The Budget said the quiet part out loud

What the 2026-27 Budget funded, what it didn’t, and the financial architecture our sector now operates inside.

The 2026-27 Budget was never going to rewrite international education. It didn’t need to. The architecture built across 2024 and 2025 was already in place: the National Planning Level, the Genuine Student framework, the Evidence Level reclassifications, the ESOS amendments.

What the Budget did was tell us which parts of that architecture the government is now willing to pay to enforce, and which parts of the migration system it is quietly redesigning around it.

Visa application charge revenue is forecast to hit $6.18 billion in 2026-27, up from $4.66 billion the year before. A $1.52 billion year-on-year lift. Not from more applications, but from higher prices on the same applications.

Stack the recent fee history. The student visa charge moved from $710 to $1,600 to $2,000 across roughly two years. The 485 doubled from $2,300 to $4,600 on 1 March 2026. The Passenger Movement Charge rises from $70 to $80 from 1 January 2027, worth $755 million over five years.

Home Affairs received 9.48 million visa applications in 2024-25, with a 7.3 per cent refusal rate across all categories. Every application carried a non-refundable fee. None of it is returned on refusal.

The arithmetic is straightforward. The system is collecting more money from fewer applicants, refusing a larger share of them in priority categories, and reinvesting a fraction of the revenue in new compliance and assessment functions rather than in faster, cheaper decisions. The $85.2 million to DEWR for skills assessments sits inside a system pulling $6.18 billion through the front door.

This is not a critique of the individual measures. The trades pathway has merit. The integrity funding is a defensible policy choice. But the direction of travel is worth naming. Revenue per applicant up. Refusal rates up in priority categories. Processing times excessive. Application volumes down. Department headcount on visa processing climbing.

The migration system is no longer cost-recovery. It is a revenue centre that funds its own expansion.

The revenue feeds the red tape

Net overseas migration: apples and oranges

NOM is a residency-based measure. It counts anyone, citizen or not, who has been in Australia for 12 of the last 16 months, minus anyone who’s been out of the country for 12 of the last 16 months.

It is not a visa grant figure. Not the permanent migration cap. Not the number of people Home Affairs let in last year. A returning Australian citizen counts. A New Zealander on the trans-Tasman arrangement counts. A student who arrived two years ago and is now on a 485 counts.

The political conversation about “cutting migration to 225,000” is a conversation derived from a statistical artefact. The actual levers are visa caps, grant rates, and the conditions that determine whether temporary holders stay or go.

The permanent program

185,000 places. Over 70 per cent reserved for skilled migrants. The headline cap hasn’t moved. The composition has.

Onshore applicants will be prioritised over offshore in both the skill and family streams. The remaining 55,110 offshore places will mostly go to high-skilled migrants addressing long-term skill needs. The points test is being “optimised”. That is the consultation language used by the Department since the Parkinson Review.

Read together, this is a quiet but substantial shift. Offshore skilled pipelines have just become a smaller, more contested pool. Onshore conversion, the 485 to 482, the 482 to 186, the 500 to 485 to 189, has become the structurally favoured pathway.

For agents whose practices are weighted offshore, the effective grant probability for a given client profile has moved against you. The clients haven’t changed. The competition has.

The international student layer

The NPL settings from August 2025, 295,000 places for 2026, remain the operating framework. No new cap. No new commission regime. No new ESOS amendments. No new student visa fee.

What changed is the resourcing behind the existing system. $19.8 million over four years buys more decision-maker capacity, more requests for further information. Probable consequence: higher refusal rates. The student visa framework is not relaxing.

For providers, the message is clear: the visa assessment rule will be applied more rigorously than before.

The English language layer

The Adult Migrant English Program is being reformed now, with a replacement program from 2029. Flexible tuition, student supports, eligibility narrowed to those “most in need”. The language suggests a tighter rather than broader program.

Means-testing AMEP without tightening upstream English requirements at the visa stage creates a structural mismatch. The cost of inadequate English doesn’t disappear when the program narrows. It just moves to employers, schools, and community services. ELICOS providers with AMEP contracts have a four-year planning horizon. The procurement design will matter more than the announcement.

What is not in the Budget

The Budget is also defined by its absences.

• ×No new commission reporting infrastructure funding, despite the active ESOS amendments debate.
• ×No specific funding for OMARA or the Migration Agents Regulations 2026 transition commencing 1 April 2026.
• ×No funded response to the agent commission cap policy proposals.
• ×No university research or HELP changes affecting international students.
• ×No movement on the student visa fee, kept at $2,000.

The CGT change nobody is talking about as a talent issue

The Budget confirmed the most substantial change to property tax in a generation. The 50 per cent CGT discount is being replaced from 1 July 2027 with a real-gain discount carrying a 30 per cent minimum on additional gains. Negative gearing is narrowed to new builds. Losses can only be offset against property income, not wages.

This is not a migration measure. But it lands in the same space as the talent-mobility argument I made earlier this year about the CGT discount and international graduates. The discount has not been removed for talent. It has been recalibrated toward new builds and away from existing-stock speculation.

For 482, 485, and onshore PR candidates modelling long-term wealth-building in Australia versus Canada, the UK, or Singapore, the calculation has just shifted. Whether it shifted enough to matter for talent decisions is a separate question, and one the sector hasn’t yet engaged with. It should.

What this means

For migration agents, model the onshore prioritisation explicitly in client conversations. A 189 candidate offshore in 2026 is not the same opportunity it was in 2025. A 485 onshore with a credible pathway has become more valuable, not less. For providers, the regulatory framework is stable, the enforcement intensity is rising, and the cohorts you onboard from 2026 will be assessed against a better-funded integrity apparatus, not a more generous one. The points test rebuild is the policy event of the next twelve months.

And for the sector as a whole, it is worth being honest about the financial architecture we now operate inside. The Department charges $2,000 to consider a student application, refuses one in five of them in current cohorts, and keeps the fee regardless. It charges $4,600 for a 485. It charges $80 to leave the country. It will collect $6.18 billion in visa application charges next year. The proceeds fund more of the apparatus that produces the charges.