The Signal was the Program.
Australia's Economic Accelerator was unwound by ministerial decision because it was created by ministerial decision. A structural read of commercialisation programs without architecture.
The Commonwealth government’s announcement that Australia’s Economic Accelerator will close to new entrants and the $800 million of unallocated funding will be redirected has been read across the ecosystem as a budget call, a signal failure, and a blow to researchers mid-application. All of that is fair. None of it is the structural question.
What was actually withdrawn was not $800 million. As the announcement states, the existing cohort will have funding towards their individual outcomes, whether successful or not. What was withdrawn was the stability of a stage-gated commercialisation signal that applicants, industry partners, and university research offices had been planning against. Stage-gated funding from proof-of-concept to proof-of-scale exists in every credible deep-tech system globally because the risk profile and time horizon do not always correspond to the needs of private capital. That is a structural market function, not market failure. When the signal goes, the planning architecture built against it goes with it.
Two questions follow from that. Was the ending appropriately governed? And is the system AEA operated inside actually designed to convert funding into commercial outcomes at all?
The ecosystem conversation is mostly running on the first. The second is the one that matters more.
A question on governance.
Innovation and commercialisation are a long game. The final outcome of a given venture may occur long after the initialisation of a commercialisation program.
Ending a research commercialisation program can be a defensible call as portfolios concentrate, priorities shift, and new mechanisms supersede old ones. The question is not whether programs should ever end. It is whether the ending is governed by architecture or by discretion.
The contrast with comparable systems is what makes this sharp.
The United States
The United States runs the Small Business Innovation Research program, established in 1982 and operating across eight presidents and both parties. The SBIR is not discretionary as it is a statutory set-aside of 3.2% of any federal agency’s extramural R&D budget above $100 million. It is also codified at 15 U.S.C. 638, and eleven federal agencies are obligated to run SBIR programs under that set-aside. The program has sunset provisions and requires periodic legislative reauthorisation.
Its authority lapsed on 30 September 2025, and the program sat in legal limbo for five months, with companies burning through reserves and tech transfer offices pausing negotiations. Then, in March, Congress passed bipartisan reauthorisation, which was signed into law in April, extending the program through fiscal year 2031.
The point is not that SBIR is untouchable. It can fall over like all programs, but the point is how it falls over. It falls over through a visible legislative process, with public debate, against a structural architecture that resumes operation when authority is restored. The funding is not subject to discretion because the set-aside is in statute.
The European Union and the United Kingdom
The European Innovation Council sits within Horizon Europe, established under Articles 173 and 182 of the Treaty on the Functioning of the European Union, and operates within the EU’s Multiannual Financial Framework.
Innovate UK is a statutory council within UK Research and Innovation, established under the Higher Education and Research Act 2017. UKRI operates at arm’s length from government as a non-departmental public body. Both have funding levels that flex with budget cycles, but neither has institutions that can be unwound by executive decision, and neither can have its core funding removed without a legislative or budgetary process running its course.
The architecture sits in primary legislation. The funding mechanism is built into the architecture.
Australia
Australia’s Economic Accelerator had a different structure. The program was created by the Higher Education Support Amendment (Australia’s Economic Accelerator) Act 2023, which established a statutorily mandated Advisory Board, a research commercialisation strategy, and an annual investment plan, all of which are prescribed under Division 42 of HESA.
What it lacked was a statutory funding set-aside. The $1.6 billion ten-year envelope was an administrative commitment, not a legislative floor. Funding could be withdrawn by executive decision because the envelope was never anchored in statute. The Advisory Board, the strategy, and the investment plan remain in place. The funding does not. The program’s architecture presumability was built into the legislation. The architecture for continued funding was not.
The structural read is not that AEA should have continued. It is that AEA could only ever end this way, through the funding mechanism, because it was never anchored in legislation. As a result, the program survives, but the funding does not.
A harder question.
The harder question is whether additional funding within the current architecture yields more commercialisation outcomes.
The international comparisons have something Australia does not, and it is not more money; they have system coordination.
The SBIR is structurally coupled to federal procurement, with the agencies that fund Phase I and II frequently serving as buyers in Phase III.
The EIC Accelerator pairs grants and equity with the Business Acceleration Services layer to support regulatory navigation and investor access.
