The deadline for 2026 capital losses is December 31. Not December 30, not January 2 on a carry-back — December 31. Losses realized after the bell close on New Year’s Eve count in 2027. Everything before it counts in 2026. The only effective planning window is October through late November, which means right now.
For high-net-worth filers sitting on a mix of unrealized gains and unrealized losses, the mechanics of this year make a systematic review worth doing before Q4 ends.
The § 1211 Math — Gains vs. Ordinary Income
Capital losses offset capital gains dollar-for-dollar, with no ceiling. Long-term losses apply first against long-term gains; short-term losses apply against short-term gains, with any excess crossing categories. The result: a taxpayer at the top bracket who harvests $400,000 in long-term losses to offset $400,000 in long-term gains eliminates $95,200 in federal tax (23.8% combined rate) and additional California tax if applicable.
The constraint kicks in when losses exceed gains. Excess losses offset ordinary income only up to $3,000 per year. Everything above $3,000 carries forward indefinitely — it doesn’t disappear — but harvesting $500,000 in losses against $100,000 in gains does not produce a $95,200 tax deduction this year. It produces a $23,800 federal deduction plus a $400,000 carryforward that is worth something, but not all at once.
The practical implication: match the harvest to the gains. Know what gains have been realized in 2026 before identifying the loss candidates. Harvesting in a vacuum produces carryforwards; harvesting against a known gain target produces immediate tax savings.
The Wash-Sale Trap — and the 30-Day Window
§ 1091 disallows a harvested loss when the same or substantially identical security is purchased within 30 days before or after the sale. The disallowed loss is not permanently gone — it is added to the new position’s basis under § 1091(d) — but the timing benefit is eliminated.
The standard workaround is to maintain economic exposure through a different instrument. Selling a position and replacing it with a comparable but non-identical security avoids the wash-sale problem while keeping market exposure. The 30-day clock runs in both directions, so the replacement purchase needs to come after a 31-day wait if you want to rebuy the original position.
The December math: if you sell on November 24, the 30-day window runs through December 24. Selling later than that risks running out of room to rebuy before year-end if you want to close out the replacement position in 2026 as well. For most investors, October through mid-November is the practical window to harvest and manage the wash-sale clock comfortably.
Crypto Is Still Outside § 1091 — For Now
The OBBBA (signed July 4, 2025) did not extend the wash-sale rule to digital assets. As of today, § 1091 applies only to securities — stocks, bonds, and options on securities. Cryptocurrency, tokens, and NFTs are not covered.
That means a Bitcoin holder can sell at a loss on December 30 and rebuy on December 31 without triggering a wash sale. The loss is fully recognized; the new position carries a stepped-up basis. For taxpayers with significant crypto exposure and unrealized losses, this is one of the more straightforward year-end moves available in 2026.
The window will likely close at some point — the IRS has been signaling expansion of § 1091 to digital assets, and legislation requiring it has circulated in both chambers. For 2026, the door is still open. In 2027, it may not be.
What Direct Indexing Actually Delivers
Direct indexing — owning individual securities in a custom account rather than a fund — is marketed as a tax-alpha strategy. The claim is that it generates more harvesting opportunities by allowing loss realization on individual constituents of an index even when the index itself is flat or up.
The claim is accurate as far as it goes. An S&P 500 index fund in a flat year may show no harvestable loss at the fund level, even if 60% of the underlying companies are down. A direct-indexed account holding those individual positions can harvest the losers while holding the winners, capturing paper losses that the fund structure obscures.
The practical caveat: the tax alpha is highest in the first few years of the account, when the entire portfolio has been purchased recently at higher basis. After several years of harvesting, most positions carry low basis and the pool of harvestable losses shrinks. Direct indexing fees — typically 25–40 basis points above a plain index fund — erode the advantage over time. It is not a magic box; it is a timing tool that front-loads tax benefits.
For investors with new concentrated cash to deploy into diversified equities, direct indexing earns its cost. For investors with existing low-basis positions being converted, the transition itself may trigger gains that offset years of projected harvesting alpha.
Concentrated Positions — Where Harvesting Interacts With Concentration Risk
Clients with single-stock concentration face a different calculus. The position may be the largest asset in the portfolio and the one most in need of reduction — but it may also carry the lowest basis and the highest embedded tax cost to unwind.
Harvested losses elsewhere in the portfolio reduce the net tax cost of a partial sale of the concentrated position. A $300,000 concentrated stock gain paired with $250,000 in harvested losses from other positions reduces the net federal tax on the transaction from $71,400 to $11,900. The harvest does not eliminate the gain; it reduces the friction on a rebalancing decision that may be warranted on its own merits.
The sequencing matters. Identify the concentrated position to reduce first, then size the harvest to offset the target gain. Working backward from the tax target produces a cleaner result than harvesting opportunistically and then figuring out what to do with the losses.
The Checklist — What to Do Before December 31
Now through November 25: - Identify all realized gains and estimated 2026 income to establish the harvest target - Scan the portfolio for unrealized loss positions, net of replacement costs - Identify concentrated positions where a partial sale makes sense; size the harvest to match - Review crypto holdings for harvestable losses under the current § 1091 carve-out - Evaluate whether direct-indexed accounts are generating expected tax alpha for their fee cost
November 26 through December 15: - Execute harvesting trades with replacement positions in place - Confirm wash-sale compliance — 31-day window must clear before year-end if rebuy is intended in 2026
After December 15: - Final review of unrealized gains that should be deferred to 2027 if realizing them this year is avoidable - Confirm estimated tax payments are adequate given any gains realized
We don’t rush October filings — but we don’t miss a December deadline either. If there is a concentrated position or a significant gains event in your 2026 picture, this is the conversation to have before Thanksgiving. Let’s talk.
Daniel R. Litvin, CPA, MBA, is Managing Partner at Laléa & Black, LLP in Beverly Hills, Calif. The firm focuses on tax planning and business management for founders, family offices, and closely held businesses. Nothing in this article constitutes investment advice; consult your investment advisor regarding portfolio decisions.