What is tax-loss harvesting and is it worth doing?

Tax-loss harvesting is selling losing positions in a taxable brokerage account to convert paper losses into a tax deduction. The IRS lets you offset capital gains dollar-for-dollar with losses, then deduct up to $3,000 of net loss against ordinary income, then carry the remainder forward indefinitely. For someone in the 32% bracket with $14,200 in unrealized losses and no offsetting gains, that’s a real ~$960 federal tax saving this year plus $11,200 banked on Schedule D. The math is more compelling the higher your marginal rate.

How does the offset waterfall actually work?

The IRS applies a three-tier waterfall in Schedule D. Tier 1: short-term losses offset short-term gains; long-term losses offset long-term gains (same-term matching). Tier 2: any net same-term losses cross over to offset gains of the other term. Tier 3: up to $3,000 of remaining net capital loss offsets ordinary income (line 21 of Form 1040). Tier 4: anything beyond $3,000 carries forward to next year and stays on Schedule D until used. The calc simplifies tier 1+2 into a single ‘matched gains’ bucket because user-facing math is identical.

What is the wash sale rule and why does it kill the deduction?

IRC §1091 disallows a capital loss if you (or your spouse, or any IRA you control) buy the same or substantially identical security within 30 days before or after the sale. The disallowed loss isn’t lost: it’s added to the cost basis of the replacement, so you eventually realize it. But it’s deferred, possibly for years. Practical workaround: wait 31 days before rebuying, or swap into a non-substantially-identical equivalent. VTI and ITOT both track total US equity but are different funds; the IRS hasn’t challenged that pair.

Why is my carry-forward so valuable?

Capital-loss carry-forwards never expire (US — different rules apply elsewhere). Every year going forward, that bank of losses can offset $3K of ordinary income, plus an unlimited amount of future capital gains. If you sell appreciated stock five years from now to fund a down payment, your carry-forward absorbs that gain dollar-for-dollar before any tax is owed. The conservative NPV the calculator shows assumes you use the carry next year at your current marginal rate; the real value is usually higher because most users keep the bank for years and apply it against gains they would have been taxed on.

Should I harvest if my loss is small?

Probably not — there's a transaction-cost floor below which harvesting is busy work. Trading commissions are zero at most major brokers now, but the bid-ask spread, the time spent rebalancing, and the audit-trail complexity still consume value. Industry rule of thumb: skip harvests under $500 of tax saved (roughly $1,500 of loss at 32 % marginal). The calculator flags 'Loss too small to matter' below $200 saved as a sharper version of that rule. If your portfolio is $20K and you have a $400 unrealized loss, your time is better spent on contributions than on harvesting that.

Does harvesting change my market exposure?

Only briefly, if you do it right. The standard pattern: sell the losing position at 10:30 AM, immediately buy a non-substantially-identical replacement at 10:31 AM. Your dollar exposure to the asset class is unchanged: you swap one large-cap US ETF for another, or one international ETF for another. The portfolio’s beta, sector weights, and dividend yield are essentially identical the next morning. If you instead sit in cash for 31 days to avoid wash-sale risk, you take market-timing risk; in years when the market is up, cash-on-the-sidelines costs more than the harvest saves.

What about state taxes?

States generally follow the federal capital-gain / capital-loss treatment but at their own rates. California, New York, Oregon, Hawaii, New Jersey, and Minnesota tax capital gains as ordinary income — meaning the harvest saves at your full state marginal rate. Texas, Florida, Tennessee, Nevada, Washington, South Dakota, Wyoming, and Alaska have no state income tax — federal-only savings. The calculator's marginal-rate input is your combined federal + state, so the output is state-inclusive when you enter the right number. If you live in a no-state-tax state, just use your federal marginal.

How does this compare to harvesting *gains* in a 0 % LTCG bracket?

Different tool, same year-end planning bucket. If your taxable income is below ~$48K single / ~$96K married (2026), long-term capital gains are taxed at 0% federal. In that bracket, you should harvest gains, not losses: sell appreciated positions, immediately rebuy them (no wash-sale rule on gains), and step up your basis tax-free. The calc covers loss harvesting only; the gain-harvest decision is a separate trade we plan to add. The two strategies don’t conflict; many investors do both in December (losses against current gains, gains against unused 0% bracket).

