CGT Reform for Crypto: Bitcoin, Ethereum and Tokens from 1 July 2027

General information only. This is not tax or financial advice. Consult a registered tax agent for advice specific to your situation.

Budget 2026 ended the 50% capital gains tax discount on cryptocurrency held more than 12 months. From 1 July 2027, individuals, trusts and partnerships disposing of Bitcoin, Ethereum, altcoins, NFTs and other crypto assets held as investments will be taxed under cost base indexation plus a 30% minimum tax on the real gain — the same regime that now applies to property and shares. Companies are unaffected (they already pay 30% on gains and never received the discount).

The short version: Bought your crypto before 1 July 2027 and selling after? Your gain splits into two parts — old half keeps the 50% discount, new half gets inflation-adjusted but pays at least 30%. Every coin you bought has its own buy-date — and the buy-date determines which bucket it lands in.

This article focuses on crypto held for investment. Tokens held under the personal use rule (if your crypto is worth less than $10,000 and you’re using it to buy things, not as investment) follow a different ATO regime that exempts certain disposals from CGT entirely, and the Budget 2026 reform does not change that carve-out.

If your crypto is a long-held BTC stack, an ETH staking position, an NFT collection, or DeFi liquidity, you almost certainly are in the investment bucket and the rules below apply.

Timeline — when the reform hits your crypto holdings

Date	What happens	What it means for you
12 May 2026	Budget 2026 announces the reform.	Crypto markets may move on the news. Integrity rules apply from this date for transactions designed to shift gains across the 1 July 2027 line — don’t try to manufacture an early disposal date for a parcel you’re not really selling.
From 12 May 2026	Tax software vendors begin updating to flag 1 July 2027.	Koinly, CryptoTaxCalculator, Syla, CoinLedger, and ATO bulk-data tools will add “pre-reform” vs “post-reform” parcel flags. Update your software to the latest version before 30 June 2027.
2026-27 income year (1 Jul 2026 – 30 Jun 2027)	Last FY where crypto disposals get the full 50% discount on the whole gain.	If you’ve held BTC, ETH or altcoins more than 12 months and were planning to take some profit anyway, this is your best tax-year window.
30 June 2027 (Wednesday, 23:59:59 UTC+10)	Last moment to lock in pre-reform treatment.	Crypto disposals are instant — no T+2 settlement. The block timestamp of your sell transaction is the disposal date. Plan around exchange withdrawal delays if cashing out to AUD.
1 July 2027	Reform start: cost-base indexation + 30% minimum tax begin for crypto.	Coins bought from this day are Bucket C; coins owned before are Bucket B. Every parcel acquisition date now matters.
Each staking reward post-1 July 2027	Each reward = new Bucket C parcel.	Staking on Ethereum, Solana, Cosmos chains creates micro-parcels by the second. Each reward gets its own cost base = AUD market value at receipt.
First crypto-to-crypto trade post-1 July 2027	Trade is a CGT event, even without AUD cashout.	Swapping BTC → ETH on a DEX or CEX after 1 July 2027 means the BTC gain is realised under new rules; the ETH becomes a fresh Bucket C parcel.
31 October 2028	First tax return applying reform rules to crypto disposals.	Self-lodging crypto investors filing 2027-28 will be first to use the new split-treatment system across (often) hundreds of parcels per year.
Mid-2030 onwards	DeFi/staking participants face exponential bookkeeping load.	Years of micro-parcels straddling the 1 July 2027 line make manual tax prep impractical — third-party tax software becomes essentially mandatory.

