The Cheapest Room in the Blueprint
In March, Ambitious Australia handed Canberra a 20-recommendation redesign of the national innovation system. Last week the government gave its first structural answer: it renovated the wing it already owned.
Good evening, reader.
Editor's note: There are two ways to read last week's release of the NCGP Policy Review. The first is as an administrative clean-up of a $1 billion grants program, which is what it is. The second is as the government's first structural answer to Ambitious Australia, the blueprint it was handed in March, which is what it has become. This piece is about the second reading, and about what a government reveals when it chooses which room to build first.
In today's Innovation Commons:
On the 22nd of July, ministers Jason Clare and Tim Ayres released the final report of the National Competitive Grants Program Policy Review, the deepest redesign of the ARC's $1 billion-a-year grants machine in decades (Ministers' Media Centre).
The first new scheme, Explore, opens in March 2027 to fund high-risk exploratory research; a final round of Discovery Projects (DP28) opens in February 2027 with the expression-of-interest stage removed (ARC).
Every new scheme is to carry collaboration potential, align with national priorities and satisfy research security expectations: a quiet conversion of the curiosity program into an instrument of national purpose.
The blueprint this answers, Ambitious Australia, landed on the 17th of March with 20 recommendations, among them a National Innovation Council, six National Innovation Pillars and a substantially redesigned R&D Tax Incentive (DISR; RSM analysis).
132 days on, every recommendation that needs new money or new legislation remains under consideration. The one that needed neither is done.
What was actually announced
Strip the communiqué language and the NCGP Policy Review does five things. It refocuses the ARC's National Competitive Grants Program on early-stage research. It collapses the scheme architecture into fewer, simpler programs. It prioritises salaries and career pathways for early-career researchers, and strengthens support for Aboriginal and Torres Strait Islander researchers. It requires the research it funds to be visible and legible to industry, with collaboration potential built into every new scheme. And it aligns the whole machine with national priorities, research integrity and research security settings.
The transition is concrete and dated. A last vintage of the old machine, Discovery Projects round DP28, opens in February 2027, minus the expression-of-interest stage that a post-implementation review quietly buried. The first scheme of the new machine, Explore, opens in March 2027, pitched at high-risk, high-reward projects. Sector webinars begin in the December quarter of 2026, and further schemes roll out progressively after that.
ARC Board Chair Peter Shergold called it "the most significant reform to the ARC's grants program in decades". On the machinery, he is right. The program funds roughly a billion dollars of discovery research a year and, on the government's own arithmetic, returns more than three dollars of economic output for every one invested. It has not been rebuilt at this depth since the ARC's modern grants architecture was assembled.
The second reading
Now read the same release as an institutional document rather than an administrative one, the way this newsletter read AEMO's transmission plan a month ago.
The old NCGP was, in institutional terms, a curiosity commons. Money was allocated by peer review against a single test, excellence, and the program's indifference to application was not a bug but its constitutional principle. The new NCGP keeps peer review and keeps the word excellence, and then adds conditions around it: alignment with national priorities, collaboration potential in every scheme, integrity and security screening that reaches into who researchers work with and what the work touches. The release states plainly that funded research should advance the country's security, defence and international standing.
The significant thing is not that the grants got simpler. It is that alignment became the admission ticket. Solar cells and quantum technologies get invoked as the fruits of undirected discovery, in the same document that ends undirected discovery as the program's organising idea. Whether that trade is wise is a live argument: focus is precisely what Ambitious Australia prescribed, and scattering a small country's research dollar across every field of human curiosity is its own kind of waste. But the trade should be named, because it completes a pattern this newsletter has been tracking all winter. Data centres may connect if they bring firmed supply. Grid-scale loads are contracted to behave on command. And now discovery research funds itself by demonstrating its usefulness to national purpose. Layer by layer, the commons is being converted into an alignment machine. Readers of The Claude Nation will recognise the direction of travel, if not yet the destination.
The blueprint and the room
Here is where the sequencing gets interesting.
Ambitious Australia, the final report of the Strategic Examination of R&D chaired by Robyn Denholm, made 20 recommendations across six elements. The heavy ones are architectural: a National Innovation Council to govern the system, six National Innovation Pillars from health to resources to technology, National Strategic Initiatives with targeted CSIRO funding, and a redesign of the R&D Tax Incentive that would lift the refundable threshold from $20 million to $50 million, create a premium startup scheme with a higher refundable rate, pay refunds quarterly instead of annually, and add a production tax credit for onshore advanced manufacturing.
Sort those recommendations by what they cost to say yes to. The Council needs cabinet. The Pillars need portfolio ministers to surrender turf. The RDTI package needs Treasury and legislation, and Treasury has spent a decade sanding the RDTI down, not building it up. Almost everything on the list carries a price tag, a bill, or a fight.
Except one. Realigning the ARC's grants program with national priorities needs no new appropriation, no legislation and no cabinet submission. It is machinery-of-government work inside an existing billion-dollar envelope, executed by a statutory agency that answers to two ministers who both got quotes in the press release. That is the recommendation that shipped first, 132 days after the report landed, proudly badged as delivering a key recommendation of Ambitious Australia.
Reviews get answered in the order of administrative cheapness, not systemic importance. This is not cynicism; it is how federations metabolise blueprints. But it means the press release and the reform are measuring different things. The press release says the government is acting on Ambitious Australia. The reform says the government has, so far, built the one room in the blueprint that came free with the house.
What the sequencing does to the system
Three practical consequences follow, and none of them appear in the announcement.
For researchers, the calendar is now the policy. Careers in this country are planned against ARC scheme dates, and the transition inserts a hinge year: the last Discovery round in February 2027, the first Explore round a month later, outcomes of each landing late that year at the earliest. The early-career salary prioritisation is real and overdue. It is also, for now, a promise about schemes whose guidelines nobody has seen. The December-quarter webinars, not last week's report, are where the actual constitution of the new machine gets written, and the thing to watch is whether collaboration potential enters as a scored criterion or a tick-box.
For startups and scaling firms, the honest message is harsher: your part of the blueprint is the part still in the queue. The refundable threshold, the premium startup rate, the quarterly refunds, the production credit, every measure aimed at business R&D spend, which is the number actually in decline, sits behind a Treasury decision that has no announced date. A firm planning its 2027 R&D program on the strength of Ambitious Australia is planning on a document the government has not yet answered.
For the system, sequencing is signal. Universities, investors and state agencies all read Canberra's order of operations as revealed preference. What has been revealed, so far, is a government comfortable directing the research it already pays for, and not yet committed to the architecture that would grow the research it does not.
The house or the room
This reading comes with a falsification test, and it is worth stating plainly. If the government's formal response to Ambitious Australia lands in the spring sittings with a named National Innovation Council and an RDTI bill, then the July reform was the vanguard of a build, and this piece was premature. If the response drifts into 2027 while the pillars stay unnamed and the tax package unlegislated, then the July reform was the build, and the blueprint has joined the national collection of ambitious documents that were agreed with in principle.
Watch three dates: the Explore guidelines out of the December-quarter webinars, the DP28 round in February, and whichever sitting week Treasury chooses to say something, or nothing, about the RDTI.
A blueprint is a promise drawn to scale. The cheapest room is now furnished, and the ribbon has been cut in front of it. Whether Australia is building the house will be decided where these things are always decided: in the appropriations.
— The Editor
P.S. Share this on LinkedIn/X to spark conversations with your colleagues.