Vault

Economics Explained — “Australia Might Have Solved Its Housing Crisis” (Negative Gearing Ends)

EE argues Australia’s May 2026 budget took the tax fuel out of its housing market by quarantining negative-gearing losses on established properties. Prices are already falling, and compulsory voting is what made the reform politically possible.

Date: reviewed 2026-10-05 Source: YouTube — Economics Explained, QtZfbWOkoIY (14:04) — Transcript (clean) Format: single-narrator explainer; sponsor read 1:55–3:07 Vault relevance: Bilyeu × EE “Wealth Is Fiction” (Chris: housing is macro, not just zoning; r > g as a risk premium), r vs g, Protection Against the Gradient (grandfathering; who gets to vote), Positional Goods

Argument Structure (as presented)

Two kinds of dirt. Australia exports minerals and trades houses among itself at ever-higher prices. Household debt is the 2nd-highest in the world, after Switzerland. (0:00–0:42)
Negative gearing. With a 47% top rate above A$190k, rental losses (interest, plus 2.5%/yr building depreciation that “never costs a dollar”) could be deducted from wage income. About 2.2M investors, half negatively geared, claimed >A$10B in losses in a year. (3:07–5:07)
Why it compounded. Limited land and restrictive zoning meant tax-driven demand bid up existing houses. Housing is now ~A$12.8T, about 3× the A$4.4T held in superannuation (pension savings). The big four banks write ~75% of mortgages; super funds hold bank shares; “the whole system fed itself.” (5:07–6:37)
The cost. Capital flowed to land instead of businesses, so the route to wealth became financing and selling property. The 2010s were the slowest productivity decade in 60 years. Net overseas migration peaked at 538k, and every arrival needs housing. (6:37–7:52)
The reform (budget night, 12 May 2026). From 1 Jul 2027, losses on established properties bought after budget night can’t offset wage income. New builds are exempt; a knockdown-rebuild counts only if it adds dwellings. Everything held before budget night is grandfathered for as long as the current owner keeps it, which EE calls “pulling up the ladder.” (7:52–9:47)
Why now: political maths, not economics. Labor lost in 2019 on the same policy. By 2025 a locked-out generation outnumbered ~2.2M investors among ~18M voters. Compulsory voting (90.7% turnout) counts young renters who would normally abstain. Labor won its biggest majority since 1943. (9:47–11:16)
Early results. Cotality national index −0.7% in July (largest drop since Dec 2022), Sydney −4% in 3 months, clearance rates below 50%. Investors are leading the retreat. There are buffers: older owners hold small mortgages and <1% of mortgaged households were in negative equity. The likely losers are recent 5%-deposit buyers. Treasury expected only a couple of percent off growth, and the market reacted more sharply. “An orderly slowdown rather than a crash”, but early days. (11:16–13:15)

Fact-check (subagent, 2026-10-05)

The figures mostly check out. The causal story does not.

Claim	Verdict	Actual
Budget mechanics (from 1 Jul 2027; new builds exempt; grandfathering)	TRUE	Losses on established properties bought after budget night offset only property income. The CGT rework EE “left alone” replaces the 50% discount from 1 Jul 2027 with inflation indexation plus a 30% minimum tax on gains, and CGT is not grandfathered. (budget.gov.au)
Prices falling because of the change	MISLEADING	*The peak was March and falls began in April, before* the 12 May budget. The RBA hiked in Feb, Mar and May, then 29 Sep: +100bp to 4.60%, the highest since 2011.** Cotality’s Sept commentary cites rates, affordability and sentiment, with no mention of the tax change. No analyst attributes the fall mainly to negative gearing. (RBA)
Cotality July −0.7%; 1.4% below peak	Mostly true / wrong figure	July −0.7% (later revised to −1.2%); actually 2.0% below peak. August −0.9%, Sept −1.1%: six straight falls, 5.2% below peak. Sydney −8.6% from its Feb peak. (Cotality Oct)
Investor loans −5.3% (Mar qtr)	True but misleading as evidence	The quarter ended before the budget and was still +18.8% y/y. June quarter −8.6%. (ABS)
Treasury: ~2% off price growth	MOSTLY TRUE	Chalmers: “about 2 per cent slower” (~$19k on a median house); CBA: just under 3%. No published Treasury paper.
Depreciation “never costs a dollar”	MISLEADING	Div 43 claims reduce the CGT cost base, so they are recaptured on sale at the discounted rate. It is a deferral plus a rate arbitrage, not free money. (ATO)
Household debt #2 after Switzerland; housing A$12.8T; super A$4.4T	TRUE / stale	~114% vs ~123% of GDP (BIS); A$12.77T. Super is now A$4.77T (June 2026).
2.2M investors, half negatively geared, >A$10B losses	TRUE (2022-23)	2,261,080; 49.4%; ~A$10.4B
47% top rate; NG paused 1985–87; NOM 538k; slowest productivity decade in 60 yrs	TRUE	NOM peak 2022-23; PC 2023 inquiry (~1.2%/yr)
Labor 2025: biggest majority since 1943; turnout 90.7%	TRUE	94/150 seats, best share since Curtin 1943
<1% negative equity	TRUE	RBA FSR Mar 2026. A further 20% fall would leave ~5% of mortgages in negative equity.
5% deposit First Home Guarantee	TRUE, omitted effect	Uncapped from Oct 2025, so it props up first-buyer demand, working against the price fall.

