A founder holding a treasury in digital assets, a business accepting crypto payment, a protocol about to issue a token and a DAO with no legal wrapper all have different commercial problems. They tend to arrive with the same finance problem: nobody wrote down a position early, the transaction history is now enormous, and the first person to ask hard questions is an auditor, a bank or Inland Revenue.
Property, not currency
New Zealand does not treat cryptoassets as currency or as legal tender. They are treated as a form of property. That single classification drives nearly everything that follows, because property has a cost base, a disposal, and a gain or loss on disposal — whereas currency in your bank account simply sits there.
The consequence founders find hardest is that a taxable event does not require New Zealand dollars to move. If you swap one token for another on a decentralised exchange, you have disposed of the first asset. If you pay a supplier or a contractor in crypto, you have disposed of the asset used to pay them. If you bridge, wrap, or convert into a liquidity position, you may have disposed of something. Depending on the structure, pledging assets as collateral can also be a disposal rather than a loan. In each case the gain or loss is measured in New Zealand dollars using the market value at the time, even though no New Zealand dollars were involved on either side.
A crypto-to-crypto swap is a disposal. A trader who never once withdrew to fiat, and who feels no richer, can still have a substantial New Zealand dollar tax liability from thousands of swaps — payable in New Zealand dollars, from a wallet that holds none. The liability and the liquidity are in different assets. Plan for that before the year ends, not after.
| Event | Commonly assumed | Position to test |
|---|---|---|
| Sell token for NZD | Taxable | Taxable. The uncontroversial case |
| Swap token A for token B | Not taxable — no cash out | Disposal of token A at market value; gain or loss crystallised |
| Pay a supplier or contractor in crypto | Just settling an invoice | Disposal of the asset used, plus the normal deduction and, where relevant, withholding and GST consequences of the underlying payment |
| Accept crypto from a customer | Revenue when converted | Revenue at the point of sale, valued in NZD; the asset then sits on the balance sheet with its own cost base |
| Provide liquidity to a pool | Still holding the same assets | Depends on the mechanics — often an exchange of assets for a pool position, and the exit is a second event |
| Pledge as collateral | Borrowing, not selling | Depends entirely on whether beneficial ownership passes. Some structures are loans; some are disposals |
Layered on top is the question of intention. New Zealand taxes amounts derived from property acquired for the purpose of disposal, and from a profit-making undertaking or scheme, without needing a separate capital gains tax to do it. In practice, most people who acquire cryptoassets do so with disposal in mind — that is generally the point of holding them — and Inland Revenue has been direct about that expectation. The instinct that "this is a long-term investment, so it is capital" travels badly here. Purpose is assessed at acquisition, on the evidence, and the evidence is your own written record. If you genuinely hold for a different reason, document it contemporaneously, because reconstructing intention years later in front of a reviewer is a losing position.
Records and cost base
This is where digital asset businesses actually fail. The tax analysis is often tractable. The record keeping is not, and no amount of good analysis survives a ledger nobody can reconstruct.
An active treasury or protocol can generate transaction volumes across multiple wallets, several chains, centralised exchanges, decentralised exchanges and bridges. Each of those produces a different export in a different format, with different timestamps, different fee treatments and different naming for the same asset. Exchange CSVs are not a sufficient record: they show what happened on that venue only, they frequently omit the cost base of assets deposited from elsewhere, and they say nothing about on-chain activity. Chain explorers show the transaction but not why it happened or what it was worth in New Zealand dollars at the time.
What you need is a proper sub-ledger — a dedicated digital asset accounting tool that ingests wallet addresses and exchange APIs, values every movement in New Zealand dollars at the time it occurred, applies a consistent cost base method, and posts summarised journals into the general ledger. Choose the cost base method deliberately, document it, and apply it consistently; switching methods mid-history is both a tax problem and an audit problem. Treat gas and transaction fees explicitly rather than letting them disappear into rounding. Label internal transfers between your own wallets so they are not misread as disposals, because an unlabelled internal transfer is the single most common source of phantom gains.
Before anything else, produce a complete inventory of every wallet address, exchange account, custodian and smart contract the business controls or has controlled — including the ones a founder set up personally in the early days and now uses for company assets. You cannot reconcile a population you have not defined, and the wallets nobody remembered are always the ones with the awkward history.
Where GST lands
GST on digital assets is not one answer. It depends on what the token is and what is actually being supplied, and the rules have been amended specifically to deal with cryptoassets, so anything you read that predates those changes is unreliable.
