Two of the most routine things a DeFi user does - dropping a token pair into an AMM pool and wrapping ETH into WETH - sit inside an open question.

As of September 26, 2026, the practitioner sources we checked report no IRS guidance that directly resolves whether an AMM liquidity-pool deposit or a one-for-one ETH-to-WETH wrap is a taxable disposition.

That is an absence of specific guidance on those two transactions. It is not a ruling that either one is taxable, and not a ruling that either one is free of tax.

What the IRS property rule settles - and what it leaves open

The part that is settled is the starting point, not the answer.

Notice 2014-21, Q&A 1 and 6 (current figure at the IRS), treats convertible virtual currency as property and says that exchanging virtual currency for other property produces taxable gain when the fair market value received exceeds adjusted basis, or a loss when it falls below basis.

What the notice does not do is classify particular transactions.

It gives no transaction-specific treatment for a liquidity-pool deposit or a token wrap, which means the disputed step is not the rule - it is whether your deposit or wrap counts as that exchange in the first place.

That gap is why the rest of this page reads like a comparison instead of an answer.

The scoped finding about missing guidance comes from CountDeFi, a commercial practitioner blog, and practitioner commentary sits below official material in the hierarchy: IRS documents govern, commercial guides and academic papers argue about how to apply them, and neither argument binds you or the examiner.

Liquidity-pool deposits: compare the two positions

Pool deposits and wrapping: the competing positions

Activity Exchange position and reasoning Continuity position and reasoning Named source and source quality Evidence needed for entry and exit
AMM liquidity-pool deposit and LP redemption The deposit exchanges the paired tokens for a different asset, the LP token, at fair market value; redemption is a further exchange back into pool assets Tax ownership of the deposited assets continues, so the LP token evidences the same position and nothing is recognized at deposit CoinTracking DeFi guide (commercial practitioner analysis); SSRN abstract identifier 4942799 (academic analysis, not an IRS authority) Per-leg amounts and USD values at deposit and at redemption, LP token quantity received and burned, fee and reward distributions logged separately
One-for-one ETH to WETH wrap and unwrap WETH is property distinct from ETH, so the wrap disposes of the ETH at fair market value One-for-one redemption preserves the same economic rights, so basis and holding period carry over and nothing is recognized CoinTracking US wrapping guide (commercial); CountDeFi (commercial practitioner analysis) Wrap and unwrap hashes with matching quantities, the ETH price used if the exchange position is taken, the same label applied in both directions

Notice 2014-21 supplies the shared property principle behind both columns. It supplies no verdict for either one.

The exchange argument is mechanical: you handed two tokens to a contract, you got back a pool token with its own address and its own price behavior, and property went out while different property came in.

Under that reading, entry is a disposal of the deposited assets measured against their basis, and redemption is a second disposal of the LP token.

The continuity argument, which CountDeFi describes as a nonrealization theory and the SSRN paper develops as academic analysis, starts somewhere else.

If the receipt token only evidences that you still beneficially own what you put in, you have not received materially different property, so there is nothing to recognize at entry and the basis simply rides through to exit.

The shared research labels the exchange reading the conservative or majority position, and that label describes how the sources present it rather than a counted survey of preparers. Neither column is an IRS holding, and neither is risk-free.

Three things get tangled in DeFi bookkeeping and they belong in separate entries. Receiving the LP token is the characterization question above.

Fee and reward-token distributions are a different item: the practitioner research treats them as ordinary income at fair market value when received or accrued, whichever the facts support, independent of how you labeled the deposit.

Impermanent loss is neither - it is an economic effect that shows up in the quantities you get back on exit, not a separate deductible loss you claim while the position is open.

ETH to WETH: why a one-for-one wrap still raises a question

Nothing about your economic exposure changes when ETH becomes WETH, and that is exactly the split.

The CoinTracking wrapping guide records the conservative commercial position: WETH is a different token, a different contract, distinct property, so the wrap is a disposition of ETH at fair market value.

