Help to Buy, the HAFF and the supply-side gap

Ask Housing Australia what happens after you apply for Help to Buy, the government's shared-equity scheme, and the honest answer for most applicants is: probably nothing, this year. A place is not granted to everyone who passes the income test and the deposit test. It is granted to a strictly limited number of buyers nationally each year, in a country where, in a typical recent year, well over a hundred thousand people took out a loan to buy their first home without any government equity contribution at all. For every applicant who receives a place, several who meet every eligibility rule will not.

Half an hour's drive away, a caseworker at a community housing provider is telling a different family a longer story: that the wait for a modest social housing property in her area can run past a decade, and that a fund named the Housing Australia Future Fund is meant to shorten it, one dwelling at a time, using not its ten billion dollars but only the investment income that ten billion dollars earns each year. Both queues are real. Both funds are genuine, hard-fought policy achievements that took years to legislate. And both raise the same plain question this article sets out to answer: do Help to Buy and the Housing Australia Future Fund put more roofs over Australian heads, or do they mostly help a lucky few compete for the roofs that already exist?

What the government did

The Housing Australia Future Fund and Help to Buy are the two centrepiece financing vehicles in the federal government's housing package, and they work in almost opposite ways. The HAFF is designed to add to the stock of social and affordable rental housing; Help to Buy is designed to help a limited number of buyers purchase a home — new or already standing — that the private market has priced them out of. Both were signature 2022 election commitments, and both took far longer to become law than the government first planned.

The Housing Australia Future Fund Act 2023 established a $10 billion investment fund, assented to in September 2023 (approx.) after a prolonged Senate standoff. The fund's capital is invested by the Future Fund Board of Guardians — the same statutory body that manages Australia's sovereign Future Fund and several related investment vehicles, including the Medical Research Future Fund and the NDIS Future Fund. This is the structural feature that gives the fund its name and its logic: the $10 billion corpus is not spent on housing. It is invested in a diversified portfolio of shares, bonds, infrastructure and other assets, and each year a portion of the return that portfolio earns — capped at $500 million a year, with a guaranteed minimum of $500 million for at least the fund's first five years regardless of how the investment portfolio actually performs in any given year (Housing Australia Future Fund Act 2023, approx.) — is transferred into a separate account, the Housing Australia Future Fund Facility, and paid out to fund housing. The $10 billion itself is intended to remain intact indefinitely, continuing to earn a return for as long as the fund exists.

The stated target for that disbursement is 30,000 social and affordable homes across the fund's first five years — reported as a split of roughly 20,000 social housing dwellings, rented to low-income tenants at below-market rates and allocated by need, and roughly 10,000 affordable homes, rented at a discount to market rent to essential workers and other moderate-income households (Housing Australia; National Housing Accord Facility material, approx.). Delivery runs through Housing Australia (the renamed National Housing Finance and Investment Corporation), which channels the money to state and territory housing authorities and community housing providers, who in turn build or acquire the dwellings. The HAFF sits alongside an older, related instrument, the Affordable Housing Bond Aggregator, through which Housing Australia issues government-guaranteed bonds to raise cheap finance for community housing providers — a mechanism that predates the HAFF but now operates under the same umbrella.

Within that 30,000-home target, a smaller share is carved out for specific acute needs: crisis and transitional housing for women and children leaving family and domestic violence, housing for older women at risk of homelessness — the fastest-growing group of people experiencing homelessness in the country, according to homelessness-sector reporting — and housing linked to veterans' services (Housing Australia Future Fund Act 2023; Housing Australia program material, approx.). These carve-outs do not sit outside the 30,000-home, 6,000-a-year ceiling; they are drawn from within it, which means every dwelling directed to one of these acute-need categories is a dwelling not available for the general social-housing waiting list this article returns to below.