Net: the policy isn’t in force until July 2027. Prices turned before it was announced, under 100bp of rate hikes. The government’s own scheme is pushing first-buyer demand up. “Might have solved it” credits a tax change with a fall that rates explain at least as well.

Discussion seeds

“Might have solved” vs EE’s own “too early to tell”. The headline outruns the body. What would count as solved: lower prices, more homes built, or less debt?
Demand-side fix, supply-side problem? EE names limited land and zoning, then the reform only removes a demand subsidy. Does that bear on Chris’s earlier “housing is macro, not zoning” position?
Grandfathering is protection against the gradient. Incumbents keep their insulation while newcomers lose it. And compulsory voting is the mechanism that let the losers outvote the protected pocket, the same who-gets-counted logic as the immigration coalitions.
Confounders. Is the price fall the tax change, or interest rates and other 2026 moves? (Fact-check pending.)
“Depreciation never costs a dollar.” Is it really free, or is it clawed back through capital gains tax on sale? (Pending.)

Discussion

Chris (2026-10-05) — clickbait again; rates explain it better.

“He is blaming a policy that wasn’t announced until after prices started falling and that doesn’t go into effect yet as a reason? There might be some speculation, but probably not of this magnitude.”

Chris’s questions on the mechanics (answered below):

Did they just remove the ability to deduct interest and depreciation?
Does it apply to every institution that owns property, i.e. are interest and depreciation no longer expenses?
If a business can still deduct them, won’t landlords just incorporate and treat it as a business expense?
Chris’s suspicion: “it sounds like it has a negative impact on new rental property, not an increase in supply.”

Answers (Claude, from budget.gov.au and Baker McKenzie):

Interest and depreciation are still expenses. The change is loss quarantining, not removal of deductions. Expenses still offset rent. Only a net loss is barred from offsetting wages. It can be used against residential rental income or residential capital gains, and carried forward indefinitely. Budget text: “won’t be able to deduct them against other income like wages.”
US analogue: this is essentially the US passive-activity-loss rule (Tax Reform Act 1986). Rental losses are “passive” and can’t offset wages, apart from a $25k allowance phased out at $100–150k income and a real-estate-professional exception. (Claude, from memory.) Australia is adopting, for new purchases, what the US has done since 1986.
Who it covers: established residential property bought after budget night. Excluded: widely held trusts, superannuation funds, build-to-rent, and investors in government housing programs. Companies are not explicitly mentioned.
Incorporating doesn’t help. A company is a separate taxpayer, so its rental loss could never offset a shareholder’s wages; incorporating would only trap the loss inside the company. Australian companies also don’t get the CGT discount. (Claude, from memory. Unverified: whether a company with other operating income may offset rental losses against it under the new rule.)
On supply, Chris’s suspicion partly holds.
- For supply: new builds keep the full benefit, so the subsidy is redirected to construction rather than removed, which is directionally pro-new-supply. Institutional build-to-rent and super funds are untouched.
- Against:
  - Investors leaving established stock sell mostly to owner-occupiers, often former renters, so rental supply and rental demand shrink together and the net effect on rents is ambiguous.
  - Builders and tradespeople are already constrained (EE says so itself). The new-build premium may be capitalized into new-build and land prices rather than producing more homes.
  - A 30% minimum tax on gains lowers the expected return on all new investment, new builds included.

Chris’s close (2026-10-05) — “it makes sense now.”

The personal analogue: Chris once got a tax surprise when he found his stock-market loss was capped against his income. (Claude: the US caps net capital losses at $3,000 a year against ordinary income, with the rest carried forward.) “Now I see that they are just trying to separate the two”: earned income on one side, investment (“passive”) income on the other.
His prediction:

“Savvy investors will find a way to separate their income from this ‘passive’ income so it will net out.”

The rule shifts behaviour, not totals. Claude notes the routes are visible in the budget’s own exclusions: buy new builds, hold through super funds or widely held trusts (both excluded), or go build-to-rent. The US version produced the same responses through real-estate-professional status and cost-segregation studies. The burden lands on the unsophisticated investor, the one-property landlord. That is the protection-against-the-gradient pattern again: the well-advised find the gap.