Ask an investor how much equity they have and most will give you a single number, stated with the confidence of someone reading their own bank balance. It isn't a balance. It's an estimate somebody else will make, on terms somebody else sets, released to the extent somebody else allows. Confuse the two and you end up planning a purchase around a number you never actually controlled.
This edition is about what equity actually is, and where it stops behaving the way most investors assume. The last two editions looked at two ways the numbers get misread: how a family home quietly absorbs the capacity you need for the next purchase, and how yield and growth answer different questions depending on where you are in a portfolio's life. This is the mechanism underneath both of them, the thing you're actually drawing on when you release equity to fund the next purchase, and the four places it stops behaving like money already in your account.
The first gate is the valuation. What you think the property is worth and what a lender's valuer says it's worth are two different questions, and only the second one turns into anything usable. Your own estimate, built off recent local sales or a portal figure, is a starting point for a conversation. It is not the number that clears.
Once a valuation exists, there is a second and entirely separate gate: serviceability. A valuation supporting a release does not mean a lender will extend against it. That depends on income, existing commitments, and the debt-to-income calculation covered below, a live constraint rather than a formality.
Key insight: paper equity clears one gate. Usable equity has to clear both, and different parties are deciding each one.
From 1 February 2026, APRA allows authorised deposit-taking institutions to lend up to 20 per cent of their new mortgage lending at debt of six times income or more. The limit applies separately to owner-occupier and investor lending, new dwellings are exempt, and APRA itself expects the constraint to bite harder on investors, who typically carry higher debt-to-income ratios than owner-occupiers.
The money is priced differently too. RBA figures for July 2026 put the discounted variable rate at 7.13 per cent for investors against 6.80 per cent for owner-occupiers, a 33 basis point gap on the same product. On the standard (non-discounted) interest-only investor product, a different rate series again, the figure runs materially higher again, at 9.59 per cent.
Key insight: the number that decides how much of your equity becomes usable is a funding cost and a prudential limit sitting behind the lender's decision, not a percentage you can read off your own numbers.
The ATO's use test, established in FCT v Munro and set out in TR 95/25, is unambiguous: deductibility of the interest is decided by what the borrowed funds are used for, not by what secured the loan. The interest on money released against the home and used to buy an investment property is deductible because of what the money did, not because of where it came from.
A redraw is treated as a new borrowing in its own right (TR 2000/2), so the same use test applies to it separately, each time. And where redrawn funds are split between income-producing and private use, the interest attributable to the private portion loses its deduction, an apportionment that compounds every year the loan runs. That is the mechanical reason a facility carrying both an investment purpose and everyday spending becomes harder to account for the longer it exists.
Key insight: the deduction was never attached to the house you borrowed against. It's attached to what the money did after it left the account, tested fresh on every draw.
From the 2027-28 income year, rental losses on established dwellings contracted after 7:30pm ACT on 12 May 2026 are quarantined under new section 26-155. They no longer offset salary in the year they occur; instead they carry forward against future rental income or capital gains. That is a cashflow and timing question, not a loss of the deduction. The deduction still exists. It arrives later, and against different income.
New residential dwellings keep both negative gearing and the 50 per cent CGT discount.
Worth naming plainly: APRA's new debt-to-income limit exempts new-dwelling lending from its 20 per cent cap, and the tax settings above favour new dwellings on negative gearing and the discount as well. Two regulators, working on entirely separate questions for entirely separate reasons, have arrived at the same tilt without coordinating with each other.
Key insight: that is a fact about the settings as they now stand, not a case for buying new. The tilt exists whether or not it changes what you do with it.
Sam is 41 here, a few years past the starting position in Issue 19 and well before the upgrade decision in Issue 22 or the harvest problem in Issue 23. One investment property, bought for 650,000, and he wants a second. (Illustrative numbers throughout.)
He'd priced his own equity off recent sales in the street, landing around 900,000. The bank's valuer came back at 845,000, a 55,000 gap that exists purely because his estimate was never the one that counted. Against 480,000 owing, that leaves paper equity of 365,000 on the number that actually clears.
He and his wife earn 195,000 combined; at six times income that puts their household debt near 1.17 million, the multiple APRA is watching. That threshold doesn't cap what he personally can borrow. It caps how much of a bank's total new lending can sit at six times income or more, which rations loans like his rather than forbidding them outright. Whether his own application clears is a separate question, one his equity figure can't answer, and he now checks it as its own gate rather than assuming the valuation was the only test.
For the release itself, he opens a separate facility for the amount rather than redrawing into the account he uses for bills and everyday spending, so the interest on it traces to one thing: the deposit and costs on property two. And because he exchanged contracts on the second property after 7:30pm ACT on 12 May 2026, the contract date, not settlement, being what fixes acquisition under that rule, he prices in that any rental loss on it from 2027-28 sits against future rental income and gains rather than his salary, a number to build into the cashflow rather than a reason to wait.
Three months later, property two settles. Sam's portfolio holds two properties and three loans: the original mortgage, the equity-release facility, and the new loan on property two, each one doing exactly one job, and the number he built the second purchase around was always the bank's, never his own.
Equity is a strange word because it sounds like ownership and behaves like an offer. It's real. It's also entirely contingent, on a valuation someone else assigns, a rate someone else sets, and a limit someone else is watching that has nothing to do with your own spreadsheet. None of that makes it useless. A high credit limit is a useful thing to hold, and equity behaves like one: it isn't a balance, and the investors who get hurt aren't usually the ones who release equity. They're the ones who planned a purchase around a number that was never fully theirs to spend. Know how it behaves before you build the next decision on it.
See you next week. — Alex
Want to see what your current position actually supports? The Property Portfolio Gap Analysis walks through where you are, where you are going, and the distance between them, in about five minutes. Find it at pbco.com.au.
This article is for educational purposes only. It does not constitute financial, legal, or tax advice. Everyone's circumstances are different. Please seek professional advice before acting on any of the strategies outlined above.