The Illusion of Property Gains — And the Tax System That Gets It Wrong
By Aiden Garrison
Series note: This series examines how decisions are actually made — not to excuse outcomes, but to understand them accurately.
Worked example provided for illustrative purposes only.
There is a basic question at the centre of property taxation that is rarely asked properly.
What is the gain?
Not the difference between purchase price and sale price. Not the number on a spreadsheet at the point of exit. The real gain — the increase in purchasing power after inflation, after financing, after time.
Once that question is asked honestly, the current system starts to look less like policy and more like approximation.
Take a straightforward case. A property is purchased for $1 million and sold 20 years later for $3.87 million. On paper, the gain is $2.87 million.
That number is not real.
Inflation at 3 per cent compounds across two decades. The original $1 million is worth around $1.81 million in future dollars. Strip that out, and the gain falls to approximately $2.06 million.
Then account for financing. With an 80 per cent interest-only loan at around 5 per cent over the holding period, interest costs total roughly $600,000. That reduces the real economic gain to around $1.46 million.
That is the number that matters to the investor — actual wealth created, after inflation, after the cost of money.
Everything else is distortion.
There are, broadly, three ways to tax this outcome.
The first is the original capital gains framework introduced under Paul Keating. It indexed the cost base for inflation and taxed the real gain in full. No discount. No approximation.
Under that approach, the indexed cost base is $1.81 million, producing a real gain of $2.06 million. Add back depreciation claimed during the holding period — building allowance and plant deductions across 20 years on a residential asset of this size would typically total around $240,000 — and the taxable gain reaches approximately $2.3 million. At the top marginal rate of 47 per cent, the tax liability is around $1.08 million.
The second is the current system. There is no indexation on assets acquired after September 1999. Instead, a 50 per cent discount is applied to the nominal gain. After depreciation recapture, the taxable gain is approximately $1.56 million and the tax liability around $733,000.
The third is a modernised version of the original — index the cost base, allow interest and depreciation deductions during the holding period, recapture depreciation at sale, and tax what remains. Run properly, it produces almost exactly the same outcome as the original indexed model. Both are attempting to measure the same thing.
Capital Gains Tax Comparison
$1 million property, held 20 years, sold for $3.87 million
Real economic gain to investor
Taxable Gain: $1.46M
Approx. Tax Liability: —
Notes: After inflation (3%) and interest-only financing at ~5%
Original indexed system (single name)
Taxable Gain: $2.30M
Approx. Tax Liability: $1.08M
Notes: Real gain plus depreciation recapture, 47% top rate
Current 50% discount system (single name)
Taxable Gain: $1.56M
Approx. Tax Liability: $733,000
Notes: No indexation
So far, the comparison appears straightforward. The indexed system extracts about $348,000 more tax on this single asset than the current discount. On its face, the existing system looks generous to investors.
That conclusion does not survive contact with how property is actually held.
The $1.08 million figure assumes the asset is owned in a single name. Most long-term property at scale is not. It sits inside discretionary trusts, with gains distributed across multiple beneficiaries — spouses, adult children, sometimes corporate entities at the company rate.
Apply the same indexed system to a trust distributing $2.3 million across four adult beneficiaries. Each receives approximately $575,000. Spread across marginal brackets, with planning, the effective rate falls considerably. The total liability can land close to $730,000.
In a similar range to the current discount system.
That is the point most worth understanding.
The headline reform — replacing the discount with an indexed real gain system — sounds like it would close a gap.
It does not. Not where it matters.
For the investor holding in their own name, with no trust, no structuring, no distribution flexibility, the change is real. They pay an additional $348,000 on this single asset. Multiplied across a market of long-term holders, the additional revenue is substantial.
For the investor holding through a properly structured trust, much of that increase can be absorbed. The system can still produce a comparable after-tax outcome in many cases. The mechanism shifts from a formula to a structure, and the destinations often converge.
This is what reform consultations rarely surface. The cost of moving from approximation to precision is not borne by those most able to absorb it. It is borne by those who cannot afford the structures that neutralise it.
The current system survives because it is simple.
It avoids tracking inflation across decades. It avoids reconstructing cost bases. It avoids disputes about how gains should be measured. It replaces precision with a rule of thumb — half the gain is real, half is not.
In periods of moderate inflation, the rule lands close enough to the right answer for political purposes. In other environments, it does not. When inflation rises, it overtaxes. When inflation falls, it undertaxes. When leverage is high, the gap widens.
It is not designed to be correct. It is designed to be workable.
The objection to a real gain system is predictable. It introduces complexity. Records to maintain. Cost bases to verify. Disputes at the margin.
All of that is true.
But it is also incomplete.
The system is already complex for anyone operating at scale. Developers, large investors and institutional capital already track these variables in detail. For that segment, complexity is not the issue. Accuracy is.
What exists today is a system that is simple for small investors and approximate for large ones. The result is a blunt instrument applied across very different circumstances — one that, by accident or design, hits hardest in the middle.
A return to an indexed, real gain system would, in most cases, increase tax revenue. Not because it captures new forms of income, but because it measures gains more precisely.
But where that revenue comes from matters more than the headline figure.
It would fall most heavily on long-term investors holding assets in their own name — the segment most exposed to bracket creep, limited structuring flexibility and the cumulative effects of inflation. The middle of the market.
At the upper end, where assets are held through trusts or company structures, the outcome can be managed. In many cases, the after-tax result under a real gain system would resemble the after-tax result under the current discount.
That similarity is not a coincidence.
It reflects the fact that any system tax engineers can plan around will, eventually, settle into roughly the same outcome at the top.
The issue is not whether property gains should be taxed. They are. They will continue to be.
The issue is who actually pays when the rules change.
A reform that replaces a flat discount with measured precision sounds like it tightens the system. In practice, it tightens it on the segment with the least capacity to respond.
The top of the market is insulated by structure. The trusts, the distributions, the company rates — they do the work the discount used to do. Call it tax planning. It functions as welfare for those who can afford the accountants who set it up.
The people on the other side of that arrangement are not the wealthy. They are working people who saved for one asset, held it in their own name, and have no accountants on retainer. They wear the difference.
They always do.
That is not a flaw in the reform.
Whether by intent or not, it is the design.