What if I have more than $3K in net losses but no gains to offset?

You take the $3K against ordinary income this year, and the rest carries forward to next year’s Schedule D. There’s no penalty and no time limit. A $50,000 unrealized loss with zero offsetting gains saves $960 (32% x $3K) this year and banks $47,000 of future deduction capacity. At a steady $3K/yr ordinary offset, the bank lasts 16 years, but in practice most users hit a year with $20K+ of capital gains long before then (house sale, RSU lot, startup exit) and the bank wipes out a much bigger tax bill in one shot. Carry-forward is the underrated half.

Why doesn't the calculator handle multiple lots?

Because the dominant variable for total tax saved is the dollar value of the loss, not the lot structure. A planned sale of 100 shares purchased in 5 lots at different prices is mathematically identical to a single $14,200 loss for harvesting purposes: the broker tracks lot-level cost basis on your 1099-B and reports the realized loss per lot to the IRS, but tax saved is unchanged. If you’re optimizing WHICH lots to sell to minimize gain instead of maximize loss, that’s a separate calculation (HIFO vs LIFO vs FIFO); the upcoming Crypto Tax Lot Optimizer covers that.

Does this apply to crypto?

Yes for the loss side: crypto sold at a loss in a taxable account generates a capital loss subject to the same Schedule D waterfall. The wash-sale rule, however, has historically not applied to crypto because the IRS classifies it as property, not a security (the 2022 Build Back Better proposal would have changed that, but it didn’t pass). So you can sell BTC at a loss at 10:30 AM, rebuy BTC at 10:31 AM, and claim the full loss, at least until Congress closes the gap, which most year-end tax-planning newsletters expect.

Is the 7 % reinvestment assumption realistic?

It's the conservative midpoint of long-run S&P 500 real returns (about 7 % nominal after a 2 % inflation drag against the 9-10 % nominal historical mean). The calculator surfaces the 10-yr compounded value because the headline 'tax saved' number understates the strategy — a $1,000 tax saving today, reinvested in the same broad-market ETF you held the loss in, becomes ~$1,967 after 10 years. That's the real return on harvesting. If you want a more conservative number, divide the 10-yr value by 2; if more aggressive, multiply by 1.4 for the historical 10-12 % nominal range.

FinanceFree · No signup · 85K+/month

Tax-Loss Harvesting Calculator — Tax Saved + Carry-Forward + 10-Yr Compound Value

Drop your unrealized losses, marginal tax rate, planned gains this year, and term type. Calculator returns tax saved this year, the dollars that carry forward indefinitely, the 10-year compounded value if you reinvest the savings at 7 %, and a wash-sale warning if you flag positions you might rebuy. Built on US capital-loss waterfall (IRC §1211(b) §1091) — same math your CPA runs in December, exposed for any year-end portfolio decision.

Instant result
Private — nothing saved
Works on any device
AI insight included

Reviewed by CalcBold EditorialLast verified May 4, 2026Methodology

Unrealized losses

Total dollar value of losing positions you could sell today. Check your brokerage's unrealized-loss column. Only positions held in taxable accounts qualify — IRA / 401(k) losses do not generate a tax deduction.

Marginal tax rate (federal + state)

Combined marginal rate — federal bracket on your top dollar plus state income tax. Typical US working professional 24-37 % federal + 0-13 % state. This is the rate that applies to the $3K ordinary-income offset.

Planned realized gains this year

Dollar value of profitable positions you expect to sell this calendar year — RSU vesting that auto-sells, rebalancing trims, or a planned sale. Set to 0 if you have none planned; the loss then drops straight to the $3K ordinary-income offset.

Gain term type

Short-term gains are taxed at your ordinary rate; long-term at 0 / 15 / 20 % federal. The calculator uses 15 % as the long-term default — adjust your marginal rate down if you sit in the 0 % LTCG bracket (single < ~$48K AGI), up if in the 20 % bracket (single > ~$520K).

