Token Vesting: Accounting Treatment for Grants & Cliffs

Token grants are everywhere in Web3 compensation. The accounting framework that applies (IFRS 2, ASC 718) is borrowed from equity-based compensation but doesn't fit cleanly. Here's where it bends and where it breaks.

Why tokens are usually treated as share-based payment (and when they aren't)

When a protocol grants its native tokens to contributors in exchange for services, the economic substance looks a lot like granting equity. The recipient performs work, the issuer settles the obligation in its own instrument, and the value of that instrument is tied to the issuer's success.

Both IFRS 2 and ASC 718 are written around equity instruments, but the substance-over-form principle pushes most preparers toward share-based payment treatment for tokens that confer governance rights, economic rights, or some claim on the protocol's value. This is the prevailing position from Big Four guidance papers issued between 2022 and 2025.

Where it stops fitting:

• Pure utility tokens with no governance or economic rights. These can look more like prepaid services or a separate revenue arrangement.
• Stablecoin-denominated grants paid out of treasury. These are simply cash-equivalent compensation expense, no IFRS 2 / ASC 718 mechanics.
• Tokens granted by a third party (e.g., a foundation grants tokens to contributors of a separate operating company). The accounting at the operating company depends on whether the foundation is a related party and whether reimbursement exists.

If the token gives the holder a residual claim on the network's value, treat it as share-based payment until your auditor tells you otherwise.

IFRS 2 vs ASC 718: the framework

The two standards agree on more than they disagree, but the differences matter at year-end.

Equity-settled awards (the protocol delivers its own tokens) are measured once at grant date and not remeasured. Cash-settled awards (the protocol promises cash equal to a token's value) are remeasured every reporting period until settlement, with changes through P&L. Most pure token grants are equity-settled in substance. Phantom token plans are cash-settled.

Vesting schedule fundamentals: cliff, linear, milestone, performance

Four patterns cover most of what's deployed in practice.

Cliff vesting. Nothing vests until a date, then a chunk vests at once. The classic "1-year cliff" inside a longer schedule.

Linear (graded) vesting. Equal portions vest over time (monthly, quarterly, or per block). Typically combined with a cliff.

Milestone vesting. Tokens vest when a defined event occurs: TGE, mainnet launch, KPI hit, audit completion. These are non-market performance conditions and they affect the number of tokens that vest, not the fair value per token.

Market-based performance. Tokens vest only if the price hits a target. Under both IFRS 2 and ASC 718, market conditions are baked into fair value at grant date and expense is recognised regardless of whether the target is hit. Counter-intuitive, but it is the rule.

The vesting condition type matters for trueup. Service and non-market performance conditions allow you to revise the expected number of vesting tokens; market conditions do not.

Grant date vs vest date vs settlement date

Three dates, three different accounting moments.

The trap is conflating vest with settlement. A contributor whose tokens "vest" on month 13 but only become transferable at TGE 18 months later has a vesting date in month 13 (that is when the service condition is satisfied) and a separate settlement date later. Cumulative expense should equal grant-date fair value of vested tokens at the vest date, not the settlement date.

Fair value at grant date: how to actually measure it

This is where token grants depart from listed-equity guidance.

Liquid token, traded on a reputable venue at grant date. Use the spot price on grant date. Apply a discount for lack of marketability if there's a transfer restriction beyond vesting (the DLOM is judgmental, often 10 to 30%).

Pre-launch token, no observable price. This is the hard case. Common approaches:

1. Last priced round. If a SAFT or token sale closed at $X recently, that's a starting point. Adjust for time, dilution, and market conditions.
2. Implied valuation backsolve. Use the protocol's equity valuation, the planned token supply, and the equity-token split to back into a per-token value.
3. Discounted token model. Project network revenue, apply a multiple, divide by circulating supply. Heavily judgmental.

Document the methodology. Auditors will challenge the input assumptions, not the technique itself. Whatever you pick, lock it down at grant date and don't revisit unless terms change.

Recognition over the vesting period: P&L impact

Once you have grant-date fair value and a vesting schedule, the recognition pattern follows.

For a 4-year linear schedule with a 1-year cliff (the most common pattern), under IFRS 2:

• Each annual tranche is treated as a separate award (graded vesting attribution).
• The 1st tranche vests over 1 year, the 2nd over 2, the 3rd over 3, the 4th over 4.
• Front-loaded expense profile: roughly 52% of total expense in year 1, 27% in year 2, 14% in year 3, 7% in year 4.

Under ASC 718, you can elect straight-line attribution: 25% per year. Most US issuers do.

Sample journal entry for a quarterly tranche, equity-settled token grant:

Dr  Compensation expense       100,000
    Cr  Equity — token reserve         100,000
(Recognise vested portion of token grant for Q2 2026,
4-year linear schedule with 1-year cliff, contributor X)

When tokens are eventually delivered on-chain at settlement, the equity reserve is derecognised against the token treasury.

