Negative gearing reform meets the housing problem it can't solve alone

The first time I tried to explain negative gearing to a friend over a beer at the Bondi Royal, she stopped me halfway and asked, “So the government is paying landlords to lose money?” Sort of, I said. The deduction is real. The loss is real, the way it sits on a tax return at the end of June. And the taxpayer-funded part is real too, because every dollar an investor offsets against their salary is a dollar the rest of us cover at the till.

It is absurd. It has also been the spine of Australian housing policy for the better part of a century. Tuesday night, give or take, that changes.

Jim Chalmers will hand down the 2026 Budget at about half past seven, and the Treasurer has stopped pretending. Treasury has been modelling two specific tax shifts that reshape who profits from holding rental property in this country. The capital gains tax discount, currently a flat 50 percent for assets held longer than a year, is reportedly headed to 33 percent. Negative gearing, which today has no upper limit on how many properties a single investor can offset against their salary, is reportedly being capped at two. Existing investors are likely to be grandfathered. That is the polite Canberra term for letting the people who already won the game keep their winnings.

I should say up front, I might be wrong about the precise shape of all this. The package isn’t released until budget night and budget leaks tend to be more aspirational than accurate. But Chalmers’s framing for the past fortnight has been unusually direct for a Treasurer the week before the speech. “From our point of view, the status quo in the housing market and the tax system is unfair, and that makes it unacceptable to us.” When a Treasurer who spent two years insisting reform was off the table now reaches for the word unacceptable, you stop ignoring him.

Two reforms in the room

About 1.2 million Australians use negative gearing. ATO data shows more than half of all investment properties operate at a paper loss, with the average annual deduction sitting at $8,702. Run those numbers across the Parliamentary Budget Office’s most recent costing and the combined cost of negative gearing plus the CGT discount lands at over $100 billion across the next decade. That is more than this country spends on aged care over the same period. Read the figure again if it doesn’t land.

The CGT discount is the bigger ticket. Chris Evans, an emeritus tax professor at UNSW, calls it “one of the biggest giveaways in the budget” and puts the cost at up to $23.7 billion a year. He is also unimpressed by the rumoured 33 percent compromise. “A reduction in the amount of the discount is a step in the right direction,” he wrote last month. “But all that does is tinker with the problem; it does not solve it.” He’d prefer phased abolition, which is the kind of academic position you can hold when you don’t have to win marginal seats.

The Grattan Institute, less ideological and more arithmetical, has been making the case for a 25 percent CGT discount phased in over five years for a decade. Its housing director Brendan Coates puts the revenue from a CGT and negative-gearing package at around $5.3 to $6.5 billion a year, and points to RBA modelling that estimates the two concessions together push prices up by one to two percent in the short run and as much as four percent over the long run. That four-percent number is the one I would ask people to sit with for a moment. On a Sydney median house price of around $1.6 million, four percent is sixty-four grand. That is a deposit.

Where the air goes thin

I used to think reforming these two tax breaks would do most of the housing-affordability work for us. It will not. The same Grattan modelling that supports the case for change estimates the price effect at one to two percent in the short term. It is a real number. It is also a small one. What the reforms shift is composition, mostly: fewer investor buyers competing with first-home buyers at auction, more households able to step in, a modest reduction in the rate at which housing wealth concentrates among the people who already own.

That isn’t nothing. It also isn’t the same as solving the problem.

The actual problem, as anyone who’s spent a Saturday queueing for a unit inspection in Marrickville knows, is supply. National rental vacancies, per SQM Research early this year, sit at 1.1 percent. In a healthy market the figure is closer to three. Australia has not built enough housing in close-in metro suburbs for two decades. Tax reform doesn’t change that. You cannot make landlords build more apartments by making it less profitable to hold existing ones.

The Property Council, which I disagree with on most things and not always wrongly, leans on a version of this point. Cut the incentives, they argue, and some landlords will sell. A Money.com.au survey of investors found 39 percent would step back if the CGT discount fell, and another 22 percent if negative gearing were capped at one property, for a combined 61 percent signalling some kind of pullback. It sounds dramatic until you remember that a household selling its investment property doesn’t make a house disappear. Almost always, it becomes someone else’s home, often a first-home buyer who’d been losing at auctions for a year.

The supply story is harder, slower, less satisfying. It is zoning fights in suburbs that haven’t seen a new typology since the second Hawke government. It is missing-middle housing on quarter-acre blocks. It is the patient grinding work of getting state planning systems to approve apartments faster than ten-year-olds can grow up. Federal tax policy can nudge any of that. It cannot drive it. The fix that would actually move the needle, the one housing cooperatives keep proving in miniature, is more dwellings built faster, closer to where people work.

What a serious budget would do

Pair the demand-side reform with serious supply moves and the picture starts to add up. Without the supply moves, you get a fairness fix and a press conference. With them, you get something that bends the next decade of housing policy toward people not yet born.

What I want to see on Tuesday is a fresh $2 billion infrastructure allocation tied to the National Housing Accord and the 65,000 dwellings the federal government has been promising for two years now, paired with build-to-rent tax incentives for institutional developers in the capital cities where the rental crunch is sharpest. And if Chalmers wants to be brave, he can offer the states the CGT savings to bankroll stamp duty abolition along the lines the WA budget has already started signalling. Coates has been pushing the stamp-duty trade for a decade. It would do more for housing affordability than any single tax change Treasury is currently modelling.

The political theatre on Tuesday will be about percentages and grandfather dates. The substance, the part that decides whether any of this matters in three years, will be about whether the demand-side measures arrive paired with supply ones, or land alone.

Where you fit in

For investors who already hold property, the grandfathering means existing arrangements stay, the CGT discount on assets held before budget night is preserved, and any new investment after the cut-off plays under the new rules. Owners of three or more properties care a fair bit more, since the two-property cap is the change to track. Quarantining the losses on properties three through whatever doesn’t take the deductions away. It just stops you using them to reduce your salary tax. That’s a meaningful cost increase for most professional landlords, but it isn’t a portfolio death sentence.

For someone trying to buy a first home, the honest answer is that the changes help, marginally. You should see slightly less investor competition at lower price points and a small dampening on price growth. You will not see the affordability problem solved by 1 July, regardless of what the Treasurer says on Tuesday.

For renters, who are the people I would ask to read this with the most caution, the politics here can mislead about timing. The ACTU’s Michele O’Neil put it as plainly as anyone last week. “The tax system should not provide the means for professional landlords to have their housing speculation subsidised by ordinary workers, who are being priced out of housing.” Whether any of that translates into rent relief depends entirely on what happens with supply. If state governments are paying attention and the federal government keeps building, you’ll feel it in two or three years. If they don’t, the next budget rolls around and we have this conversation again.

The signal worth watching on Tuesday isn’t a percentage. It is whether Chalmers tags his tax changes to a larger housing-supply package or hands them down on their own as a fairness fix and hopes the supply side sorts itself out. He’s been hinting all week that he knows which version is policy and which is theatre. We’ll know which one he chose by Wednesday morning.