Quick answer: Possibly — but a funded development contract puts two important R&D claim questions in play at once. Expenditure is not notionally deductible to the extent it is not at risk (ITAA 1997, s 355-405): the rule has two limbs — the consideration must be expected as a direct or indirect result of the expenditure being incurred, and expected regardless of the results of the activities — and where both are met that part drops out. Separately, the activity must be conducted for you (s 355-210) — which you assess yourself on the whole of the circumstances, applying the legislation and ATO guidance, with the ATO able to review that assessment later. Milestone payments for a delivered article are not automatically consideration caught by the at-risk rule, and a cost-reimbursement line often is. ("Recoupment" is a different concept, belonging to the separate clawback provisions.) You self-assess.
28 July 2026 — this article describes the current rules. The 2026-27 Federal Budget announced R&DTI changes scheduled to start 1 July 2028; those are not yet law, and we cover them separately in the proposed $50,000 Budget measure.
Adelaide is a place where this question frequently arises. The Osborne shipyard sits at one end of the city and the Australian Space Agency is headquartered at Lot Fourteen at the other — the same innovation precinct Ignition Research works from. Between them sit several hundred small suppliers: machine shops, RF and sensor houses, autonomy software teams, materials and coatings specialists. Many of them work on technically challenging development programmes. Many of them assume the R&D Tax Incentive (R&DTI) is closed to them because "the prime paid for it".
That assumption is too fast in one direction and too slow in the other. The rules do not turn solely on whether money changed hands. They ask, in two separate places, what that money was for.
We assume throughout that there is genuine R&D — an outcome that could not be known in advance without a systematic progression of work. Whether your programme crosses that line is a separate question, and for a lot of defence and space work it is the harder one; we come back to it below.
The Two Conditions a Funded Contract Puts in Play
1. Was the expenditure at risk?
Section 355-405 of the Income Tax Assessment Act 1997 is the provision that catches funded development. On the ATO's guidance, the at-risk rule has two limbs, and both must be met: the consideration must be received, or reasonably expected, as a direct or indirect result of the expenditure being incurred (the nexus limb), and it must be expected regardless of the results of the activities on which the expenditure is incurred (ATO, TR 2021/5). Where both limbs are satisfied, to that extent you cannot notionally deduct the amount.
The nexus limb: The link the rule looks for is between the consideration and the incurring of the expenditure — not between the payment and the project in a loose commercial sense.
The regardless-of-results limb: This is the limb that does the commercial work, and the one most often missed. Consideration you would receive whether the research succeeded or failed points to expenditure that was never at risk. Consideration contingent on a technical outcome — payable only if the article passes, or only if a performance target is met — does not sit as comfortably in the rule, because it is not expected regardless of the results. A supplier invoicing actual engineering hours plus a margin, with the customer bearing the cost of every rework cycle, is much closer to the rule than one paid a fixed price only on delivery of a qualified article.
"To the extent": This is not an on/off switch for the programme. A single contract can carry a reimbursed non-recurring engineering line and a genuinely at-risk development effort the supplier funds from its own balance sheet in the hope of a production award. The rule bites on the reimbursed part.
There is also a relieving consequence worth knowing. Where s 355-405 already stops you claiming expenditure related to a recoupment, the ATO's clawback guidance says you do not then also make a clawback adjustment for that amount — the two mechanisms are not stacked on the same dollar. Clawback is covered separately in our Insights.
2. Who was the activity conducted for?
Even where the money is at risk, an R&D entity can only notionally deduct expenditure on activities meeting the conditions in s 355-210 — broadly, that the activity was conducted for it. The ATO frames that as asking who receives the major benefit, informed by who effectively owns the results, who controls the conduct of the activities and who bears the financial risk, weighed together on the whole of the circumstances.
Defence contracting makes this live because the standard contracting suites deal with intellectual property expressly, and often generously to the Commonwealth or the prime. Note what the test actually reads, though: effective ownership, described by the ATO as looking at practical as well as formal rights to the results, not proprietorship of a particular registered right. A supplier that licenses foreground IP to a prime for the platform, keeps the underlying know-how and can apply it to the next customer, is in a different position from one that assigns everything and retains nothing.
We unpack the "conducted for" factors in detail in Who claims when a contract manufacturer makes your product — the analysis is the same one, applied to a different supply chain.
What Is Usually the Harder Question: Is It R&D at All?
