The 2026–27 Budget replaces Australia's 50% Capital Gains Tax discount with cost-base indexation and a 30% minimum tax on capital gains, from 1 July 2027. Most headlines have framed this as a substantial tax rise on investors. Read the transitional arrangements more carefully and a different picture emerges: the reform creates three different regimes depending on when you bought and when you sell, and for long-held investors the maths may actually run in your favour.
This piece walks through what changes, how the apportionment works for assets held across the cutover, and what the numbers actually show.
From 1 July 2027:
The transitional arrangements are the part most coverage has skipped. They're also the part that determines which "regime" applies to your specific situation.
The transitional arrangements create three distinct cases. Which one applies to you depends on two dates: when you bought the asset, and when you sell it.
Nothing changes. The 50% discount applies in full. Tax is calculated on half the gain at your marginal rate.
Full new regime. Your cost base is indexed by CPI from purchase to sale. The indexed gain is taxed at your marginal rate, with a 30% floor (waived if you're on income support).
This is where it gets interesting. Your gain is apportioned into two parts:
The split is determined by the asset's value at 1 July 2027. Two methods are available to establish that value:
You choose per asset. The choice matters when your asset's actual growth was concentrated in one period rather than spread evenly over the holding period.
Assets acquired before 20 September 1985 ("pre-CGT") lose their full exemption only prospectively. Pre-cutover gain stays exempt; post-cutover gain falls under the new regime, with the asset's value at 1 July 2027 setting the new cost base. The exemption isn't being clawed back retroactively.
The ATO time-based formula uses pro-rata allocation. For an asset bought before 1 July 2027 and sold afterwards:
Value at 1 July 2027 = Cost base + (Sale price − Cost base) × (days held to 1 July 2027 ÷ total days held)
Take a property bought 1 July 2010 for $500,000 and sold 1 July 2030 for $1,000,000.
Applying the formula: Value at 1 July 2027 = $500,000 + ($1,000,000 − $500,000) × 0.85 = $925,000.
That number becomes the boundary. The pre-cutover slice of the gain ($425,000, before selling-cost apportionment) gets the 50% discount. The post-cutover slice gets indexation from the $925,000 baseline forward to the sale date, plus the 30% floor.
Indexation does the heavy lifting on that second slice. If you held from 1 July 2027 to mid-2030, that's roughly three years of inflation. At 2.5% per year, the indexed baseline would be approximately $925,000 × 1.077 = $996,500. The indexed gain on the post-2027 portion: a few thousand dollars or less. Often the indexed base exceeds the sale price and the post-cutover slice contributes nothing to your tax bill.
This is the counter-intuitive result that drives the next section.
Take the canonical scenario: investment property bought 1 July 2010 for $500,000. Sold 1 July 2030 for $1,000,000. Selling costs $15,000. Other income that year: $100,000. Not your main residence. Not on income support.
Under current law — if you sold in 2026, just before the cutover:
Under reform — same scenario sold in 2030, straddling the cutover:
The difference: reform pays $15,633 less than current law would have on the same asset.
The reason is straightforward once you see the apportionment. The 85% pre-cutover slice gets the 50% discount — identical to current law treatment. The 15% post-cutover slice gets indexation from a baseline of approximately $925,000 (the apportioned value at 1 July 2027). Project CPI to 2030 at 2.5% per year, and the indexed baseline reaches around $996,000. The remaining taxable gain on that portion is small — a few thousand dollars at most — and marginal tax applies at the top of your income stack, producing roughly $740 in tax on it.
This pattern is genuinely surprising the first time you see it. When the reform was first floated, most commentary assumed it would be a flat tax rise on long-held assets. The transitional arrangements changed that completely.