Innovate UK sits within UKRI alongside the seven research councils and Research England, with a coordinated regulatory architecture across the Medicines and Healthcare Products Regulatory Agency, the British Standards Institution, and the Catapult network.
In each system, funding is one instrument inside a coordinated commercialisation architecture.
Australia funds, but the rest of the architecture is partial or absent. AEA was a funding instrument inserted into a system that was not designed to receive it efficiently.
There is data that paints a contrasting picture, worth sharing directly. The widely cited framing that Australia ranks last in the OECD for industry-research collaboration draws on OECD data series going back to the mid-2010s and has been repeated in ecosystem commentary for nearly a decade. More recent Australian Government tracking suggests improvements in adjacent metrics, with Australia ranking 7th out of 39 in the 2023 OECD innovation survey on the share of innovation-active businesses.
All in all, signalling that the trajectory for Australia is not flat.
The numbers that hold across both readings are the input-output gap and the cluster picture.
Australia ranks 16th globally on innovation inputs and 27th on innovation outputs in the 2025 WIPO Index, with the gap most pronounced in Knowledge and technology outputs, where Australia ranks 29th. The 2025 WIPO cluster ranking shows Australia with two clusters in the global top 100, down from three in 2024. Strong inputs, weaker outputs, particularly in knowledge and technology commercialisation, narrowing cluster footprint.
The pattern is consistent regardless of which collaboration metric is preferred.
The point is not that Australia is uniquely broken, but even with improving collaboration metrics and a respectable overall innovation ranking, the conversion of research investment to commercial output continues to lag. Funding is not the only constraint, which indicates that the surrounding system architecture and market appetite also need attention.
Signals do not execute themselves; systems do. The kill-switch frame that emerged from the corporate venture capital wind-downs of the last 24 months applies here without modification. A program that sits inside the wrong system architecture is fragile by default and can be unwound by the next budget cycle, the next leadership shift, or the next reprioritisation.
Having a kill switch should not be considered a flaw in the program. It is the system letting the program be the only thing carrying the load. Consideration of building a durable architecture, and funding levels that can flex without the system or the program collapsing. Without architecture, the program becomes the architecture, and the architecture ends when the program does.
What should the next program consider?
If the next attempt is going to do different work, three structural design choices matter more than the funding number attached to it.
The first is statutory architecture rather than departmental discretion. A program established under primary legislation, with an arms-length body owning it institutionally, can have its funding levels flexed without the institution itself unwinding. Durability of architecture is what allows funding to flex without the system collapsing. International comparisons all share this property to varying degrees, and the evidence is consistent: the institution survives because the architecture survives.
The second is coordinated system design rather than standalone funding. The regulatory pathway, the IP framework, industry partnership norms, procurement’s willingness to absorb new technology, and capital follow-on are not background conditions; they are key inputs to the commercialisation system. A program that funds in isolation will likely produce the conversion rate AEA produced. A program designed within a coordinated system could do better.
The third may matter more than its position in the list suggests. Portfolio logic should be defined at design time. Some programs operate as open-stage, gate-controlled funnels, in which every applicant can, in principle, progress. Others run as concentrating portfolios where the cohort narrows deliberately, with reserves and follow-on rules visible from the start. Both are defensible.
What is not defensible is starting one and ending up as the other through an external decision. Applicants can plan against either logic, but they cannot plan against an undeclared logic.
What should we watch for?
The signal worth watching over the next 12 months is whether the successor mechanism is announced with or without statutory architecture.
A multi-year departmental envelope is the same program with a different number attached. A program established under primary legislation, sitting inside coordinated commercialisation architecture, is a different proposition entirely. The framing of the announcement will tell the ecosystem which one is on the table.
For operators within the system, the move is to stop planning against single instruments. Stage-gated funding from a discretionary program is a fragile input. Consideration needs to be given to planning against architecture, not against the program. Where the architecture is coordinated, the work can compound. Where it is not, expect the rug to move, even without any early market indications.
Australia does fund research at scale. However, it has not yet been asked whether Australia is willing to build the architecture that converts funding into outcomes. Until that conversation starts, the kill switch will keep firing.
This article is based on the author's professional experience and interpretation of publicly available information. It is provided for general informational purposes only and does not constitute advice. Any views expressed are the author's own and do not refer to any specific organisation, program, or individual.