Defence and space programmes are documentation-rich, which fools people into thinking eligibility is easy to evidence. It is not, because most of that documentation is compliance evidence, and compliance is exactly what the R&DTI does not reward.
Core R&D activities are experimental activities whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience, but can only be determined by applying a systematic progression of work based on principles of established science, and that are conducted for the purpose of generating new knowledge (business.gov.au).
Applied to this sector, that tends to sort work as follows — indicatively, not as a rule:
Typically closer to core R&D
Typically not core R&D
Developing a material, coating or joint whose behaviour in the required environment cannot be predicted from existing data
Applying a qualified material to a new part in a known way
Resolving a thermal, vibration or EMI interaction that current engineering practice cannot predict for your configuration
Analysis that confirms a design already expected to comply
Developing an autonomy or sensor-fusion approach where achievable performance is genuinely unknown
Integrating proven components to a specification
Trials designed to find out whether an approach works, with a stated hypothesis
Qualification testing to demonstrate that it does
That last row is the one that costs suppliers the most. Testing whose purpose is to demonstrate conformity with a standard generally will not satisfy the statutory experimental and new-knowledge tests, and the fact that it is expensive, mandatory and technically demanding does not change that. It is fact-dependent, not automatic: qualification work can be core R&D where those statutory tests are independently met on the facts, and it may qualify as a supporting activity where they are not. You self-assess. The same logic runs through building-code compliance testing, which we treat at length in a companion article in our Insights.
None of which means qualification spend is always lost. Activities that are not core R&D may still qualify as supporting R&D activities where they are directly related to core activities — and, if they produce goods or services, are directly related to producing goods or services, or are of a kind excluded from being core, only where they are conducted for the dominant purpose of supporting a core activity (business.gov.au). The dominant-purpose question is decided activity by activity.
A Worked Shape: Two Lines on the Same Contract
Illustrative only — a hypothetical, not a ruling.
An Adelaide supplier wins a subcontract from a prime. The contract has two payment lines:
Line A — Non-recurring engineering, reimbursed at cost plus a fixed margin
Invoiced monthly against actual hours. Rework cycles are payable. If the technical approach fails, the supplier is still paid for the hours.
Line B — Company-funded development run in parallel
To establish whether a manufacturing route the supplier believes in can hold tolerance at rate. Nobody has agreed to pay for it. It exists because winning the production award depends on it.
The at-risk analysis reads these differently, and it should. Under Line A, at the moment each month's cost is incurred, the supplier can reasonably expect consideration as a result of incurring it — and can expect it whether or not the technical approach works, which is the second limb. Under Line B, the supplier wears the loss if the route does not work; a later production award is a commercial hope, not consideration for having incurred the expenditure.
So the practical consequence is not "defence suppliers can't claim". It is that the claim has to be built from the contract's payment mechanics upward, and that the boundary between funded and unfunded work needs to exist in the accounting records before year end — not be reconstructed from a spreadsheet in month eleven.
Three Sector-Specific Traps
Work performed overseas: Where part of the development happens outside Australia — a foreign parent's test facility, an overseas launch or trials campaign — the domestic paragraph of s 355-210 does not cover it. Activities conducted overseas need an overseas finding from AusIndustry, applied for in the income year the activities are conducted. Export-control or security constraints on where work may physically be performed do not create an exception; they are a reason to plan the finding early.
Registration is not a place to hide detail, and it is not published either: Registration with AusIndustry requires a description of the activities, the unknown and the systematic work — a real technical narrative. Both administering agencies are constrained in what they may disclose, but under different regimes: the ATO is bound by confidentiality provisions, while DISR is constrained by the Industry Research and Development Act 1986 framework. There is, however, one deliberate exception: the ATO publishes an annual R&D tax transparency report listing each claimant's name, ABN or ACN and total R&D expenditure claimed, roughly two years after the relevant income year, and entities cannot opt out (ATO). The report discloses a total, not what you were working on — but the fact of claiming, and the scale, becomes public.