Wash-sale-window positions

Number of the losing positions you intend to rebuy within 30 days — including identical buys in your spouse's account or your IRA. Any positive number triggers a wash-sale warning. The fix is either to wait 31 days or to swap into a non-substantially-identical replacement (VTI ↔ ITOT works; SPY ↔ VOO does not).

Embed builderDrop the Tax-Loss Harvest on your site →Free widget · 3 sizes · custom theme · auto-resizes · no signupGet embed code

What This Calculator Does

The Tax-Loss Harvesting Calculator answers the December question every taxable-account investor faces: are my unrealized losses worth anything if I harvest them now?Drop the dollar value of your losing positions, your combined federal-plus-state marginal rate, any planned realized gains for the year, the term of those gains, and the count of positions you intend to rebuy within 30 days. The calculator returns the tax saved this year, the dollars that flow to next year’s Schedule D as a carry-forward, the 10-year compounded value if you reinvest the savings at 7 %, and a wash-sale warning when the rebuy count is non-zero.

Most calculators stop at “tax saved this year” and miss the entire long-tail value of the carry-forward. Capital-loss carry-forwards never expire in the US — once they’re on Schedule D, they offset future gains dollar-for-dollar plus another $3,000/yr against ordinary income until the bank is exhausted. For a working professional in their 30s with 30+ years of investing ahead, the carry-forward is usually worth more than the first-year deduction. This calculator surfaces both numbers so the harvest decision is made on real economics, not on the visible-only headline.

The Math

The waterfall is straight from IRC §1211(b). Capital losses first match capital gains of the same term (short-term against short-term, long-term against long-term). Excess crosses term, then up to $3,000 of net loss reduces ordinary W-2 / 1099 income on Form 1040 line 21. Anything beyond $3,000 carries forward to next year’s Schedule D, where it stays — usable against future gains plus another $3,000 of ordinary income each year — until exhausted. There is no time limit, no carryback, no wash of value.

The 10-year compounded value uses 7 % nominal — the conservative midpoint of long-run S&P 500 real returns (about 7 % nominal after a typical 2 % inflation drag against a 9-10 % historical mean). This number is the real return on harvesting: the tax savings, reinvested in the same broad-market exposure that produced the loss, become roughly twice the headline at the 10-year mark. Skip the harvest, and that compounded delta is the silent cost of leaving the deduction on the table.

A Worked Example — “$14,200 unrealized, no offsetting gains”

Suppose your unrealized losses total $14,200, your combined marginal rate is 32 %, you have $0of planned realized gains this year, term doesn’t apply (no gains to match), and you’re willing to swap into a non-substantially-identical replacement so wash-sale risk is 0 positions.

Matched against gains: $0 (no gains planned)
Ordinary offset: $3,000 × 32 % = $960 saved this year
Carry-forward: 14,200 − 3,000 = $11,200 banked on Schedule D
10-year compounded value of the $960: $1,888 if reinvested at 7 %
Loss utilization this year: 21.1 % — the carry-forward is most of the value

The verdict line says “Harvest worth doing.” The action implied: sell the losers, immediately rebuy a non-substantially-identical replacement (VTI for ITOT, QQQ for VGT, etc.), book the $960 deduction, and bank the $11,200 carry-forward against any future capital gain you might realize over the next 30 years.

When This Is Useful

Three high-value moments. Year-end rebalancing. The classic December move — losses accumulated over a down market year are worth far more crystallized than left on paper. Most CPAs only have time to flag the obvious, big losses; this calculator lets you run the math on borderline harvest decisions in 60 seconds. Before a planned gain. An RSU lot vesting and auto-selling, a stock-option exercise, a rebalancing trim — anything that produces a realized gain this year is a candidate to be paired with a loss harvest in the same calendar year. After a market drawdown. A 20 % broad-market correction creates harvest opportunities even in long-held positions. Harvesting against the corrected price + immediately rebuying the same exposure resets your basis at the lower market level — a pure-tax win that costs nothing in market exposure.

