The Zero-Percent Harvest — Pay $0 in Capital Gains Tax

Most investors pay capital gains tax. A subset pay zero — legally — by harvesting gains in the 0% long-term capital gains bracket. In 2026 a married couple can realize up to ~$96,700 in long-term gains with no federal tax owed, every year that their ordinary income is low enough.

The 0% long-term capital gains bracket is one of the most generous provisions in the U.S. tax code and one of the least exploited. It exists because the LTCG schedule is bolted on top of the ordinary income schedule rather than running parallel to it. If your ordinary income is low enough in a given year, your first dollars of long-term capital gain stack onto the bracket below the 15% LTCG threshold — and that bracket taxes at 0%. Sell gains, pay no tax, repurchase the same shares the same day, walk away with a stepped-up cost basis.

How the bracket actually works

The mistake most people make is thinking of LTCG as a separate tax schedule. It isn't. LTCG is taxed at 0%, 15%, or 20% depending on your total taxable income — ordinary income plus the gains you realize. The 0% bracket isn't "$0 of ordinary income plus $96,700 of gains"; it's "total taxable income up to $96,700 for MFJ in 2026." If you earn $40,000 of ordinary wages, your headroom for 0%-bracket gains is roughly $96,700 minus $40,000 minus the standard deduction, which is the figure the sandbox to the right is computing live.

harvest_cap = zero_pct_top − (ordinary_income − standard_deduction)
harvest_cap = max(0, harvest_cap)

The mechanic: sell, buy, step up

Capital gains harvesting works because the wash-sale rule applies only to losses. You can sell a gain and repurchase the same security on the same day with no tax penalty and no waiting period. The brokerage records the sale at the current price, the IRS taxes the gain at your applicable LTCG rate (zero, in the 0% bracket), and your new cost basis is the repurchase price. Future appreciation starts from the higher basis.

The compounding effect is meaningful. A position with $100,000 of embedded gain, harvested in the 0% bracket and stepped up, produces $0 of current tax and saves 15% on the gain when you eventually sell at a higher bracket — $15,000 of permanent tax savings on that one harvest. Repeat for ten years and you've extracted $150,000 of tax-free basis step-up that would otherwise have been owed at retirement.

When CGH works best

Three windows are ideal: sabbaticals (a year off work, low or zero ordinary income); the early-retirement bridge between FI age and Social Security claiming; and geographic-arbitrage years like a "nomad year" with residence in a low- or no-tax jurisdiction. In all three cases the ordinary-income side of the bracket is small and the room for 0% gains is correspondingly large.

When it fails

Three traps in particular. Earned income too high. A six-figure W-2 alone puts most filers above the 0% bracket ceiling; there's no harvest room. State tax. California, Oregon, Minnesota, and a few others tax LTCG as ordinary income at state rates of 6–13%, partially clawing back the federal benefit. The strategy is much more powerful in no-tax states (FL, TX, WA, NV, TN, SD, WY). ACA subsidy cliffs. If your insurance comes from the marketplace, harvesting a gain that pushes your MAGI past 400% of poverty level can vaporize thousands in premium tax credits — easily more than the LTCG savings.

The Roth-conversion coordination problem

This is where most one-strategy-at-a-time advice falls apart. Roth conversions and capital gains harvesting compete for the same low-income years. Each dollar of conversion fills the bracket that each dollar of harvested gain would have filled at 0%. They aren't additive — they share a single budget.

The optimum is rarely "all conversions" or "all harvesting." More often it's a yearly mix that fills the 0% bracket with gains first (since they're tax-free outright) and uses the next bracket up (12%) for conversions. The exact split depends on your traditional-IRA balance, your unrealized gain, your retirement spending, and your Social Security claiming year. The Roth Conversion Playbook covers the conversion side; running them together is what the full simulator does.

Step-up at death: the silent edge case

If you plan to leave appreciated assets to heirs, harvesting may backfire. Inherited assets receive a step-up in basis at the date of death, eliminating all unrealized gain tax-free. Aggressive late-life harvesting pre-pays tax that would have been forgiven anyway. The rule of thumb: harvest aggressively in your early retirement years, slow down or stop once you're confident the assets won't be sold during your lifetime.

What to do today

1. Pull your cost basis. Most brokerages display cost basis per lot in the unrealized gains section. Know the number before tax season.

2. Model a low-income year. Plug a realistic ordinary-income figure into the sandbox to the right and see how much room you'd have. For most pre-retirees, the answer is zero. For most early retirees, it's substantial.

3. Coordinate or compete. Pick whether you're prioritizing conversions or harvests this year — running both full-throttle on the same income window leaves money on the table.

Where to go from here

The full optimization problem is multi-year, multi-strategy, and depends on numbers you wouldn't want to recalculate by hand each year. The full simulator handles it; the AI Advisor proposes the annual mix against your actual data. The Roth Conversion Playbook is the partner blueprint to this one.