Using the calculator on the same scenario ($1M sale, $500k cost, $15k selling, $100k income, sold 2030), here's how the reform compares to current law across purchase years:
| Purchase year | Pre-cutover % | Current law (sold 2026) |
Reform (sold 2030) |
Direction |
|---|---|---|---|---|
| 2000 | 90% | $99,475 | $88,559 | Saves $10,916 |
| 2005 | 88% | $99,475 | $86,377 | Saves $13,098 |
| 2010 | 85% | $99,475 | $83,842 | Saves $15,633 |
| 2015 | 80% | $99,475 | $90,164 | Saves $9,311 |
| 2018 | 75% | $99,475 | $96,488 | Saves $2,987 |
| 2019 | 72.7% | $99,475 | $99,359 | Near break-even (saves $116) |
| 2020 | 70% | $99,475 | $102,816 | Costs $3,341 more |
| 2022 | 62.5% | $99,475 | $112,294 | Costs $12,819 more |
| 2024 | 50% | $99,475 | $128,116 | Costs $28,641 more |
| 2025 | 40% | $99,475 | $140,757 | Costs $41,282 more |
| 2027 | 2.5% | $99,475 | $191,299 | Costs $91,824 more |
Break-even for this specific scenario sits at around 2019 purchase year. Earlier than that, reform pays less. Later than that, reform costs more, and the cost grows steeply as the post-cutover slice gets larger.
This is one scenario. Different gain sizes, different incomes, different sale dates all shift the break-even year. The reform's impact for any specific asset depends on the interaction of: how long you held before the cutover, how long you'll hold after, your marginal tax rate, and the size of the indexed gain on the post-cutover slice. The calculator handles all four.
Every reform number above — the break-even table, the worked example, the $15,633 figure — assumes 2.5% annual CPI growth from now to the sale date. That's the midpoint of the RBA's 2–3% target band and is close to the long-term Australian average. But future CPI doesn't exist yet, and the result is meaningfully sensitive to that assumption.
Here's how the canonical scenario ($1M sale, $500k cost, $100k income, 2010 purchase, sold 2030) responds to different inflation assumptions:
| CPI assumption | Reform tax | vs current law ($99,475 baseline) |
|---|---|---|
| 1.5%/yr | $105,285 | +$5,810 (costs more) |
| 2.0%/yr | $94,669 | −$4,806 (saves) |
| 2.5%/yr (article baseline) | $83,842 | −$15,633 (saves) |
| 3.0%/yr | $83,102 | −$16,373 (saves) |
| 3.5%/yr or more | $83,102 | −$16,373 (saves, plateau) |
At 3% inflation or above, the post-cutover slice is fully absorbed by indexation and additional inflation can't help further — the number plateaus. Below 2%, the direction reverses: reform costs slightly more than current law even for long-held assets, because indexation no longer wipes out the post-cutover gain.
For a more recent buyer in the same setup (2024 purchase, otherwise identical), the inflation sensitivity is steeper because the post-cutover slice is much larger:
| CPI assumption | Reform tax | vs current law |
|---|---|---|
| 1.5%/yr | $145,500 | +$46,025 |
| 2.0%/yr | $136,893 | +$37,418 |
| 2.5%/yr (article baseline) | $128,116 | +$28,641 |
| 3.0%/yr | $119,165 | +$19,690 |
| 4.0%/yr | $100,736 | +$1,261 (near break-even) |
The pattern: higher inflation softens the reform's impact for everyone. Lower inflation tightens it. The 2.5% midpoint isn't a forecast — it's a working assumption near the centre of the credible band.
The calculator includes an editable "Inflation projection" field that appears whenever indexation is in play. Try your scenario at 2%, 2.5%, and 3% to see how much the answer moves. Treasury costings, RBA forecasts, and bond-market implied inflation rates all sit in roughly the 2–2.75% range right now, so 2.5% is a defensible middle — but the right approach is to test your specific scenario across a range.
For people in the "costs more" half of the table, the cost is real and substantial. The reform bites hardest on:
Recent buyers (2022 onwards). Most of your gain falls in the new regime. You lose the 50% discount on the majority of the asset's appreciation. The 30% floor adds further pressure if your marginal rate is below 30% on the top of the gain.
Future buyers. Pure new regime applies in full. Whether you save or lose depends on how cumulative inflation compares to the 50% discount over your holding period. At 2.5% inflation per year, indexation roughly matches a 50% discount somewhere in a 15-25 year holding range — earlier than that, the old 50% discount was better; later than that, indexation may pull ahead. There's a long tail of uncertainty here that depends on actual inflation outcomes.