Common Mistakes

Triggering a wash sale by rebuying too fast.IRC §1091 disallows the loss if you (or your spouse, or any IRA you control) buy the same or substantially identical security within 30 days before or after the sale. The fix is either a 31-day wait or a replacement swap. VTI and ITOT both track total US equity but are different funds from different issuers — the IRS has not challenged that pair. SPY and VOO are riskier — both track the S&P 500, and the conservative consensus avoids the swap.
Harvesting in a tax-advantaged account. Losses in your IRA, 401(k), Roth IRA, HSA, or 529 do NOT generate a tax deduction. Only taxable brokerage accounts qualify. If your only losses are inside an IRA rebalance, the harvest is decorative — there is nothing to deduct on Schedule D.
Believing the $3,000 cap is a hard ceiling on value. The carry-forward is the under-reported half of the math. A $50,000 unrealized loss with zero offsetting gains saves $960 (32 % × $3K) this year and banks $47,000 of future deduction capacity. Most investors hit a year with $20K+ of capital gains long before that bank is exhausted (a house sale, a startup exit, an RSU vesting cliff) and the carry-forward wipes out a much bigger tax bill in one shot than the $960 first-year saving.
Over-harvesting tiny losses.Below ~$200 of tax saved (~$600 of loss at a 32 % rate), the transaction-cost floor — bid-ask spread, time spent, audit-trail complexity — outweighs the deduction. The calculator flags this as “Loss too small to matter.” Better to skip the harvest, hold the position, and wait until the loss pile is meaningful.
Ignoring the term type.Short-term gains are taxed at your ordinary rate; long-term at 15 % default federal. The calculator’s gainsRate input matters because matching $5K of long-term gains saves $750 (15 %) versus matching $5K of short-term gains at a 32 % marginal saves $1,600 — same loss, same calendar year, more than 2x the tax savings. If you have a choice of which gains to realize this year, matching short-term first is almost always correct.
Treating the calculator’s output as a tax opinion. It is not. State variations, AMT exposure, NIIT (3.8 % on investment income above $200K single / $250K married), bracket-creep interactions, and the section-1411 rules for high-earners all shift the real number. The calculator captures the dominant 90 %; the last 10 % is a CPA conversation. Use it to decide whether harvesting is worth the conversation, not as a substitute for it.

Related Calculators

Pair the harvest calc with the RSU Tax Calculator when your year has both vesting income and unrealized losses — RSU auto-sale lots are the most common offsetable gains source for working professionals. Stock option exercises layer in AMT-preference income; the ISO / NSO Calculator handles that side of the year-end planning. After harvesting, the Investment ROI Calculator projects the new cost-basis trajectory for the swapped-in replacement fund. And because the marginal-rate input is the load-bearing variable in the harvest math, run the Tax Bracket Calculator first to confirm your top-dollar federal + state combined rate before committing to the harvest.

Frequently Asked Questions

The most common questions we get about this calculator — each answer is kept under 60 words so you can scan.