How time changes your tax bill

Crypto is the asset class most sensitive to the reform’s time math, for three reasons:

1. Holding-period variance is extreme. Many investors have 2017–2018 parcels worth 50× their cost base; others bought in 2024 and have modest paper gains. The legacy/reform split is dramatically different across these profiles — far wider than for property (where most parcels are 5–15 year holds) or shares (where most are 1–10 year holds).
2. Growth rates are very high. Bitcoin has historically returned 60–90% CAGR over 4–10 year windows. Indexation at 2.5%/yr CPI is a rounding error against that. For property at 5%/yr nominal growth, indexation recovers a big slice of the real gain; for crypto at 40%/yr, it barely registers.
3. Parcel proliferation is unique. Staking rewards, airdrops, hard forks, DEX swaps, and DCA-style buys create dozens of micro-parcels per year. Each has its own acquisition date and its own bucket classification. No other asset class generates parcels at the cadence crypto does.

Holding-period split — worked table

For a $100,000 nominal gain parcel, the legacy share of the gain (the slice that keeps the 50% discount under Bucket B day-count apportionment) depends entirely on when you bought versus when you sell:

Bought	Sold	Hold (yrs)	Legacy share	Reform share	Comment
1 Jul 2017	1 Jul 2030	13	76.9%	23.1%	Long-held legacy maximalist
1 Jul 2020	1 Jul 2028	8	87.5%	12.5%	DeFi-era buyer, near-term sale
1 Jul 2024	1 Jul 2032	8	37.5%	62.5%	Bull-cycle buyer caught by reform
1 Jul 2028	1 Jul 2034	6	0%	100%	Bucket C — full new rules

Plain-English takeaway: DeFi-era (2020–2022) buyers benefit most from selling soon after 2027 — they retain ~80%+ legacy share. Recent (2024 onwards) buyers face mostly reform-bucket treatment regardless of when they sell, because the pre-1-July-2027 slice of their hold is too short to dominate.

Indexation by years past reform (2.5%/yr CPI assumed)

For the reform-portion gain, the cost base gets indexed by CPI. Standard uplift factors:

Years past 1 Jul 2027	Cost base uplift
3 yrs	7.7%
5 yrs	13.1%
10 yrs	28.0%
15 yrs	44.8%

For crypto, indexation matters less than for any other asset class — because crypto growth rates dwarf CPI. A 10-year indexation uplift of 28% is a fraction of typical 5-year crypto returns. A property investor selling after a 10-year hold under reform gets meaningful relief from indexation; a Bitcoin holder gets a token gesture.

Crypto’s high-growth problem

Sample tax bill on an $80,000 nominal gain (10-year hold, 5 yrs pre-reform + 5 yrs post), at a 45% marginal rate:

Scenario	Annual growth	Old-rules tax	New-rules tax	Diff
BTC long-hold	40%/yr	$18,000	~$23,400	+30%
ETH multi-year	25%/yr	$18,000	~$21,600	+20%
Altcoin gambler	10%/yr nominal	$18,000	~$19,800	+10%

Plain-English takeaway: crypto’s hyper-growth means indexation barely dents the real gain. Reform-portion tax is brutal for fast-growing tokens. Slow-growth altcoins (or anything tracking inflation) fare slightly better but still cost more under reform than under the old 50% discount.

Staking parcels — a thousand mini-buckets

Imagine Tom stakes 32 ETH on Lido from January 2024. He receives ~0.32% APR weekly. Over 7 years to 2031 ≈ 365 staking-reward parcels. Each parcel has a market-value-at-receipt cost base. Each parcel falls into a bucket based on receipt date:

Receipt period	Bucket	What it means
Jan 2024 – Jun 2027	Bucket B	~180 parcels with varying legacy shares (depending on receipt date)
Jul 2027 onwards	Bucket C	~185+ parcels, all pure new-rules

Sell the whole stack in 2031 and your tax software must apportion every parcel individually. The earliest 2024 reward parcel has ~50% legacy share; the May 2027 reward parcel has well under 5%; every post-July-2027 parcel is 100% reform. Manual calculation is impossible — every weekly reward has its own day-count math.