The competing continuity position holds that one-for-one redeemability preserves the same rights, with basis and holding period carrying over to the wrapped units and no gain measured at the wrap.

What the research flags hardest is asymmetry. Picking the exchange reading on the way in and the continuity reading on the way out is not a position, it is two positions chosen for their results, and the file shows it.

Keep the scope narrow too: this comparison covers one-for-one ETH and WETH, and the reasoning that supports it does not automatically travel to every bridge token, receipt token, or derivative wrapper.

Myth: "I can redeem it one-for-one, so it must be tax-free." Redeemability is the core of the continuity argument, not proof that the argument wins. It remains an open question for these transactions, which is why the reasoning you relied on has to live in the file next to the transaction.

Myth: "A different token address settles the tax result." A new contract address is the factual hook for the exchange position, and it does not close the question either. Both readings look at the same address and disagree about whether the property received is materially different.

Myth: "If wrapping is treated one way, unwrapping can use whichever result is cheaper." This one is not open. Whichever reading you adopt, apply it to the wrap and the unwrap, and keep the same basis and holding-period consequences on both sides.

Lending yield: separate the deposit from the income

reedcorp.tax describes the common practitioner handling of a lending deposit: the deposit continues your beneficial ownership, interest received is ordinary income at fair market value, and the return of your original principal is not itself taxable.

The receipt token is where that tidiness stops.

Whether swapping your asset for a lending receipt token is an exchange is expressly unresolved in the same material, and CountDeFi separately identifies a second question - what happens when yield accrues inside the receipt token and no reward transaction ever hits your wallet.

The evidence supports neither a universal "income accrues" answer nor a universal "income only on withdrawal" answer, so that is a question for your preparer with the contract behavior in front of them.

One mechanic underlies all of it.

The value you include in income when a reward or interest payment is received becomes your cost basis in those units, and a later sale or exchange produces a separate capital gain or loss measured against that basis.

The same batch can carry income tax at receipt and a distinct gain or loss at disposal, and the two can point in opposite directions: income recorded at a high value, then a loss when you sell.

Liquid staking timing has its own page, including what the staking ruling does and does not cover.

Why a missing 1099-DA does not answer the tax question

Notice 2024-57 excuses brokers from filing Form 1099-DA, pending further guidance, for a listed set of transactions: wrapping and unwrapping, liquidity-provider transactions, staking, lending, short sales, and notional principal contracts.

Two things follow, and only two. Those exceptions are about the broker's filing duty, and they are temporary by their own terms. The taxpayer's substantive treatment and reporting duty sit outside them entirely.

The 2026 Instructions for Form 1099-DA also exclude rewards and staking payments from that form. So a pool position can generate reward income all year and produce no matching form anywhere. An empty mailbox is a gap in third-party reporting, not a classification.

For how the form itself works, the cluster hub is the place to return to the digital-asset tax guides.

Build a record for the whole position

Most people asking whether a wrap is taxable are really asking how anyone is supposed to track this.

The workload is real: one pool position can generate a deposit, a receipt token, a dozen fee claims, a partial withdrawal, and a final exit, with gas on every leg, and an import that merges them into a single "trade" has thrown away the structure you need.

When a tool's output and the chain disagree, the chain is the record and the tool is an interpretation.

CountDeFi's recordkeeping guidance draws the line that matters: what the explorer shows is transfers, and the tax position is the interpretation you build on top of them. Keep both, separately.

Source-record checklist (CoinReckon's synthesis of the CountDeFi recommendations retrieved September 26, 2026, as a record template rather than a statutory form):

  1. Preserve the originals before any tool touches them: wallet exports, protocol dashboards, explorer pages for each deposit, claim, and withdrawal.
  2. Capture per leg: hash ______, chain ______, timestamp ______, signing wallet ______, contract ______, tokens and amounts in ______ / out ______, gas amount and token ______, USD value ______, valuation source ______.
  3. Connect the legs into one position: deposit → receipt or LP token → reward and fee receipts → withdrawal, each as its own labeled entry.
  4. Annotate the treatment you selected and the dated published reasoning you relied on.
  5. Carry each recognized reward amount forward as the basis of those exact units.
  6. Compare ending token quantities wallet by wallet against your book.
  7. Retain unmatched items and interpretation questions in an exception log.