Broadly, the supply of cryptoassets that function as a means of payment or investment has been carved out of the normal GST net, which prevents the absurd outcome of GST arising every time a token changes hands. That relief does not extend to everything with a token attached. A token that entitles the holder to goods or services is closer to a voucher or a prepayment than to a currency. Non-fungible tokens conveying rights to artwork, media, access or utility are frequently supplies of something, and the analysis follows the underlying right rather than the wrapper. Mining and staking rewards, transaction validation services, and businesses providing exchange, custody or development services have their own treatment — and where the customer is offshore, the zero-rating rules and the remote services rules both matter.
If you accept crypto as payment for a normal taxable supply, the supply is taxable in the ordinary way. The GST is calculated on the New Zealand dollar value of the consideration at the time of supply and is payable in New Zealand dollars. Movement in the token's value between the sale and the day you convert is your problem, not the Commissioner's.
Staking, mining, airdrops and yield
The recurring question is whether a receipt is income when it arrives, or whether nothing happens until you dispose of it. The general shape of the answer is that rewards received in return for an activity — validating, mining, staking, providing a service — look like income at the point of receipt, valued in New Zealand dollars, and the asset then carries that value as its cost base for a later disposal. Unsolicited receipts with no service behind them sit on less settled ground, and the treatment turns on the facts and on why you held the underlying asset in the first place.
| Receipt | Question to answer | Practical difficulty |
|---|---|---|
| Mining rewards | Is this a business or a profit-making activity? Almost always income on receipt | Valuing many small receipts, and deducting hardware and power correctly |
| Staking rewards | Income on receipt in most fact patterns | Rewards that accrue continuously but only become transferable later — when is receipt? |
| Airdrops | Was it received in connection with a business or an existing activity? Was it solicited? | Valuing a token with no market at the moment it lands in the wallet |
| Liquidity provision and yield | Return on an asset, or proceeds of an exchange of assets? | Pool mechanics differ; impermanent loss has no clean tax analogue |
| Governance token grants | Consideration for services, or a shareholding-like interest? | Often also an employment or contractor payment, with PAYE consequences |
The valuation-at-receipt problem deserves its own attention. Where a token has deep liquidity on major venues, valuing a receipt is administrative. Where it does not — a new project token, a governance token before listing, an airdrop into an illiquid market — you are being asked to put a New Zealand dollar figure on something with no reliable price. There is no clever answer. What defends the position is a written valuation policy adopted in advance: the price source hierarchy, the timestamp convention, what you do when there is no market, and who approves an exception. A consistent, documented, defensible method applied to every receipt is worth far more under scrutiny than a more precise method applied selectively.
The accounting position
Tax and financial reporting diverge here, and the reporting side surprises boards more.
Under IFRS, cryptoassets held by most businesses are generally not cash, not a cash equivalent, and not a financial asset — a typical token gives its holder no contractual right to receive cash or another financial asset from anyone. That leaves two homes. If you hold digital assets for sale in the ordinary course of business, they may be inventory, and a broker-trader may be able to measure at fair value less costs to sell. Everything else generally falls to intangible assets.
Intangible is where the reporting problem lives. Under the cost model, an intangible asset is carried at cost less accumulated impairment. When the market falls, you impair. When the market recovers, an impairment loss on this kind of asset is not reversed. A treasury that has ridden a cycle down and back up can therefore be carried at a written-down figure that everyone in the room knows is far below what the assets are worth, with the loss in the profit and loss account and no corresponding recovery. That is not a rounding issue for a company whose balance sheet is mostly digital assets — it distorts equity, it distorts covenant calculations, and it makes the statutory accounts a poor description of the business.
Impairment under the cost model runs one way. If a material part of your balance sheet is digital assets, the reported equity in your statutory accounts can drift a long way from economic reality, and that number is what a lender, an acquirer or a covenant test will use. Decide how you will explain the gap — in the notes, in board reporting and in a raise — before the first reporting period in which it matters. Confirm the applicable standard and any revaluation option available under your reporting framework with your auditor; the treatment of digital assets is an area of active development internationally.
Issuing a token
The question at issuance is simple to ask and difficult to answer: when the money comes in, is it revenue, a liability or equity? The accounting answer is not determined by what you call the raise. It is determined by what the token obliges you to do.
If the token entitles holders to future goods, services or platform access that you must deliver, the proceeds look like a customer prepayment — a liability, released to revenue as you perform. If the token conveys a residual interest in the entity, or a right that behaves like ownership, it moves towards equity. If it conveys a right to receive cash or a return, it starts to look like a financial liability, and it may be a financial product for regulatory purposes as well, which is a materially larger conversation involving the Financial Markets Authority and licensed advice. If it conveys nothing at all, that is a position too, and it needs to be stated plainly rather than left ambiguous.