Crypto-to-crypto trades trigger time-cuts mid-hold

Trading BTC → ETH on 12 March 2028 is a CGT event. Your BTC gain is realised under new rules at that moment. The ETH you receive becomes a fresh Bucket C parcel from that date. If you swap again in 2030 (ETH → SOL), another CGT event with another Bucket C parcel creation. Active traders on Solana, Ethereum, or Bitcoin L2s compound parcel proliferation each trade — every swap resets the clock on the destination token and locks the new acquisition firmly into Bucket C.

Stablecoins — time mostly irrelevant

USDC, USDT, AUD-pegged stablecoins: gain/loss usually ≈ AUD-fluctuation only. The time math is theoretically the same as for BTC/ETH but practical impact is tiny. Indexation on a cost base ≈ proceeds produces a near-zero adjustment. The 30% minimum rarely binds because gains are typically $0–$50 per swap. The compliance burden remains (each swap is a CGT event), but reform vs old-rules tax differences are negligible.

Personal-use crypto and time

If you bought crypto worth less than $10k for purchases (not investment) and used it within a reasonable period, the ATO personal-use rule may exempt the disposal entirely — regardless of when it falls relative to 1 July 2027. Time on its own doesn’t help you here; the use case does. Reform changes nothing about the personal-use carve-out.

Bottom-line summary

• 2017–2020 buyers selling 2027–2030: legacy share ~75–85%, reform impact modest (10–20% more tax than under old rules).
• 2024–2026 buyers selling 2032–2034: reform share dominates, expect 20–30% more tax than under old rules — and worse for high-growth tokens.
• Stakers and yield farmers: parcel proliferation makes tax software mandatory (Koinly, CryptoTaxCalculator, Syla, CoinLedger). Manual spreadsheets stop working around 100 parcels.
• Active traders: every CGT event creates fresh Bucket C parcels; long-term holding becomes more tax-efficient than churning, since each swap forfeits any accumulated legacy share on the destination token.
• Personal-use crypto: time irrelevant if you qualify for the exemption.

The three-bucket transition, applied to crypto

Treasury structures the transition the same way for every CGT asset class:

Bucket	Description	Treatment
A	Acquired AND disposed before 1 July 2027	Unchanged. 50% discount applies as before.
B	Owned before 1 July 2027 and disposed after	Split treatment. Pre-1 July 2027 portion of the gain keeps the 50% discount; post-1 July 2027 portion runs through indexation + 30% minimum.
C	Acquired on or after 1 July 2027	Wholly new rules. Cost base indexation + 30% minimum across the full holding period.

Crypto has one advantage over property and unlisted shares: parcel acquisition dates are typically known to the second from on-chain block timestamps or exchange trade confirmations. The split is therefore deterministic — if a parcel was received on 15 May 2019 and disposed on 15 December 2029, the pre/post-1-July-2027 day count is fixed.

There is no need to commission a 1 July 2027 valuation (as a property investor would) because the spot AUD price at 1 July 2027 is observable from any reputable price source. The Treasury split apportions the gain by days held rather than by mark-to-market value at the reform date, which matters when prices ran hard either side of 1 July 2027.

Bucket A worked example — sold before reform date

Sarah bought 0.5 BTC in March 2020 at AUD 14,000/BTC. Her cost base is $7,000. She decides to sell ahead of the reform on 25 June 2027 at AUD 190,000/BTC. Proceeds = $95,000. Nominal gain = $88,000. Holding period ≈ 7.3 years. Her marginal rate is 37%.

Because both the buy and the sell happened before 1 July 2027, the old rules apply in full:

• Discounted gain = $88,000 × 50% = $44,000.
• Tax at 37% = $16,280.

Plain-English takeaway: if you can crystallise a long-held parcel before the reform date, the legacy 50% discount applies cleanly with no apportionment, no indexation, and no 30% minimum check. For a high-conviction long-term hold, this is rarely worth doing on its own (Sarah pays $16,280 now versus letting the reform’s split treatment apply to a later sale). But for parcels you were already planning to sell — or for loss-harvesting candidates — 30 June 2027 is a hard deadline worth marking on the calendar.