References & further reading
Australian Government, Department of Education. Australia’s Economic Accelerator - Program Update. https://www.aea.gov.au/australias-economic-accelerator-program-update
Capital Brief. Commonwealth government axes $800m research commercialisation program. https://www.capitalbrief.com/briefing/commonwealth-government-axes-800m-research-commercialisation-program-39051371-d3ee-440e-b2a6-5d5c7c49e10b/
InnovationAus. ‘Sends the wrong signal’: Research sector blasts AEA axing. https://www.innovationaus.com/sends-the-wrong-signal-research-sector-blasts-aea-axing/
Science and Technology Australia. Federal Budget “savings” create more uncertainty without a plan. https://scienceandtechnologyaustralia.org.au/federal-budget-savings-create-more-uncertainty-without-a-plan/
AEA Advisory Board. 2024-25 Annual Report on research translation and commercialisation. https://www.aea.gov.au/download/813/aea-advisory-boards-annual-report-2024-25/384/annual-report/pdf
On international governance architecture
United States:
SBIR / STTR Congressional Research Service. Small Business Research Programs: Overview and Issues for Reauthorization in the 119th Congress. https://www.congress.gov/crs-product/IF12874
Congressional Research Service. Small Business Research Programs: Selected Issues for Reauthorization. https://www.congress.gov/crs-product/R48629
CSIS. SBIR and STTR Reauthorization and the Future of Small Business Innovation. https://www.csis.org/analysis/sbir-and-sttr-reauthorization-and-future-small-business-innovation
SBIR.gov. The History of the SBIR and STTR Programs. https://www.sbir.gov/tutorials/program-basics/tutorial-5
European Union:
EUR-Lex. Horizon Europe - the framework programme for research and innovation. https://eur-lex.europa.eu/EN/legal-content/summary/horizon-europe-the-framework-programme-for-research-and-innovation-laying-down-its-rules-for-participation-and-dissemination.html
European Commission. Horizon Europe - Research and Innovation. https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-europe_en
European Innovation Council. EIC Accelerator. https://eic.ec.europa.eu/eic-funding-opportunities/eic-accelerator_en
European Parliament Fact Sheets. Multiannual Financial Framework. https://www.europarl.europa.eu/factsheets/en/sheet/29/multiannual-financial-framework
United Kingdom:
Legislation.gov.uk. Higher Education and Research Act 2017. https://www.legislation.gov.uk/ukpga/2017/29/notes/division/3/index.htm
GOV.UK. UKRI Framework Document 2025. https://www.gov.uk/government/publications/ukri-framework-document-2025/ukri-framework-document-2025
UKRI. Our relationship with the government. https://www.ukri.org/who-we-are/how-we-are-governed/our-relationship-with-the-government/
Australia's Performance
Current data (2024-2025)
Australian Government, Department of Industry, Science and Resources. How is the Australian innovation system tracking in 2024? https://www.industry.gov.au/news/how-australian-innovation-system-tracking-2024
Australian Government, Department of Industry, Science and Resources. What do international indicators say about innovation? https://www.industry.gov.au/news/what-do-international-indicators-say-about-innovation
WIPO. Global Innovation Index 2025: Australia. https://www.wipo.int/edocs/gii-ranking/2025/au.pdf
WIPO. Global Innovation Index 2025: Cluster ranking. https://www.wipo.int/web-publications/global-innovation-index-2025/en/cluster-ranking.html
Historical Context
Public Sector Network. State of the Australian Innovation System. https://publicsectornetwork.com/insight/state-of-the-australian-innovation-system-how-government-industry-academia-connect-innovation
Australian Government. Research Translation and Commercialisation Agenda. https://www.education.gov.au/research-translation-and-commercialisation-agenda
OECD Economics Department. Boosting Productivity in Australia (Working Paper No. 1025). https://www.oecd.org/content/dam/oecd/en/publications/reports/2013/02/boosting-productivity-in-australia_g17a224d/5k4c0vt9xhf3-en.pdf
Deloitte Access Economics. The path forward for commercialising university research. https://www.deloitte.com/au/en/services/financial-advisory/perspectives/the-path-forward-commercialising-university-research.html
Research commercialisation system design
OECD. University-Industry Collaboration. https://www.oecd.org/en/publications/university-industry-collaboration_e9c1e648-en.html
Australian Government. Action Plan to supercharge research commercialisation. https://www.minister.industry.gov.au/ministers/taylor/media-releases/action-plan-supercharge-research-commercialisation