International students and the rental frontline

On a wet Tuesday evening outside a two-bedroom terrace in Sydney's inner west, forty people wait in a line that runs past the corner shop and around the block, phones out, rental applications already filled in twice over. Some are nurses and hospitality workers who have been searching for six weeks. Others are international students, newly arrived from India or China, handing over a share-house agent's holding fee before they have even seen the room, because everyone in the queue has heard what happens to people who hesitate. None of them set the settings that put this many renters in front of this few properties. Those settings were written in Canberra, changed, unmade, and changed again, on a timetable set by ministers — not by the people standing in the rain.

That queue is the sharp end of a policy story that has changed direction five times in three years. Australia reopened its border to international students in December 2021, watched enrolments climb past every previous record, then spent 2023 and 2024 trying to slow that rebound down — first through visa-integrity rules, then through a legislated national cap that failed in the Senate, then through a ministerial direction that achieved much the same effect without a vote, then through visa fees that more than doubled inside twelve months. Each change was defended, at the time, as the fix. None of them, on the government's own figures, has yet reduced the number of students competing for a lease in Australia's tightest rental markets.

What the government did

International students enter Australia chiefly on the Student visa (subclass 500), which lets a person study a registered course full-time, work limited hours during term, and depends on settings — provider risk rating, financial and English-language thresholds, processing priority — set entirely by the Department of Home Affairs and the minister of the day. A related visa, the Temporary Graduate visa (subclass 485), lets graduates stay on and work for two to four years after finishing a degree. Together these two visa classes drove most of the post-pandemic migration rebound documented elsewhere on this site, and they are the settings this article follows in detail.

The border reopened to international students in December 2021, after almost two years in which new student arrivals were effectively barred. Enrolments recovered quickly, helped by lengthened post-study work rights and faster visa processing designed to rebuild a sector that had lost its largest single source of revenue almost overnight. By late 2023, the recovery had gone further and faster than the government's own forecasts, and ministers began looking for ways to slow it.

The speed of the rebound was not solely a matter of pent-up demand. Universities, having lost their largest revenue stream almost overnight in 2020, marketed aggressively to rebuild enrolments once the border reopened, and the government's own settings — faster processing, lengthened graduate work rights — were designed explicitly to help them do it. Both sides of that arrangement had a strong, entirely legitimate incentive to move quickly: universities needed the fee revenue to fund research and teaching they had already committed to, and the government needed the export income and the skilled-worker pipeline the sector provides. Neither side had an equally strong, equally fast-moving incentive to expand the number of beds available to house the students that recovery brought back.

The first lever was the Migration Strategy, released in December 2023, which tightened the "genuine student" test and lifted financial and English-language requirements, aimed chiefly at low-quality private colleges functioning as a visa-extension pathway rather than a place of study — a policy examined in more detail in this site's coverage of the broader migration-housing collision. The second lever, Ministerial Direction 107 (MD107), took effect in January 2024 and instructed the visa-processing system to prioritise applications from "low-risk" providers, giving the department an administrative tool to slow overall volume without changing any legislation.

The third and most consequential attempt was legislative. In 2024 the government introduced the Education Services for Overseas Students Amendment (Quality and Integrity) Bill, which would have formally capped new international-student commencements at a "National Planning Level" of 270,000 for 2025, allocated provider by provider. A Senate committee inquiry through mid-2024 drew heavy opposition from universities, private colleges and peak bodies including the Group of Eight and Universities Australia, who warned the caps would cost billions in lost revenue and thousands of jobs. A majority Senate committee report in October 2024 recommended the bill proceed, but a dissenting report from Coalition senators foreshadowed the vote to come. On 18 November 2024 the Coalition signalled it would vote against the bill; on 28 November 2024, the final sitting day of the year, the bill failed to pass the Senate — defeated by a rare alliance of the Coalition, which argued the design did not fix the underlying migration settings and unfairly punished individual private providers, and the Greens, who argued the bill scapegoated international students for a housing shortage they had not caused and that the real fix was more housing supply, not fewer students.

Within three weeks of that defeat, the government moved again. On 19 December 2024 it issued Ministerial Direction 111 (MD111), replacing MD107 with a new priority-processing framework: each education provider gets an indicative allocation set at 80% of its share of the National Planning Level, and once a provider reaches that threshold, its remaining applications fall to the back of the processing queue. In practice, this achieved administratively — through a direction signed by the minister, with no vote, no amendments and no committee hearing — much of what the failed legislation would have done in law. The 2025 National Planning Level was set at 270,000 new commencements, the same headline figure the failed bill had proposed. For 2026, the level was raised to 295,000, split roughly between higher education (about 196,000) and vocational training (about 94,000) — a loosening, not a further tightening, within twelve months of the figure first being set.