Bucket B worked example — long-held Bitcoin

Daniel bought 2.0 BTC on 15 May 2019 at AUD 11,200/BTC. His cost base is $22,400. He disposes of the full 2.0 BTC on 15 December 2029 at AUD 240,000/BTC for proceeds of $480,000. Nominal gain = $457,600. Holding period ≈ 10.6 years. His marginal rate at disposal is 47% (top bracket).

Day count apportionment:

• 15 May 2019 → 1 July 2027 = ~2,969 days (legacy bucket).
• 15 May 2019 → 15 December 2029 = ~3,866 days (total).
• Legacy share = 2,969 / 3,866 ≈ 76.8%.

Legacy portion (pre-1 July 2027 share):

• Legacy gain = $457,600 × 76.8% ≈ $351,437.
• After 50% discount = $175,718.
• Tax at 47% = $82,587.

Reform portion (post-1 July 2027 share):

• Reform gain = $457,600 − $351,437 ≈ $106,163.
• CPI uplift over 2.5 years at 2.5%/year ≈ 6.4% (compounded). Indexation adjustment ≈ $106,163 × (1 − 1/1.064) ≈ $6,387.
• Real reform gain ≈ $99,776.
• Tax at max(47%, 30%) = 47% × $99,776 ≈ $46,895. The 30% minimum doesn’t bind because Daniel’s marginal rate is higher.

Total tax = $82,587 + $46,895 ≈ $129,482.

Comparison with old rules: $457,600 × 50% × 47% = $107,536. The reform adds about $21,946 of tax (+20%). The bulk of Daniel’s gain still falls under the legacy 50% discount because he bought the parcel nearly a decade before the reform date; only the last ~23% of his holding period falls under the new rules.

Model your own parcel in the Crypto CGT Calculator — it surfaces the legacy share, indexation adjustment, and minimum-tax binding flag separately.

Bucket C worked example — Ethereum bought after reform

Aiden buys 10 ETH on 5 September 2028 at AUD 7,500 each, for a total cost base of $75,000. He sells the lot on 5 September 2033 at AUD 14,000 each for proceeds of $140,000. Nominal gain = $65,000. Holding period = 5 years exactly. His marginal rate at disposal is 32.5% (the middle bracket).

Because both the buy and the sell happened on or after 1 July 2027, the new rules apply across the whole holding period — no apportionment needed.

Step 1 — index the cost base by CPI. Assume CPI runs at 2.5% per year over the 5-year hold. The compound uplift factor is 1.025⁵ ≈ 1.1314.

• Indexed cost base = $75,000 × 1.1314 = $84,855.
• Real gain = $140,000 − $84,855 = $55,145.

Step 2 — compare the marginal-rate tax to the 30% minimum.

• Tax at Aiden’s marginal 32.5% on the real gain = $55,145 × 32.5% = $17,922.
• Tax at the 30% minimum = $55,145 × 30% = $16,544.
• Aiden’s marginal rate is higher than 30%, so the 30% minimum doesn’t bind. He pays $17,922.

Comparison with old rules: $65,000 nominal gain × 50% discount × 32.5% = $10,563. The reform adds about $7,359 of tax (+70%). Indexation gave Aiden back some of the loss to inflation, but losing the 50% discount on a 5-year hold with strong real growth is a big swing. This is the “pure new rules” scenario — what most crypto bought from FY 2027–28 onwards will look like.

Low-income example — when the 30% minimum bites

Casey earns $42,000/year from a part-time job (her marginal rate is 18%). She bought 1 BTC in 2019 for $8,000 and sells it in 2030 for $185,000. Nominal gain = $177,000. Holding period ≈ 11 years. Most of the holding period is pre-reform.

Day count apportionment:

• 2019 → 1 July 2027 ≈ 8 years (legacy bucket).
• 2019 → 2030 ≈ 11 years (total).
• Legacy share ≈ 73%.

