outcome was known to be achievable, and in part, it was known not to be achievable because of the conveyancing requirements of two of the three States”
  • In regards to Activity 1.2. – “That it (the objectives) may have been achievable using AI software but probably not otherwise”, and that the statements made “do not say for what purpose the applicant was experimenting with AI software, or what experiments were done”.

It was also discussed that:

  • Insufficient evidence was provided that specific work, that would otherwise have been R&D (specifically in regard to Activity 1.2), was being undertaken during the period;
  • On the evidence, the outcome of the activities could only have been successful had there been changes to the Legislative requirements (implemented by each individual State);

In this regard, the uncertainty stemming from an absence of control by the applicant is not uncertainty of outcome consistent with s.355-25(1)(a). It is the outcome of the experimental activities that must be uncertain in order to satisfy s.355-25(1)(a)

  • There was a glaring inability to ascertain, from the third-party provider, any information about the specific technical challenges that were evident within the project.

It is noted that no specific technical issues were registered, and the outcome of examination/cross-examination may not have swayed the tribunal in any event

  • At the time of development there were a range of potentially suitable solutions already in the market, but, that they accepted, just because another entity already achieved the desired outcome but kept the means by which it was achieved secret, then whether the outcome is known will depend on whether others desiring to achieve the same outcome can know that by using readily available software tools (or perhaps by reverse engineering consistently with intellectual property rights) they can achieve the same result. Otherwise, they might try and fail, even though another has succeeded by some unknown means.

Learnings from Decision

This decision highlights the following points for applicants registering software activities:

  • Sufficient detail must presented in applications to highlight the specific technical uncertainty that exists, and why experiments must be undertaken to close a knowledge gap.
  • It is not sufficient to merely state that there is no ‘off-the-shelf’ product/solution available for the purpose/intended outcome. Applicants must explain why the outcome of attempting to build their desired function was unknown or could not be determined in advance based on existing knowledge.
  • Whilst S.27M variations may be granted (subject to the satisfying of s.32A IR&D Act) they are not likely to be accepted unless made within 10 months after the end of the income year, (per Clause5.2(1) s.30E IR&D Act). Further, subject to Clause 5.2(3)(c) the applicant cannot make a variation whilst a reviewable decision is the subject of external review. It is accordingly important that correct and sufficient detail is included in the original application lodged.

Click here to view the H2O Exchange Pty Ltd and Innovation and Science Australia case.

Please contact our office if you wish to discuss in greater detail.


William Ferguspn / Friday, December 6, 2019 - 09:01

So it appears that they are ruling that only the Research part counts. The Development part is ineligible for the tax offset. Research is about whether it can be done, Development is about how it can or should be done. When creating software, almost all of the effort is in the Development. Looks like we can't claim that now.

Post a Comment

(*) indicates required field.