The Act took two attempts to pass. The Coalition opposed it outright, arguing an investment fund exposed to market returns was an unnecessarily complicated way to fund something as urgent as social housing. The Greens, holding the balance of power in the Senate, blocked the bill twice through late 2022 and into 2023, arguing the $500 million annual cap was too low against the scale of the housing crisis and pushing for a national rent freeze or rent cap as a condition of support — a demand examined in full in this site's coverage of the Greens' balance-of-power role. The government did not concede the rent freeze. It did concede the annual-minimum guarantee described above, and separately announced a one-off $2 billion Social Housing Accelerator payment directly to the states — a distinct, on-budget program, not part of the HAFF itself — intended to fund additional social housing more quickly while the HAFF ramped up. With those changes, the Greens allowed the bill to pass in September 2023 (approx.).

Help to Buy followed a similar path. The Help to Buy Act, passed by Parliament in late 2024 (approx.) after more than a year stalled in the Senate by a combination of the Coalition and, for a period, the Greens, establishes a shared-equity scheme under which the Commonwealth co-purchases a home alongside an eligible buyer. The government contributes up to 40% of the purchase price for a newly built home or up to 30% for an existing home, and the buyer needs a minimum deposit of just 2%, with no lenders' mortgage insurance charged on the government's share. No rent is charged on the government's portion of the equity, but it is not free money: when the home is sold, or at any point the buyer chooses to "buy back" a share, the government receives the same proportion of the sale price or valuation that it contributed — capturing its share of any capital gain, and absorbing its share of any loss, rather than simply forgiving a grant. Eligibility is means-tested, with an income cap of roughly $90,000 a year for a single applicant and $120,000 for a couple (approx.), and the purchase price itself is capped at a different level in each state and territory, broadly tracking each capital city's median dwelling price. The scheme is limited to approximately 10,000 places a year (approx.), a cap set out in the enabling legislation and confirmed in Housing Australia's own scheme material.

The evidence

Start with the arithmetic the HAFF's own structure invites. A fund with a $10 billion corpus, targeting 30,000 homes over five years, is targeting an average of 6,000 homes a year (approx.) — a number bounded not by how much housing Australia needs, but by the $500 million annual disbursement cap the Act itself sets, regardless of how the underlying $10 billion actually performs in markets in any given year.

That annual figure also needs to be read against the backdrop it is trying to reverse. Commonwealth funding for social housing declined for decades under governments of both major parties, delivered through a series of five-yearly national housing agreements that were repeatedly criticised, including by the Productivity Commission, for lacking clear targets and for real funding that failed to keep pace with population growth. Social housing's share of Australia's total housing stock fell from roughly 6% in the early 1990s to somewhere near 4% by the early 2020s (approx., AIHW housing assistance data), even as the number of households needing it grew. The HAFF is better understood, on this reading, as an attempt to arrest and partially reverse a multi-decade decline that both sides of politics presided over at different times, rather than as a response to a problem this government alone created. That context matters for judging the fund fairly. It does not change the arithmetic of what 6,000 additional homes a year can and cannot do against a shortfall built up over thirty years.

Set that 6,000-homes-a-year average against the size of the queue it is meant to shorten. State and territory social housing waiting lists nationally run to well over 160,000 households (approx., Productivity Commission, Report on Government Services, social housing data) — people who already qualify for social housing and are simply waiting for a property to become available. At an average delivery rate of 6,000 additional dwellings a year, and setting aside that some of the fund's homes go to entirely new categories of need such as crisis accommodation, the HAFF's own target would take close to three decades to work through the existing waiting list alone, before counting a single new household that joins the queue in the meantime as rents keep rising elsewhere on this site's Cost of Living tab. The fund was never designed or marketed as a complete solution to that queue; even the government's own material describes it as one contribution among several. But the scale mismatch between a queue in the hundreds of thousands and an annual delivery rate in the low thousands is the first fact a reader needs to hold in mind when judging what "the government funded social housing" actually means in practice.

A fund built to run forever is still capped, every single year, at a number small enough to count on one hand's worth of zeroes.