Legacy portion:

• Legacy gain = $177,000 × 73% = $129,210.
• After 50% discount = $64,605.
• The discount + her low base income pushes her into the 32.5% bracket once the gain stacks on top. For simplicity here, assume the legacy slice pushes her to a blended ~30% effective rate on the discounted gain. Tax on legacy ≈ $19,381.

Reform portion — this is where the 30% minimum bites:

• Reform gain = $177,000 − $129,210 = $47,790.
• CPI uplift over ~3 years at 2.5%/year ≈ 7.7%. Indexation adjustment ≈ $47,790 × (1 − 1/1.077) ≈ $3,416.
• Real reform gain ≈ $44,374.
• Tax at Casey’s marginal rate (18% on this slice, ignoring bracket-creep effects) = $44,374 × 18% = $7,987.
• Tax at the 30% minimum = $44,374 × 30% = $13,312.
• The 30% minimum binds. Casey pays $13,312 on the reform portion, not $7,987.

Plain-English takeaway: the 30% minimum is the part of the reform that hurts low-income holders the most. Casey’s marginal rate of 18% would normally apply, but the reform ratchets her up to 30% on the reform-portion gain — adding $5,325 of tax purely because of the floor.

If Casey were on a higher income (37% or 47% bracket), the 30% floor would be irrelevant — she’d pay her marginal rate anyway. The 30% minimum is designed to stop high-net-worth holders sheltering large gains by realising in low-income years (e.g. a sabbatical, parental leave, retirement). The unintended consequence is that genuinely low-income holders also lose the bracket-discount benefit on their reform-portion gains.

Staking-rewards example — the bookkeeping nightmare

Tom stakes ETH on Lido starting in January 2024. He receives roughly 0.1 ETH per month in staking rewards. By the time he decides to sell his entire ETH position in January 2031, he has accumulated 84 separate reward parcels — one per month — plus his original staking deposit.

Each reward parcel has:

• Its own buy-date (the day the reward was received).
• Its own cost base = market price the day you received them.
• Its own Bucket A / B / C classification based on receipt date.

When Tom sells in January 2031:

• The reward parcels received Jan 2024 → Jun 2027 (≈ 42 parcels) are Bucket B — they straddle 1 July 2027 and need split apportionment.
• The reward parcels received Jul 2027 → Jan 2031 (≈ 42 parcels) are Bucket C — pure new rules.
• His original staking deposit (Jan 2024) is Bucket B.

That’s 85 separate CGT line items, each with its own day-count math, its own indexation calculation, and its own 30% minimum check. The Bucket B parcels have wildly different legacy shares — the Jan 2024 parcel has ~74% legacy share, the June 2027 parcel has well under 1%.

Plain-English takeaway: there is no way Tom is doing this manually in a spreadsheet. The math itself is mostly arithmetic, but the bookkeeping is the killer. Crypto tax software has always been valuable for stakers; under the reform it becomes non-negotiable. Tom also needs to have declared each monthly reward as ordinary income at the market price the day he received them — that’s a separate income-tax obligation that happens in real time, not at disposal.

Why your tax software will save your bacon: Manual parcel matching across thousands of crypto trades is genuinely impossible. If you don’t already use Koinly, CryptoTaxCalculator, Syla, or similar — start before 30 June 2027. The split-treatment math gets harder with every passing year of staking rewards, DCA buys, and DeFi swaps.

Parcel matching becomes essential

Under old rules, matching which coins you’re selling to which coins you bought (parcel matching) was mainly about choosing FIFO vs LIFO vs specific identification to optimise the discount-eligible portion. Under reform, every parcel needs a “straddles 1 July 2027?” flag so the split apportionment runs correctly.

A long-term BTC stack accumulated through dollar-cost averaging may have hundreds of parcels — some pre-reform only (Bucket B), some entirely post-reform (Bucket C), some sold before the reform date (Bucket A).