The $20,000 floor, and the RSP route past it: R&D expenditure for the income year must generally be at least $20,000 — that is the lower bound of the entitlement in s 355-100(1) of the ITAA 1997 (ATO) (business.gov.au). A small supplier whose only genuinely at-risk work is one company-funded trial can land under that floor after the funded lines are stripped out. Eligible R&D expenditure incurred to a qualifying RSP may still qualify even where the usual $20,000 R&D expenditure threshold is not met — read precisely: where total notional deductions are below A$20,000, the offset base is generally limited to qualifying expenditure incurred to a non-associate Research Service Provider for services in a field for which it is registered, together with eligible CRC Program contributions. Other in-house amounts do not automatically join that below-threshold base. See claiming R&D under $20,000. Using an RSP does not guarantee eligibility — you still self-assess.
What to Fix Before the Next Contract Is Signed
Separate funded and unfunded work in the contract: Distinct scope, distinct pricing basis, distinct milestone structure.
Cost-code from day one: A project code that maps to activities, so at year end the at-risk boundary is a report rather than an argument.
Deal with foreground results expressly: Assignment, licence-back, and rights to the underlying know-how and test data all feed the "conducted for" factors.
State the unknown before you test: A hypothesis recorded before the trial is the difference between core R&D evidence and a qualification report.
Plan overseas activities in the year they happen: Findings are applied for in the income year, not afterwards.
Where an RSP helps is upstream of all of this: helping companies document the unknown, hypothesis and evaluation process arising from their technical work. Local context matters here too — see the R&D Tax Incentive in Adelaide for how the SA ecosystem interacts with a claim. Two caveats stand: an RSP supplies research capability, not tax advice, and engaging one does not answer the at-risk or "conducted for" questions by itself.
What the 2026-27 Budget Announced
The 2026–27 Federal Budget announced proposed R&DTI changes that are intended to apply to income years starting on or after 1 July 2028. These measures are not yet law, and current-year claims continue to be assessed under the existing rules (ATO). See our Budget update for further details on the proposed changes and their status.
Frequently Asked Questions
Q: Can a defence subcontractor claim the R&D Tax Incentive on work paid for by a prime?
A: Potentially, to the extent the expenditure was at risk. Section 355-405 of the ITAA 1997 removes the notional deduction where two limbs are met: the consideration is received or reasonably expected as a direct or indirect result of the expenditure being incurred, and it is expected regardless of the results of the activities. A cost-reimbursed engineering line is much more exposed than company-funded development the supplier would lose if the approach failed. It applies to the extent of the funding, not to the whole programme. You self-assess.
Q: Does a cost-plus contract stop an R&D claim in Australia?
A: Not automatically, and not necessarily for the whole contract — but a pricing basis that pays you because you incurred the cost, whatever the research shows, is the fact pattern the at-risk rule is aimed at. The analysis is done on the amounts, so a contract with both reimbursed and unreimbursed development can produce a claim over the unreimbursed part.
Q: Who claims the R&D when the Commonwealth or the prime takes the IP?
A: Whichever entity the activities were conducted for, decided on the whole of the circumstances under s 355-210. Effective ownership of the results is one factor alongside control and financial risk, and the ATO looks at practical as well as formal rights — so a supplier that licenses foreground IP but retains the know-how and can reuse it is not in the same position as one that retains nothing.
Q: Will my company's name be published if I claim the R&DTI?
A: Yes. The ATO publishes an annual R&D tax transparency report showing each claimant's name, ABN or ACN and total R&D expenditure claimed, about two years after the income year, and there is no opt-out. It shows a total, not a description of your activities; other entity-level program details are protected by the confidentiality provisions of the tax law.
Sources & Further Reading
legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-210, 355-405
Related: what an RSP is · R&D Tax Incentive in Adelaide · what does not qualify · refundable vs non-refundable offset · claiming R&D under $20,000
Talk to Ignition Research before you sign the next development subcontract — as a Registered Research Service Provider at Lot Fourteen in Adelaide, we assist defence and space suppliers in documenting the experimental work undertaken within development contracts, state the unknown before the test article is built, and keep cost records that separate funded delivery from company-funded research. Whether expenditure is "at risk" is a statutory tax characterisation for your company and its tax adviser, not something we determine. We are not a registered tax agent: your company self-assesses and remains responsible for its own claim, with advice and lodgement handled by your tax adviser. Get in touch.
This article is general information from a Registered Research Service Provider about the R&D Tax Incentive. It is not tax, legal or financial advice; eligibility depends on your circumstances and you should self-assess and seek your own advice.
Thinking about a project like this?
If you're weighing up an AI, software or technical improvement project and can't tell yet whether it's implementation or research, start with a quick read on where it sits.