Step 6 is the cheap test that catches the expensive errors.

If your book says a wallet ends the year with 4.1 of a reward token and the chain says 3.6, something is missing, duplicated, or valued twice, and you find it in minutes instead of reconstructing the year in April.

Keep two columns in the exception log, because missing data and unresolved treatment are different problems - "reconciled" means the quantities agree, not that the law is settled.

And keep protocol movements out of the gift bucket: if any of this leaves your control as an actual transfer to another person, check the rules for transferring crypto as a gift instead of labeling it as protocol activity.

Red flags to resolve before filing

Incompatible entry and exit labels. A deposit treated as an exchange and a redemption treated as a continuation, or a wrap and unwrap handled by different logic. Bring both legs side by side with the basis you used on each.

Silence or a refund read as authority.

When an agency concedes and pays a refund before a court rules and the case is dismissed as moot on that basis, the dismissal creates no binding precedent on the underlying question, and a separate agency ruling issued independently is not overridden by it.

The Sixth Circuit's Jarrett opinion of August 18, 2023 supports that narrow procedural point and nothing broader. Bring the published reasoning you actually relied on, not a news summary.

Untraceable values. A USD figure with no price source and no timestamp cannot be defended later. Bring the valuation source per leg.

Unmatched legs. A withdrawal with no matching deposit, or reward income with no receiving transaction. Bring the wallet-by-wallet quantity comparison that surfaced it.

A software label accepted as the analysis. An import that stamped a position "swap" made a guess about a contested question. Bring the underlying transactions and your own annotation.

Consistency is what documents your reasoning. It does not guarantee that the IRS agrees with it, and it does not become an argument that one reading is settled. If new contrary authority appears, or a leg simply cannot be reconstructed, that is the moment to hand the file to someone licensed to sign it rather than pick the label that lowers the number.

Bring the complete deposit-to-withdrawal record and the dated position you used to a qualified tax preparer before filing an unresolved DeFi position.

Sources

  1. IRS Notice 2014-21, Q&A 1 and 6, property and exchange principles. Published 2014; as of September 26, 2026. Official.
  2. CountDeFi, "How Is DeFi Taxed In The US? Pools, Lending And Staking", pool positions, wrapping, and the recordkeeping sections "The Records a DeFi Position Actually Needs" and "The test your book has to pass." Retrieved September 26, 2026. Commercial practitioner analysis; source of the scoped absence finding, the continuity theory, and the checklist recommendations.
  3. CoinTracking, DeFi tax guide, liquidity-pool treatment. Research snapshot September 26, 2026; publication date not supplied. Commercial practitioner analysis.
  4. SSRN paper, abstract identifier 4942799, ownership argument for pool deposits. Research snapshot September 26, 2026; title and publication date not supplied. Academic analysis, not an official holding.
  5. CoinTracking, US wrapping-crypto guide, exchange position. Research snapshot September 26, 2026; publication date not supplied. Commercial practitioner analysis.
  6. reedcorp.tax, crypto tax basics 2026, lending and receipt-token discussion. 2026 material in the September 26, 2026 snapshot. Commercial commentary.
  7. IRS Digital Assets hub, Notice 2024-57 broker-filing exceptions. Notice dated 2024; snapshot September 26, 2026. Official scope reference, not a ruling on the disputed transactions.
  8. IRS Instructions for Form 1099-DA, exclusion of rewards and staking payments. Tax year 2026; snapshot September 26, 2026. Official.
  9. Sixth Circuit, Jarrett v. United States, No. 22-6023, procedural mootness. Opinion dated August 18, 2023; snapshot September 26, 2026. Court opinion reproduced by Justia; supports the narrow procedural point only.