Crypto tax software (Koinly, CryptoTaxCalculator, Syla, CoinLedger) will need to enrich parcel records with the 1 July 2027 indicator and emit the split apportionment per disposal. Expect provider updates through 2026–27 as the legislation finalises.

In the meantime, the ATO-acceptable cost base methods do not change. You can use:

• FIFO (sell-oldest-first) — the default, and what most software does automatically.
• LIFO (sell-newest-first) — sometimes useful for harvesting short-term losses.
• Specific identification (pick which coins you’re selling) — if you can demonstrate exactly which parcels were sold.
• Weighted-average (one average cost per coin) — for interchangeable (like dollars — one Bitcoin is the same as another Bitcoin) tokens with proper documentation.

Whatever method you used before reform, keep using it consistently — switching methods to chase the reform timing is the kind of thing the ATO will look at on audit.

Hard forks and airdrops — cost base questions

The ATO’s existing position is that tokens received via an airdrop or hard fork generally have a $0 cost base for CGT purposes. The receipt itself can be assessable as ordinary income at fair market value on the date of receipt — particularly if you are carrying on a business or holding crypto for personal investment purposes where the airdrop is treated as ordinary-income receipt. (This is a long-running area of ATO guidance via Tax Determinations; check the current TD references with your tax agent.)

Under reform, the date logic is unchanged. A 2024 airdrop of token X with a $0 cost base sold post-1 July 2027 follows the standard bucket logic: the receipt date is the CGT acquisition date, so a 2024-received airdrop sold in 2029 is Bucket B (split treatment). A post-reform airdrop received and disposed entirely after 1 July 2027 is Bucket C. The combination of $0 cost base and indexation is unfavourable — there is no nominal cost base to index up — so the entire post-reform gain becomes real gain that runs through the 30% minimum test.

Staking rewards and DeFi yield

Staking rewards (PoS validation rewards, liquidity-pool rewards, lending yield) are typically assessable as ordinary income at receipt at the market price the day you received them. That receipt-date market price becomes the cost base of the new tokens; the subsequent CGT event runs separately when those tokens are eventually disposed.

Each staking reward is a separate parcel with its own acquisition date — and this is where long-term stakers face a recordkeeping problem. A validator earning weekly ETH rewards over 5+ years accumulates hundreds of micro-parcels, many of which will straddle 1 July 2027. Each one needs its own cost base, acquisition date, and (eventually) its own split calculation when disposed.

Two practical implications:

• Reward parcels received before 1 July 2027 are Bucket B when disposed post-reform. The pre-1-July-2027 hold period on a reward parcel is the gap between reward receipt and 1 July 2027 — often quite short for late-cycle rewards.
• Reward parcels received on or after 1 July 2027 are Bucket C in their entirety. The reform applies across the full hold period.

DeFi liquidity-pool deposits raise a separate ATO question — whether the deposit itself is a CGT event (because you have given up the underlying tokens for an LP token). The ATO’s position remains evolving on this; conservative practice treats LP entry/exit as CGT events. Whatever you decide on the disposal side, the reform applies to LP token disposals exactly as it does to underlying-token disposals.

Stablecoin disposals

USDT, USDC, DAI and other AUD-priced stablecoin disposals are CGT events, but the gain or loss typically reflects AUD/USD currency drift rather than crypto-market movement. Reform applies, but two features mean the practical impact is small:

• Indexation rarely makes a material difference because cost base ≈ proceeds. There is very little nominal gain to index.
• The 30% minimum is unlikely to bind because the gains themselves are small, not because of indexation.

What stablecoin trading does generate is large numbers of micro-parcels through routine on/off-ramping and DeFi rebalancing. Each one is a CGT event under both old and new rules. The Bucket B split for any stablecoin parcels held across 1 July 2027 will produce tiny dollar amounts but still need to be recorded.

NFTs

A unique non-fungible token is a CGT asset like any other crypto. Acquisition date = mint date (for primary purchases) or buy date (for secondary-market purchases). Disposal triggers a CGT event at the AUD-equivalent of the sale proceeds (less platform fees and gas).