Help to Buy's scale sits in a similar place relative to the market it enters. ABS Lending Indicators data shows well over 100,000 first-home-buyer loans are written in Australia in an average year (approx., ABS Lending Indicators) — a figure that moves with interest rates and confidence but has stayed in that broad range across the period Help to Buy was being designed and debated. Against that backdrop, a scheme capped at roughly 10,000 places a year, even fully subscribed every year, would touch under one in ten of the first-home buyers who enter the market annually without any government equity contribution at all. Housing Australia's own commentary at launch acknowledged the scheme was never intended to be the primary pathway into home ownership for most buyers; it was pitched as a targeted assist for a specific band of moderate-income earners who fall just short of qualifying for a standard loan on a full-priced home. On the numbers, that is an accurate description of its own ambition. The risk this article returns to below is what happens when a modestly scaled demand-side scheme meets a market where the underlying constraint is not who can get a loan, but how many homes exist to be bought at all.

Help to Buy's rollout also depended on a step easy to miss in the national debate over its design: each state and territory had to sign a bilateral agreement with the Commonwealth before the scheme could operate for buyers in that jurisdiction, because state governments hold the land-titles and conveyancing frameworks the shared-equity structure relies on. Not every state signed at the same time; some states took considerably longer than others to finalise their agreement after the Act passed federal Parliament (state government and Housing Australia announcements, approx.), meaning the scheme's effective start date, and therefore its first full year of the roughly 10,000-place cap, varied by postcode for reasons that had nothing to do with an applicant's eligibility. A national scheme whose availability depends on separate state-by-state sign-off is a reminder that even the demand-side half of this article's story is not purely a Commonwealth decision — it is also a federation-coordination problem, of the kind this site's States tab documents in other contexts.

It is also worth noting Help to Buy is not the first shared-equity scheme Australians have been able to access. Several states already run their own versions — Western Australia's Keystart, South Australia's HomeSeeker, and Victoria's Homebuyer Fund among them — each with its own income and price caps, place limits and repayment terms, established well before the federal scheme was legislated (state housing agency material, approx.). The federal scheme sits on top of, rather than replaces, these state programs, and the overlap raises its own coordination question — how many of Help to Buy's roughly 10,000 places a year go to buyers who would have qualified for a state scheme regardless — that neither the Act nor Housing Australia's published material fully answers.

That underlying supply constraint is measurable, and it sits upstream of both funds. ABS Building Activity data shows annual dwelling completions running at roughly 175,000 to 180,000 a year through the period both funds were legislated (approx.) — a rate the National Housing Accord itself judged inadequate when National Cabinet set an implied target of 240,000 completions a year to reach 1.2 million new homes over five years from mid-2024 (National Housing Accord communiqué, Aug-2023). Actual completions have run below that implied rate in every year measured since. Whatever finance either fund supplies, it still has to be converted into a finished dwelling by the same construction sector responsible for that shortfall — a sector ASIC's own insolvency data shows was under acute strain over the same years. Construction consistently accounted for the largest share of corporate insolvencies of any Australian industry through 2022-23 and 2023-24, of the order of a quarter or more of all company insolvencies nationally in some periods (approx., ASIC insolvency statistics) — driven, industry bodies report, by fixed-price contracts signed before materials and labour costs rose sharply, and by builders unable to reprice contracts already underway. Jobs and Skills Australia's Skills Priority List has continued to list core residential building trades — carpenters, bricklayers and several other trades — among the country's most persistent shortage occupations across the same period (approx.).

Put together, a fund that pays out $500 million a year and a scheme that finances up to 10,000 purchases a year are both entering a construction sector that was already failing to build enough homes for the population it already had, using a workforce ASIC and Jobs and Skills Australia both describe as stretched. Neither fund adds a carpenter, a bricklayer or a project manager to that workforce. Both compete for the same finite supply of them.