Which is why the whitepaper is an accounting document. Whatever it promises — utility, access, revenue share, governance, a redemption mechanism, a buyback — is the substance that determines the accounting and much of the tax and regulatory analysis. Marketing language written to excite a community becomes, later, evidence of an obligation. Have finance and legal read the whitepaper before it is published, not after the token is in circulation and the promises cannot be unmade.
DAOs, entities and cross-border
A DAO raises a question that has to be answered before anything else can be: who is the taxpayer. Software does not file returns. If there is no legal entity, the obligations and the liability do not disappear — they attach to people, and in the least attractive analysis a group of participants acting together in a profit-making enterprise without a legal wrapper can be exposed on a general partnership analysis, where liability is joint and personal. Practical consequences follow immediately. Who signs the accounts. Who holds the treasury keys and in what capacity. Who is liable if a contract fails or a counterparty sues. Who is a director for the purposes of the duties that attach to that role. Most serious projects end up wrapping the DAO in a recognisable legal entity, precisely so these questions have answers.
The cross-border layer compounds it. Token entities are routinely established in foreign jurisdictions on advice that focuses on the token and not on the group. If the people making the decisions are in New Zealand, several things follow. Central management and control can make a foreign-incorporated company a New Zealand tax resident. A New Zealand-resident team operating for a foreign entity can create a permanent establishment. New Zealand's controlled foreign company rules can attribute income back to New Zealand shareholders. And if the intellectual property was developed in New Zealand and now sits offshore, transfer pricing applies to how it got there and to what is charged for it — a question both revenue authorities are equipped to ask.
None of this makes an offshore structure wrong. It makes an offshore structure a group tax decision that needs New Zealand advice as well as local advice, taken before incorporation. Structures assembled to satisfy a foundation lawyer in one jurisdiction, without anyone modelling the New Zealand consequences, are expensive to unwind once a token is live and holders are distributed.
Auditors, banks and the sequence
Two audiences will test all of the above, and neither is interested in the technology.
An auditor will want evidence of existence, ownership and completeness. Existence and ownership generally means proving control of the keys — a signed message from each wallet, or independent confirmation from a custodian — not a screenshot of a balance. Completeness means demonstrating that the wallet population is complete, which is why the inventory matters. Beyond that: your valuation policy and its consistent application, the sub-ledger reconciled to the chain and to the general ledger, the classification decision and its rationale, the impairment assessment, related-party and insider wallet transactions, and your key management and segregation of duties controls. Nearly all of that has to be built as you go. It cannot be assembled retrospectively from exchange exports.
A bank is running a different test. Crypto businesses get debanked because they present anti-money-laundering and countering financing of terrorism risk that the bank cannot cheaply assess, not usually because anyone thinks the business is dishonest. The bank needs to know the source of funds, the identity of counterparties, what compliance obligations you carry in your own right, and whether your controls would survive its regulator's questions. The businesses that keep banking relationships are the ones that treat this as a standing relationship to manage: proactive disclosure, documented AML/CFT policies, a named compliance officer, blockchain analytics on incoming funds, and a second banking relationship established while the first one is healthy. Do not wait until an account is being closed to start assembling that file.
| By when | What has to exist | Why then |
|---|---|---|
| Before the first material transaction | Wallet inventory, sub-ledger tool selected, cost base method and valuation policy written down | History is cheap to capture live and very expensive to reconstruct |
| Before the entity structure is fixed | New Zealand tax advice on residence, PE, CFC and transfer pricing — alongside any offshore advice | Structure is the hardest thing to change once a token is in circulation |
| Before the whitepaper is published | Finance and legal review of every promise the token makes | The promises determine revenue, liability or equity, and the regulatory perimeter |
| Before the first token event | Accounting position papers on classification, issuance and measurement, discussed with the auditor | Agreeing treatment in advance costs a fraction of arguing it in an audit |
| Before the first audit | Key control evidence, reconciled sub-ledger to GL, impairment assessment, AML/CFT documentation | These are evidence obligations that only work if built contemporaneously |
| Before a raise | Clean cap table including token holders, tax positions documented, banking stable, statutory equity gap explained | Diligence tests the weakest of these, and unresolved tax exposure discounts valuation |
The pattern across all of it is the same. The technical questions in this area are genuinely unsettled, and reasonable advisers disagree. What is not unsettled is the standard of evidence. A defensible position taken early, written down, applied consistently and supported by a reconcilable ledger will hold up. The same position reconstructed two years later from partial exports will not, however correct it happens to be.
XLCFOWe build the finance function underneath digital asset businesses — sub-ledger implementation, accounting position papers, treasury reporting and audit readiness — working alongside specialist tax counsel where a licensed opinion is required.