The reform applies identically. There is no NFT-specific carve-out, no “collectible” treatment in the BP2 measure, and no equivalent of the personal use rule (if your crypto is worth less than $10k and you’re using it to buy things, not as investment) unless the NFT was genuinely acquired for personal use (e.g. a profile-picture NFT used as a social identifier, under $10,000). Investment-held NFTs straddle 1 July 2027 the same way as ERC-20 tokens.

Crypto-to-crypto trades

Since 2014, the ATO has been clear that a crypto-to-crypto trade is a CGT event — you have disposed of one CGT asset (the source token) and acquired another (the destination token). Each trade triggers:

• A CGT event on the disposed token, at the AUD-equivalent of the proceeds (usually the AUD-equivalent of the destination token received).
• A new acquisition of the destination token, with cost base = AUD-equivalent at trade date + any platform/gas fees.

Under reform, a BTC → ETH trade on 5 June 2027 is a Bucket A event for the BTC parcel (disposed before 1 July 2027 — legacy 50% discount applies in full). The newly acquired ETH parcel has an acquisition date of 5 June 2027 (pre-reform) — when that ETH is later disposed post-1 July 2027, it falls into Bucket B with a very small legacy share (the ~26 days between 5 June and 1 July 2027 divided by the total hold period).

The same logic for a trade on, say, 5 August 2027: the source token is disposed under reform (Bucket B or C depending on its acquisition date); the destination token’s acquisition date is 5 August 2027, putting it firmly in Bucket C for any future disposal.

Worked example. Mia trades 5 ETH for 50,000 USDC on 12 March 2028. She bought the 5 ETH in 2022 at AUD 4,500 each ($22,500 total). At the time of the swap, ETH is worth AUD 10,000 each, so she receives USDC worth $50,000. This is a CGT event even though Mia hasn’t cashed out to AUD — her gain on the ETH is realised the moment the swap settles on-chain.

• Disposal proceeds for the ETH = AUD-equivalent of the USDC received at trade time = $50,000.
• Cost base of the ETH = $22,500.
• Nominal gain on ETH = $27,500.
• Because Mia bought the ETH before 1 July 2027 and disposed after, the ETH disposal is Bucket B — split treatment applies.
• The USDC she received now has its own cost base of $50,000 and acquisition date of 12 March 2028 — that’s a fresh Bucket C parcel for any future disposal.

Plain-English takeaway: every time you swap one crypto for another — including stablecoins, including layer-2 wrapped tokens — you trigger a taxable event on the token you’re giving up. Many investors are surprised by this. Under the reform, the rules don’t change but the consequences get more expensive on long-held positions.

Lost private keys and stolen crypto

The ATO permits a CGT event of “lost crypto” — broadly, you can claim a capital loss equal to the cost base of the asset where you can demonstrate the keys are irretrievably lost or the crypto was stolen. Documentation requirements are stringent: wallet addresses, evidence of attempted recovery, police reports for theft, blockchain evidence that the affected wallet has not transacted since the loss date.

Under reform, the loss can offset capital gains (legacy or reform portion) in the year claimed and any carried-forward amount can offset future gains. The reform doesn’t change how the loss itself is calculated — your cost base is your nominal AUD acquisition cost, not an indexed cost base. Gains are subject to the new rules; losses stay nominal. This asymmetry is one reason why locking in losses before 30 June 2027 is more attractive than it might first appear (see below).

Crypto-business vs crypto-investment

If you carry on a crypto trading business — high-frequency trading, market-making, professional arbitrage — your profits are ordinary income, not CGT gains. The reform is irrelevant to you because you never received the 50% discount in the first place. You pay your marginal rate on the full nominal gain regardless of holding period.