The debate did not end with MD111. At the May 2025 federal election, the Coalition — having helped defeat Labor's legislated cap only months earlier — campaigned on a tougher one: at most 115,000 new commencements a year at public universities and 125,000 across the VET, private-university and non-university higher-education sectors, a combined ceiling below the National Planning Level Labor had just set administratively, alongside a 25% cut to the permanent migration program over two years. The Coalition claimed its stricter settings would free up more than 100,000 homes over five years (approx., Coalition 2025 election policy). Labor was returned to government at that election, and the National Planning Level it had already set stood. The episode is nonetheless instructive: the argument in Canberra was never really about whether to cap international student numbers. Both major parties agreed a cap of some kind was warranted. The disagreement — and the reason the legislated version collapsed — was over mechanism, design and degree, which makes the resulting three years of administrative improvisation harder to excuse as an unavoidable consequence of political disagreement over the principle itself.

Running alongside these processing settings, the government also raised the price of applying. The Student visa application charge rose from $710 to $1,600 from 1 July 2024, then to $2,000 from 1 July 2025 — a rise of more than 180% across two increases in a single year, and one of the highest student-visa fees of any comparable study destination; a comparable United States student-visa application costs roughly a tenth as much (approx.). Unlike the processing settings, the fee increase applies to every applicant regardless of provider risk rating, and its stated purpose has shifted between visa-integrity, cost-recovery and Budget-repair framing across the period it has been in force.

The evidence

The scale of the rebound is not in dispute. Department of Home Affairs data shows the number of Student visa (subclass 500) holders in Australia climbed from a pandemic-era trough of roughly 400,000 in 2021, as the closed border cut off new arrivals, to 568,753 by the end of the 2022-23 financial year — already a record — and to 608,262 by the end of 2023-24, a further increase of about 7% (Department of Home Affairs, Study visa statistics; 2021 figure approx.). Counting the related Temporary Graduate visa, roughly 852,000 people held a student or graduate visa in Australia in July 2023 (approx., Department of Home Affairs) — a population larger than the city of Canberra, almost all of it needing somewhere to live inside a small number of capital-city rental markets.

None of this was starting from zero. Australia had spent the better part of two decades building international education into a mainstream part of its economy, and student numbers were already substantial by international standards before the border closed in March 2020 (approx., Department of Home Affairs historical series). The 2021 trough was not a return to some pre-growth baseline; it was a sudden, almost total interruption of an already-large and still-growing sector, which is part of why the subsequent rebound felt, and was, so fast. To put the resulting cohort in context: Australia's private rental market comprises a few million dwellings nationally, but a student population in the high hundreds of thousands, concentrated in a handful of inner-city postcodes rather than spread evenly across the country, can move the vacancy rate in those specific postcodes far more than the national totals suggest. A renter in regional Tasmania felt almost none of this pressure. A renter three kilometres from the University of Sydney or RMIT felt all of it.

Students are not spread evenly across the country, and that concentration is the heart of this topic. The overwhelming majority live and study in a handful of inner-city precincts around Australia's largest universities — inner Sydney, inner Melbourne and inner Brisbane chief among them — competing directly in the general private rental market, because the country has never built enough dedicated student housing to take them out of it. Purpose-built student accommodation covers only around 6% of demand nationally, roughly one bed for every fifteen or sixteen students, and in Sydney, Melbourne and Brisbane the ratio is under one bed per ten students, compared with more than two-and-a-half beds per ten students in London (approx., Cushman & Wakefield; Property Council of Australia). Industry analysis has estimated the resulting shortfall at roughly 200,000 beds against 2023 enrolment levels alone (approx., Cushman & Wakefield). The consequence follows directly: the great majority of the 600,000-plus student cohort is not absorbed by a dedicated housing stream. It is competing, lease by lease, against nurses, graduates and families for the same shrinking pool of ordinary rental listings — concentrated in the same postcodes that recorded the country's lowest vacancy rates through 2023 and 2024.

Vacancy rates in those postcodes tell the same story from the supply side. Capital-city rental vacancy rates fell to around 1% nationally through the second half of 2023 (approx., CoreLogic/SQM Research), and Sydney's own rate hit a record low of 0.8% in early 2024 (SQM Research, approx.) — a level that leaves a renter with, in practical terms, no genuine choice of listing. The timing lines up closely with the student-visa recovery: vacancy rates had been elevated during the border closure, when inner-city rentals emptied out as students left, and tightened sharply as students returned from 2022 onward.