Tax-deferral plays. The policy rationale most explicit in the Treasury commentary is the 30% floor's role: it stops taxpayers from sheltering gains by timing realisations in low-income years. If you've been planning to sell a big asset in a retirement year with deliberately low taxable income, the floor will catch most of it — except for income-support recipients, who are explicitly exempt.
Recipients of Age Pension, JobSeeker, Disability Support Pension and other income support payments are exempt from the 30% minimum tax. Marginal rates still apply to the indexed gain.
Using the calculator with low income ($25,000) and a modest gain under the new regime:
| Indexed gain | Non-pensioner | Pensioner | Saving |
|---|---|---|---|
| $30,000 | $5,963 | $3,180 | $2,783 |
| $60,000 | $14,963 | $12,162 | $2,801 |
| $100,000 | $26,963 | $24,162 | $2,801 |
The pattern: pensioners save in the $700–$2,800 range in typical low-income scenarios. The exemption matters most when the marginal rate on the top of your gain would otherwise be below 30%.
It's a narrower exemption than it sounds. For pensioners with significant assets — say a $1M+ investment property that wasn't a main residence — the gain itself usually pushes total income high enough that marginal rates exceed 30% anyway, and the 30% floor doesn't bite. The exemption is most useful for retirees realising moderate gains while staying in lower tax brackets.
Assets acquired before 20 September 1985 ("pre-CGT") have been completely outside the CGT system since CGT was introduced. The 2027 reform doesn't change that retroactively. It applies prospectively.
What that means:
If you hold pre-CGT assets and may sell after July 2027, one practical option is to obtain a formal valuation as close to 1 July 2027 as possible. That valuation determines how much of any future gain is taxable. A professional property valuation costs a few hundred dollars; for assets potentially worth millions, getting this right matters.
This is less of a "retrospective grab" than some commentary suggested. It is, however, the end of an unusual quirk in the Australian tax system, and worth planning around if you hold these assets.
Alongside the CGT changes, the Budget also limits negative gearing on residential property to new builds, applying to contracts entered from 7:30pm AEST on 12 May 2026. Existing investments are grandfathered.
For property investors, the two reforms interact:
Property is the asset class with the most policy attention here. For shares, managed funds, and other investments, only the CGT changes apply.
For someone holding investments and trying to think about what to do:
Timing. If you're in the "saves money under reform" half of the table, there's no rush to sell before 1 July 2027. If you're in the "costs more" half, selling before 1 July 2027 locks in the 50% discount on the whole gain — but only if you can sell at a price that holds up, and only if your other circumstances haven't changed meaningfully.
Valuations. For assets where the ATO time-based formula will materially understate or overstate actual market value at 1 July 2027, get a formal valuation. For shares, quoted prices at 1 July 2027 will be available automatically — no action needed. For properties that have appreciated unevenly over the holding period (e.g., a renovation that drove most of the value in one short window), a formal valuation may produce a more favourable apportionment than the time-based formula.
Retirees. Model your specific scenario. The interaction of marginal rates, gain size, and post-cutover holding period determines whether you're better or worse off. The pensioner exemption helps in narrow cases; don't assume it solves all your CGT problems.
The 12-month rule. Holding for 12+ months still matters for the pre-2027 discount on the pre-cutover slice. If you're close to that threshold, waiting may add value.
Companies, super funds, trusts. Different rules. Companies pay 30% on capital gains regardless of holding period; super funds pay 15% (or 10% if held 12+ months) regardless. The reform doesn't directly change either. Trusts are generally treated like individuals for CGT purposes and are affected.
The numbers in this article all come from the live calculator. You can run your own specific scenario:
It handles all three regimes automatically, runs the apportionment formula, and shows the pre-cutover and post-cutover breakdown for straddling assets.
Open the calculator →As at 13 May 2026. The reform was announced 12 May 2026 and final legislation is pending. Specific figures may change once draft legislation is released. The pre-CGT prospective treatment, apportionment mechanics, pensioner exemption from the 30% floor, and new-build election are all confirmed in primary Treasury and PM's office sources at the date of publication. Indexation projections use 2.5% per year as a working assumption (the calculator lets you change this); actual CPI from 2027 onwards is unknown and will affect every post-cutover number above.