The 30% minimum tax also doesn’t apply because it is a CGT-specific rule. A trader on $25,000 of other taxable income who makes $10,000 of trading profit pays at the ordinary brackets (which can be below 30% effective). A long-term holder making the same $10,000 of CGT gain on a Bucket C post-reform parcel would hit the 30% minimum and pay $3,000 of tax — more than the trader pays at brackets.

If you have always treated yourself as an investor, the reform tightens your treatment relative to the trader. If you have always treated yourself as a trader, nothing changes — but you remain exposed to the existing question of whether the ATO agrees you are carrying on a business (the relevant factors are scale, frequency, sophistication, business-like operation, profit motive). Don’t reclassify yourself one way or the other to chase reform timing without professional advice.

Planning levers for crypto investors

Loss-harvest before 30 June 2027. Bring forward realised losses to the 2026–27 income year so they offset gains that still enjoy the full 50% discount or any final pre-reform Bucket A disposals. Carried-forward losses remain usable post-reform but offset gains taxed under stricter rules. The Crypto CGT Calculator and CGT Harvest Calculator can rank candidates by year-end deadline.

Audit your parcel records now. Every parcel needs an accurate AUD cost base and acquisition date — without that, you cannot run a Bucket B split. Long-term stakers, DeFi power-users, and traders with multi-wallet histories should reconcile their tax-software exports against on-chain data before the reform date, not after. Reconstruction in 2030 from incomplete records is much harder than reconciliation in 2026.

Consider whether to dispose pre-reform. For each material parcel, model the disposal under both rules. The reform raises tax on most high-return long-term holds (Daniel’s BTC above), but for low-return parcels (stablecoin gains, slow-growth altcoins) indexation can leave you slightly better off. Don’t blanket-sell.

SMSF holders — no separate carve-out. If your crypto is held in a self-managed super fund, the SMSF tax rates differ (15% accumulation, 0% retirement-phase pension assets), but the reform still applies. The 50% discount in super becomes a 1/3 discount (legacy); the reform replaces that with indexation + minimum tax in super too, with the minimum applied to the fund’s effective rate. Specialist advice recommended.

Don’t reorganise into a company purely to escape the reform. Companies pay 30% on gains and have no CGT discount — for high-return long-term holds the reform leaves individuals worse off than they were, but still typically better off than a company structure. Reorganisation also triggers its own CGT events on the contributed crypto.

Calculators and scenarios

• Crypto CGT Calculator — Bucket A/B/C apportionment, indexation, 30% minimum.
• Crypto Tax Calculator — FIFO/LIFO parcel matching, swaps, staking income, AUD conversion.
• Crypto DeFi & Staking Tax Calculator — staking-reward parcels with auto-population of acquisition dates.
• CGT Calculator — general CGT engine for non-crypto assets in the same portfolio.
• CGT Hold vs Sell scenario — model pre vs post 30 June 2027 disposal timing.

Sources

• Treasury Budget Paper No. 2, Tax Reform — Boosting Home Ownership measure (p.21), 12 May 2026 — confirms reform applies to “all CGT assets, including pre-1985 CGT assets, held by individuals, trusts and partnerships”.
• Treasury fact sheet: Negative Gearing and Capital Gains Tax Reform (12 May 2026).
• ATO Tax Determinations on crypto assets — TD series covering staking, airdrops, hard forks, lost keys, personal-use exemption. Check the ATO website for the current versions of TD 2014/25 (crypto as CGT asset), TD 2014/26 (crypto-to-crypto), and the recent staking determinations referenced in prose above.
• ATO Cryptocurrency Data-Matching Program — basis for the ATO’s existing visibility into Australian exchange activity.

Related reading

• 50% CGT Discount Reform: Cost Base Indexation + 30% Minimum Tax from 1 July 2027 — master article covering all asset classes.
• Budget 2026 Explained: Winners and Losers — full Budget breakdown.
• CEX vs On-Chain Crypto Trading: Tax Differences in Australia — compliance burden by venue.
• How Budget 2026 Changes Your EOFY Plan — combined EOFY 2026 and 2027 implications.