The geographic concentration compounds the pressure rather than spreading it. New South Wales and Victoria host the substantial majority of international enrolments, chiefly in Greater Sydney and Greater Melbourne, the same two capital cities that recorded the country's tightest rental markets through this period (approx., Department of Home Affairs; Department of Education). Within those cities, enrolment concentrates further still, around the handful of universities — the University of Sydney, UNSW, the University of Melbourne and RMIT among the largest — that draw the bulk of new international commencements each year. A national vacancy-rate average understates the local reality: the postcodes immediately surrounding these campuses recorded some of the lowest vacancy rates of any part of the country, well below both the national and even the tight capital-city averages already cited.

The same concentration explains why the export-revenue figures in the steel-man above translate so directly into university balance sheets. At several of Australia's largest universities, international student fees are reported to account for more than a quarter of total revenue, funding research infrastructure and domestic teaching places that government funding and domestic fees alone do not cover (approx., university annual reports). That reliance cuts both ways for this article's argument: it is the clearest evidence the export-revenue case in the steel-man is not overstated, and it is also the clearest evidence why any sudden change to enrolment settings has an immediate, material effect on university finances — a fact that helps explain why the sector lobbied so hard against the legislated cap, and why the government, once that cap failed, moved quickly to an administrative substitute rather than simply letting numbers run.

Purpose-built accommodation houses roughly one in fifteen international students. The rest compete, lease by lease, in the same rental market as everyone else.

The two curves in that chart are not a coincidence of timing so much as two readings of the same underlying number. Every additional student visa holder is, in the overwhelming majority of cases, an additional renter — and the years in which that number climbed fastest are the same years in which the vacancy rate fell fastest. None of this means students caused the entire shortage; the next section sets out why that claim goes further than the evidence supports. But the concentration is real, it is measurable, and it landed hardest in exactly the postcodes this site's other housing coverage already shows were under the most acute pressure.

The caps themselves, whatever their intent, have not yet reduced that concentration. Despite MD111's queue-throttling and the fee increases, Australian universities reported a record 545,000 international student enrolments in 2025 (approx., sector reporting) — higher, not lower, than the year before the legislated cap was even proposed. The National Planning Level for 2026 was then set at 295,000 new commencements, above the 270,000 level applied in 2025. On the numbers the government itself publishes, three years of visa-integrity rules, a failed legislative cap, a replacement ministerial direction and two rounds of fee increases have not yet produced a fall in the underlying population competing for a lease near an Australian campus.

The critique

Judged against its own stated purpose — easing pressure on rental markets under acute stress — the sequence of policy in this article reads less like a considered strategy and more like a government reaching for whichever lever was still available after the last one failed. Visa-integrity rules in December 2023, an administrative processing direction in January 2024, a legislated cap introduced and defeated in November 2024, a replacement ministerial direction within three weeks, and two rounds of fee increases either side of that defeat: five distinct instruments, deployed across three years, each justified at the time as the fix for the same problem. A single change, however imperfect, invites scrutiny of its design. Five changes in three years invites a different question — whether the government understood the scale of the challenge when it first reopened the border, or is instead improvising in response to headlines and a housing-stress figure that kept climbing regardless of which lever was pulled last.

The pivot from legislation to ministerial direction deserves particular scrutiny, because it changed how much scrutiny future changes would receive. When the ESOS Amendment Bill failed in the Senate, it failed after a public committee inquiry, submissions from universities and student bodies, and a recorded vote — the ordinary machinery of parliamentary accountability working as intended, however contested the outcome. MD111, which achieved much the same practical effect three weeks later, required none of that. A minister signed it. There was no committee hearing, no amendment process, and no vote a future government would need to legislate to reverse. Major settings governing one of Australia's largest visa classes and export industries are now being run by administrative direction specifically because the legislative route failed — a precedent that trades transparency for speed, and one this site's readers should weigh regardless of which party is signing the next direction.

Australia was not alone in facing this collision, and the comparison sharpens the critique of process rather than the underlying goal. Canada capped new study permits from January 2024, and the United Kingdom restricted most postgraduate taught students from bringing dependants from the same month, both moving in a single, direct step rather than through a legislate-fail-then-improvise sequence. Australia's own attempt to legislate a comparable cap took most of 2024, failed, and was replaced within three weeks by an administrative direction achieving a similar practical effect without the same public scrutiny. The direction of travel — tightening international student settings in response to domestic housing pressure — was common across several English-speaking study destinations facing similar problems at the same time. The manner in which Australia got there, changing mechanism five times rather than committing to one and adjusting it, was not.

A government that tries five different levers on the same problem in three years is not managing a crisis. It is reacting to one, in public, one headline at a time.

The fee increases compound the concern rather than resolving it. Raising the Student visa application charge from $710 to $2,000 across two increases inside twelve months (1 July 2024 and 1 July 2025) does not add a single bedroom to the national housing stock; it raises Commonwealth revenue and prices out the most price-sensitive applicants, disproportionately those from lower-income source countries, while wealthier applicants absorb the cost and keep coming. A tool that filters by ability to pay rather than by the actual pressure a given enrolment adds to a given rental market is a blunt instrument dressed up as a housing-pressure fix — and, on the enrolment and Planning Level figures set out above, it has coincided with rising rather than falling numbers.

