The 2026-27 Federal Budget retargeted the R&D Tax Incentive in a way that most claimants have not yet absorbed. The headline rate uplift of 4.5 percentage points across every tier is the part that has travelled. The structural changes underneath it are the part that will actually determine outcomes.

From 1 July 2028, the way you are structured will materially change the value of the same R&D work.

What is changing.

The SME turnover threshold lifts from $20 million to $50 million. The refundable offset becomes a function of company age: SMEs under ten years old keep the refund, SMEs ten years and older get the same 23% rate but only as a non-refundable offset (a tax asset that requires future profit to use). The large business intensity threshold drops from 2.0% to 1.5%, with the premium tier rising from 16.5% to 21%. The minimum expenditure floor lifts from $20,000 to $50,000 unless the work runs through a Research Service Provider or a Cooperative Research Centre. The maximum expenditure cap lifts from $150 million to $200 million.

The single most important change sits behind those numbers. Supporting activities are being removed from the eligibility base entirely. From 1 July 2028, only core R&D expenditure under section 355-25 will qualify. Treasury-commissioned analysis found that subsidies for supporting activities did not produce additional R&D, and the policy intent is to concentrate the offset on the experimental work that generates new knowledge.

Supporting activities currently account for around 29% of registered claims. For most claimants, that is the number to focus on.

Why structure now matters.

Under the current rules, corporate structure is largely a hygiene question. Two companies running identical R&D programmes in different structures get broadly similar outcomes. Under the new rules, identical R&D programmes will produce materially different outcomes depending on the age of the claiming entity, the aggregated turnover of the group, the composition of core versus supporting work, and how the R&D entity sits inside the broader corporate structure.

A genuine new venture or properly-formed spin-out will access the 23% refundable offset (immediate cash, regardless of tax position) for the first ten years. An older trading company doing the same R&D in the same structure as today will receive the same 23% as a non-refundable offset, which only converts to value when the entity is profitable enough to use it.

That gap is large enough that multi-year R&D programmes spanning the 1 July 2028 commencement should be designed with the new architecture in mind, not retrofitted at claim time. FY27 is the window to get this right.

The reclassification opportunity.

The removal of supporting activities is being read by most of the market as a loss of eligible expenditure. We read it differently.

A significant proportion of expenditure historically classified as supporting can in fact be properly characterised as core, provided the experimental purpose and the hypothesis-to-evaluation progression are clearly demonstrated. Items have defaulted to the supporting category over the years because it was the safer or simpler classification, not because the underlying activity failed the core test.

Categories that warrant particular attention in the lead-up to commencement:

  • Data collection where the data acquisition is itself the experimental observation phase of the hypothesis test.
  • Iterative testing where each iteration is a discrete experiment evaluating a stated hypothesis.
  • Production runs conducted for a dominant experimental purpose and generating observations against the hypothesis.
  • Software build work where the technical uncertainty being resolved sits inside the build itself, not separately characterised.
  • Calibration and methodology development where the methodology is itself the new knowledge being generated.
  • Failure analysis and pivot work that drives a documented revision to the hypothesis or experimental design.

In each case the technical question is the same: can the activity be tied to a documented experimental progression with a defined hypothesis, observation and evaluation? Where it can, the activity is properly core, and the reclassification is defensible. Where it cannot, the expenditure will fall out of eligibility from 1 July 2028.

This reclassification exercise is going to be the dominant technical task on FY26 and FY27 claims. The work done now to establish the experimental architecture is what determines whether the post-2028 claim profile holds up.

What still needs to be resolved.

Budget Paper No. 2 sets the policy parameters. Several mechanical questions wait on the exposure draft legislation:

  • How the ten-year age test is defined, and whether it applies to the registering entity, the connected group, or the underlying business operation.
  • Whether predecessor-entity or continuing-business rules will be introduced to prevent restructuring around the age test.
  • The transitional treatment of expenditure straddling 1 July 2028.
  • The interaction with the existing R&D consolidated group rules under Subdivision 355-J.
  • The treatment of in-flight multi-year programmes commenced under the current rules.

We will publish a fuller piece once the exposure draft is released. The structural decisions that need to be made before then are independent of those mechanical questions.

What to do now.

If you have a multi-year R&D programme that will run past 1 July 2028, the work to do in FY27 sits in three places.

First, a systematic review of currently-claimed supporting activities against the core definition, with a view to documenting the experimental progression that supports reclassification.

Second, a structural review of how the R&D entity is positioned inside the broader group, against the new age and turnover settings.

Third, if a spin-out or new-venture structure is on the table for unrelated commercial reasons, the timing decision is now also an R&D decision.

If you'd like a working session on the implications for your specific programme, get in touch.