The HAFF's accounting treatment adds a second, more technical layer to the evidence. Because the $10 billion corpus is recorded as a financial asset — an investment the Commonwealth owns, rather than a grant it has spent — establishing the fund did not appear as a $10 billion hit to the underlying cash balance the way a direct social-housing grant of the same size would have. Only the annual disbursement from the fund's returns counts against the Budget's cash balance each year, at a fraction of the corpus (Australian Government Budget Paper No. 1, 2023-24 and following years, approx.). This is not improper — the same structure is used for several other Commonwealth investment vehicles, and it is disclosed in the Budget papers rather than hidden — but it means the $10 billion figure that features prominently in the fund's name and its political marketing is not the number that determines how many homes get built in a given year. That number is $500 million, and the gap between the headline figure and the operative figure is examined in more detail in this site's dedicated look at off-budget funds.

The critique

Judge the two funds against the question this article opened with — do they add supply, or subsidise demand into a market that cannot expand to meet it — and they fail that test in two different, instructive ways.

Help to Buy fails it almost by design. When the government's equity contribution goes toward an existing home — and nothing in the scheme's rules requires it to go toward a newly built one — the transaction adds a buyer to the market for a home that already exists. It does not create a new dwelling; it changes who can afford to bid for one that is already there. Comparable shared-equity and first-home-buyer assistance schemes, in Australia and internationally, have a long and well-documented history of doing exactly this: a government contribution that increases a cohort of buyers' purchasing power, dropped into a market where the number of homes for sale in a given period does not expand to match, shows up at least partly as a higher price for the same home rather than as an additional buyer successfully housed (Productivity Commission and Grattan Institute research on first-home buyer assistance and stamp-duty concessions, cited as secondary analysis, approx.). Australia's own first-home-owner grants, in earlier decades, are the domestic example this independent literature most often points to. Help to Buy's income caps, price caps and equity-repayment structure are more carefully targeted than a blunt cash grant, and they reduce — without eliminating — this effect. But the basic economics does not change because the assistance takes the form of equity rather than a grant: assistance that expands what a fixed pool of buyers can pay for a fixed pool of homes tends to bid up the price of that fixed pool, especially for the established-home purchases the scheme does not exclude.

Help to Buy does not require a single new home to be built. It only requires an existing one to be sold.

Australia is not the first country to try a shared-equity scheme under this name. The United Kingdom ran a larger and longer-running Help to Buy equity-loan scheme from 2013 to 2023, and the independent research examining it — including analysis published by the UK's National Audit Office and by housing economists — found the scheme was associated with higher prices for new-build homes in areas where take-up was concentrated, as developers priced in the additional buying power the government's equity contribution gave purchasers (UK National Audit Office; academic housing-economics literature on the UK Help to Buy scheme, cited as secondary comparative evidence, approx.). The mechanics of the Australian and UK schemes differ in scale and detail, and Australia's tighter income and price caps are a deliberate design response to some of the UK criticism. But the underlying economic mechanism — a subsidy that increases what an eligible buyer can pay, dropped into a market that does not expand supply in response — is the same mechanism in both countries, and the UK's decade of experience is the closest real-world precedent available for judging what a demand-side scheme does to prices when it is not matched by an equivalent expansion in supply.

The HAFF's failure mode is different and less about economic theory than about arithmetic. Nothing about its structure is dishonest: it does fund genuinely new social and affordable homes, not just a change of ownership on an existing one, which is a real point of difference from Help to Buy. But the annual cap that gives the fund its "perpetual, recession-proof" character — the very feature the steel-man above rightly credits — is the same cap that keeps its yearly output small relative to the waiting list it is meant to shorten. The government chose the $500 million figure, and chose to guarantee it regardless of investment performance, as a trade-off for keeping the $10 billion corpus off the underlying cash balance and intact for the long term. That is a legitimate fiscal design choice. It is also a choice that puts a hard ceiling on how many homes the fund can deliver in any given year, a ceiling set by an accounting decision rather than by the actual size of the social-housing shortfall.