Perhaps the clearest evidence of incoherence is the mismatch between the government's own settings and its own outcomes. The National Planning Level, the government's flagship replacement for the failed legislative cap, was raised for 2026 rather than held or lowered — an increase of 25,000 places within twelve months of the figure first being set, after a year in which enrolments hit a fresh record regardless of the restrictions already in force. If the purpose of the National Planning Level was to relieve pressure on the rental markets this article documents, raising it in year two, before that pressure had visibly eased, sits awkwardly with the stated rationale for introducing it in year one.

The sector's own reaction captures the trap the government built for itself. Universities and peak bodies including the Group of Eight argued throughout 2024 that the caps, however implemented, would cost billions in export revenue and damage Australia's standing as a study destination. At the same time, renter and housing advocates argued the settings — administrative queue-throttling and a doubled visa fee — still fell well short of what an undersupplied rental market needed, given commencement numbers kept climbing regardless. Both critiques being live at once, from opposite directions, is not evidence the government found a sensible middle path. It is evidence of a policy that satisfied neither the export industry it was meant to protect nor the renters it was meant to help.

Meanwhile, the one lever genuinely capable of relieving pressure without trading off export revenue — building more purpose-built student accommodation — moved slower than any of the demand-side levers above. Industry pipelines put total purpose-built student beds on track to reach only around 144,300 by 2027 (approx., sector estimates), against a shortfall measured in the hundreds of thousands. The government reached fastest for the tool that changed who could come, not for the tool that would have changed how many beds existed for them to live in once they arrived — even though the housing shortfall, not the enrolment number itself, is what put the renter in the queue outside that Sydney terrace.

What they got right — and the honest caveats

Credit is due in several places. The December 2023 Migration Strategy's tightened "genuine student" test targeted a real integrity problem: a small number of low-quality private colleges had been operating as a visa-extension pathway rather than a genuine place of study, exploiting vulnerable students and adding little to either the education sector or the country's skills base. Raising financial and English-language thresholds protects prospective students from being sold a course they cannot complete or afford, as much as it manages migration volume.

It is also worth stating plainly, in line with this site's standing commitment to treat migration as a policy number and never as a slur: international students are not the antagonists of this story. Community legal centres and student advocacy groups have reported cases of overcrowding and exploitative share-house arrangements affecting international students specifically, priced out of stable leases by the same shortage this article documents (approx., community-sector reporting). The renter queuing outside that Sydney terrace in this article's opening is, as often as not, a student rather than a local — competing in the same undersupplied market, subject to the same rent rises, with less bargaining power and less familiarity with tenancy protections than most. A policy response that treats students only as a demand figure to be capped, rather than also as renters bearing the cost of the same shortage, misses half of who this issue actually affects.

MD111 was also, on its own terms, a genuine correction rather than a cosmetic one. MD107, its predecessor, had been criticised for disproportionately favouring a small number of large "low-risk" providers over smaller, regional and public institutions competing for the same indicative allocation; MD111's shift to a fixed percentage of each provider's own National Planning Level share was a direct response to that criticism, spreading the processing advantage more evenly across the sector rather than concentrating it further.

The most important caveat, and the one the fairness doctrine of this site demands be stated plainly, is that international students did not create Australia's rental shortage by themselves, and treating enrolment numbers as the primary cause risks obscuring a much larger and older failure. Independent analysis has pushed back specifically on the claim that students are driving the crisis, pointing out that the structural shortfall in dwelling completions — examined in detail elsewhere on this site — pre-dates the post-pandemic student rebound by well over a decade and will outlast it regardless of where visa settings land. Students are a real, measurable, geographically concentrated addition to demand in a small number of postcodes. They are not the reason Australia has spent a generation building fewer homes than a growing population needs, and a policy response that treats visa settings as a substitute for a housing-supply fix will not close that older gap no matter how many times the settings are changed.

It is also fair to note the purpose-built accommodation pipeline, while still short of what is needed, is at least growing rather than stalled: the 144,300-bed pipeline figure cited above is a real increase on today's stock, even if it will not be built quickly enough to catch up with recent enrolment growth. And it remains genuinely uncertain, as of the most recent data available, whether softening visa-application volumes reported through 2026 reflect the accumulated weight of these settings finally working, a genuine cooling in global demand for Australian study places, or reputational damage from three years of policy instability and a visa fee that has nearly tripled. Another year or two of enrolment and vacancy data will answer that question more reliably than any single Budget forecast can today.