Modifications, cancellations, and forfeitures

Grants get changed. Common modifications: extending vesting, accelerating on a liquidity event, repricing after a token crash.

Modification. You measure the incremental fair value (new fair value minus original fair value at modification date). If positive, expense it over the remaining vesting period. If negative, you generally cannot reduce expense; original fair value is a floor.

Cancellation with replacement. Treated as a modification. The replacement award's fair value is compared to the cancelled award's fair value at cancellation date.

Cancellation without replacement. Accelerate the remaining unrecognised expense to the cancellation date. This is counter-intuitive: cancelling a grant doesn't save you P&L, it pulls expense forward.

Forfeitures (service condition not met). Reverse cumulative expense for unvested tranches. Under IFRS, you must estimate forfeitures up front and true up. Under ASC 718, you can choose to recognise forfeitures as they occur.

The recipient's tax view: when does income hit?

This is jurisdiction-specific and the area where contributors get surprised most often. The general patterns:

• US (employee). Income recognised at vest based on FMV at vest. If subject to a substantial risk of forfeiture and an 83(b) election is filed within 30 days of grant, income is taken at grant on grant-date FMV, which can be a huge tax saving for pre-launch tokens.
• UK. Generally taxed at vest as employment income at FMV. Restricted Securities Election (s431) can move tax to grant.
• France. Complex. Depends on whether the grant qualifies under the BSPCE / AGA regimes. If not, taxed as wages at vest.
• Germany / Switzerland. Typically taxed at vest at market value, but Switzerland has favourable rulings on illiquid tokens at grant.
• Singapore / UAE. Light or no tax on vest for many structures.

The mismatch between accounting expense (issuer side) and taxable income (recipient side) is normal and expected. They are not the same number and they don't hit at the same time.

Practical patterns: 4-year linear with 1-year cliff, milestone unlocks

The two patterns covering 80%+ of Web3 grants:

4-year linear with 1-year cliff.

• Year 1: nothing vests until month 12, then 25% vests at the cliff.
• Months 13 to 48: monthly linear vesting of the remaining 75%.
• Accounting: graded attribution under IFRS 2 (front-loaded), straight-line option under ASC 718.

TGE + 12-month cliff + 24-month linear.

• 0% pre-TGE.
• 25% at the cliff (12 months post-TGE).
• 75% over 24 months thereafter.
• TGE is typically a non-market performance condition. Expense recognition starts at grant, but the number of vesting tokens depends on TGE actually happening; true up if probability changes.

Document the schedule precisely. "4-year vesting" without a cliff specification is a future audit problem.

Common reporting mistakes

Patterns we see repeatedly when reviewing Web3 books:

• Recognising expense at settlement, not vest. A contributor whose tokens unlock at TGE 18 months after vesting should not generate expense in month 18.
• Using current market price for measurement. Grant-date fair value is fixed. Subsequent price movements don't change equity-settled expense.
• Ignoring the cliff in the attribution. Under graded vesting, the cliff doesn't delay recognition; expense accrues from grant date even though no tokens have vested yet.
• Treating phantom tokens as equity-settled. If the protocol promises cash equal to token value, that's cash-settled. Liability remeasurement applies.
• Forgetting employer payroll tax. In many jurisdictions, the employer owes payroll tax on the FMV at vest, regardless of whether the contributor is paid in tokens or fiat.

FAQ

Q: Does IFRS 2 actually apply to tokens, or are we stretching the standard? The IASB hasn't issued token-specific guidance. Substance-over-form analysis under IFRS 2 is the prevailing position from Big Four guidance, but it remains an interpretation, not a rule.

Q: We're pre-token. How do we book grants if there's no fair value yet? You still book at grant date based on the best available estimate. Common inputs: last SAFT round price, implied per-token value from equity raise, or a documented internal valuation. Don't wait until TGE to start recognising expense.

Q: What happens if the token price crashes 90% before vest? For equity-settled grants, nothing on the issuer's books; fair value was locked at grant. For the recipient, taxable income at vest is based on FMV at vest, so the tax bill drops with the price. Cash-settled grants are remeasured, so the issuer's liability also drops.

Q: Can we true up if the team votes to cancel a grant? Cancellation without replacement accelerates the remaining unrecognised expense. You don't get to reverse what's already been recognised.

Q: Is an 83(b) election always a good idea? For pre-launch tokens with negligible FMV, often yes. For tokens with meaningful FMV at grant, the upfront tax bill can be significant and non-recoverable if the tokens never vest. Run the numbers with a tax adviser before electing.

Further reading

• Wag3s HR — payroll and contributor compensation tooling for Web3 teams.
• DAO Contributor Compensation — practical patterns for paying contributors.
• IFRS vs GAAP for Crypto Assets — broader framework comparison for digital assets.
• IFRS 2 Share-based Payment (IFRS Foundation).
• ASC 718 Compensation — Stock Compensation (FASB).