Both funds, whatever their design differences, draw on the same scarce construction capacity documented above. Every additional government-funded dwelling — social, affordable, or Help to Buy new-build — competes for the same tradespeople, the same building materials and the same project-management capacity as every private and state-funded project already straining against a construction sector recording some of its highest insolvency rates on record. Without a matching expansion in that capacity — more apprentices completing trades, more building-materials manufacturing capacity, faster planning approvals feeding projects into the pipeline at a steady rate rather than in surges — additional government-funded demand for construction services, however well-intentioned, risks bidding up construction costs and wages rather than translating cleanly into additional finished homes on the timetable either fund promises. Housing Australia's own delivery has, in its first years, run behind the pace implied by the funds' five-year targets, consistent with a sector working near capacity rather than one with spare room to absorb extra government-funded projects instantly (Housing Australia progress reporting; state housing authority updates, approx.).

There is a further, more political dimension to the critique. Both instruments carry names and framing — a $10 billion "Future Fund" for housing, a scheme literally called "Help to Buy" — that sound considerably larger and more comprehensive than their operative scale. A citizen who hears that the government has established a $10 billion housing fund could reasonably assume $10 billion is being spent on housing; the operative figure is $500 million a year, a 2% draw on the headline number. A citizen who hears the government is helping people buy homes could reasonably assume the scheme reaches a meaningful share of the market; it reaches under one in ten first-home buyers in an average year, even at full capacity. Neither gap is a fabrication — both structures are disclosed accurately in the legislation and the Budget papers — but the distance between the headline figure that generates the political announcement and the operative figure that determines how many Australians are actually housed is a pattern worth naming plainly, because it recurs across several of this site's Budget and Debt topics.

What they got right — and the honest caveats

Credit is due on several fronts. Getting the HAFF through the Senate at all, after two failed votes, was a genuine legislative achievement — Commonwealth funding for social housing has a long history of being negotiated afresh, and sometimes cut, at every change of government or Budget cycle, and a legislated, perpetual funding stream is a structurally different and more durable commitment than any of its predecessors, whatever its annual scale. The guaranteed minimum disbursement, secured through the Greens' negotiation, is itself a genuine improvement on the original bill: it removes the risk that a bad year in investment markets would leave the housing sector with nothing to build on, converting a variable return into a predictable one.

The government also did not rely on the HAFF and Help to Buy alone. The $2 billion Social Housing Accelerator, paid directly and quickly to the states outside the HAFF structure, was designed to deliver additional social housing while the HAFF's slower investment-and-disbursement cycle ramped up — a genuine acknowledgement, in the funding design itself, that the HAFF's own annual cap would not be enough on its own in the short term. The National Housing Accord also expanded the National Housing Infrastructure Facility, directing finance toward the enabling infrastructure — water, power and road connections — that can hold up a new estate regardless of how much construction finance is available, and secured planning-reform commitments from the states through National Cabinet. The government also established the National Housing Supply and Affordability Council, an independent statutory body tasked with publishing an annual assessment of the housing system and tracking progress against the Accord's targets — a genuine, if modest, transparency mechanism that lets outside scrutiny, including this site's, check delivery against promise (Treasury; National Housing Supply and Affordability Council Act 2023, approx.). Together, these measures are genuine supply-side steps that sit alongside, rather than instead of, the two funds this article examines, and they complicate any claim that the government's entire housing response is demand-side or purely financial in nature.

Help to Buy's design also deserves credit against the alternative it most resembles. Unlike a first-home-owner grant, which hands a buyer cash with no mechanism for the taxpayer to recover it, Help to Buy's equity structure means the Commonwealth's contribution is genuinely repaid, with its proportional share of any capital gain, when the home is eventually sold or the buyer chooses to repurchase the government's share. That is a more fiscally disciplined instrument than the blunt grants it replaced in the political conversation, even where its effect on the price of the home at the point of purchase resembles the effect economists attribute to those older grants.

The honest caveat is that both funds are still early in their delivery cycle. The HAFF's first full disbursement round only began in 2024, and Help to Buy's first purchases settled later still; there is not yet enough delivery data to say with confidence whether either program will hit its published five-year or annual targets, let alone whether delivery accelerates as community housing providers and state agencies build capacity to use the funding faster. The critique above is a critique of scale and structure, established on the funds' own published targets — not yet a verdict on execution, which the next two to three years of Housing Australia reporting will settle more reliably than any analysis written today.