What is tax-loss harvesting and is it worth doing?
Tax-loss harvesting is selling losing positions in a taxable brokerage account to convert paper losses into a tax deduction. The IRS lets you offset capital gains dollar-for-dollar with losses, then deduct up to $3,000 of net loss against ordinary income, then carry the remainder forward indefinitely. For someone in the 32% bracket with $14,200 in unrealized losses and no offsetting gains, that’s a real ~$960 federal tax saving this year plus $11,200 banked on Schedule D. The math is more compelling the higher your marginal rate.
How does the offset waterfall actually work?
The IRS applies a three-tier waterfall in Schedule D. Tier 1: short-term losses offset short-term gains; long-term losses offset long-term gains (same-term matching). Tier 2: any net same-term losses cross over to offset gains of the other term. Tier 3: up to $3,000 of remaining net capital loss offsets ordinary income (line 21 of Form 1040). Tier 4: anything beyond $3,000 carries forward to next year and stays on Schedule D until used. The calc simplifies tier 1+2 into a single ‘matched gains’ bucket because user-facing math is identical.
What is the wash sale rule and why does it kill the deduction?
IRC §1091 disallows a capital loss if you (or your spouse, or any IRA you control) buy the same or substantially identical security within 30 days before or after the sale. The disallowed loss isn’t lost: it’s added to the cost basis of the replacement, so you eventually realize it. But it’s deferred, possibly for years. Practical workaround: wait 31 days before rebuying, or swap into a non-substantially-identical equivalent. VTI and ITOT both track total US equity but are different funds; the IRS hasn’t challenged that pair.
Why is my carry-forward so valuable?
Capital-loss carry-forwards never expire (US — different rules apply elsewhere). Every year going forward, that bank of losses can offset $3K of ordinary income, plus an unlimited amount of future capital gains. If you sell appreciated stock five years from now to fund a down payment, your carry-forward absorbs that gain dollar-for-dollar before any tax is owed. The conservative NPV the calculator shows assumes you use the carry next year at your current marginal rate; the real value is usually higher because most users keep the bank for years and apply it against gains they would have been taxed on.
Should I harvest if my loss is small?
Probably not — there's a transaction-cost floor below which harvesting is busy work. Trading commissions are zero at most major brokers now, but the bid-ask spread, the time spent rebalancing, and the audit-trail complexity still consume value. Industry rule of thumb: skip harvests under $500 of tax saved (roughly $1,500 of loss at 32 % marginal). The calculator flags 'Loss too small to matter' below $200 saved as a sharper version of that rule. If your portfolio is $20K and you have a $400 unrealized loss, your time is better spent on contributions than on harvesting that.
Does harvesting change my market exposure?
Only briefly, if you do it right. The standard pattern: sell the losing position at 10:30 AM, immediately buy a non-substantially-identical replacement at 10:31 AM. Your dollar exposure to the asset class is unchanged: you swap one large-cap US ETF for another, or one international ETF for another. The portfolio’s beta, sector weights, and dividend yield are essentially identical the next morning. If you instead sit in cash for 31 days to avoid wash-sale risk, you take market-timing risk; in years when the market is up, cash-on-the-sidelines costs more than the harvest saves.
What about state taxes?
States generally follow the federal capital-gain / capital-loss treatment but at their own rates. California, New York, Oregon, Hawaii, New Jersey, and Minnesota tax capital gains as ordinary income — meaning the harvest saves at your full state marginal rate. Texas, Florida, Tennessee, Nevada, Washington, South Dakota, Wyoming, and Alaska have no state income tax — federal-only savings. The calculator's marginal-rate input is your combined federal + state, so the output is state-inclusive when you enter the right number. If you live in a no-state-tax state, just use your federal marginal.
How does this compare to harvesting *gains* in a 0 % LTCG bracket?
Different tool, same year-end planning bucket. If your taxable income is below ~$48K single / ~$96K married (2026), long-term capital gains are taxed at 0% federal. In that bracket, you should harvest gains, not losses: sell appreciated positions, immediately rebuy them (no wash-sale rule on gains), and step up your basis tax-free. The calc covers loss harvesting only; the gain-harvest decision is a separate trade we plan to add. The two strategies don’t conflict; many investors do both in December (losses against current gains, gains against unused 0% bracket).
What if I have more than $3K in net losses but no gains to offset?
You take the $3K against ordinary income this year, and the rest carries forward to next year’s Schedule D. There’s no penalty and no time limit. A $50,000 unrealized loss with zero offsetting gains saves $960 (32% x $3K) this year and banks $47,000 of future deduction capacity. At a steady $3K/yr ordinary offset, the bank lasts 16 years, but in practice most users hit a year with $20K+ of capital gains long before then (house sale, RSU lot, startup exit) and the bank wipes out a much bigger tax bill in one shot. Carry-forward is the underrated half.
Why doesn't the calculator handle multiple lots?
Because the dominant variable for total tax saved is the dollar value of the loss, not the lot structure. A planned sale of 100 shares purchased in 5 lots at different prices is mathematically identical to a single $14,200 loss for harvesting purposes: the broker tracks lot-level cost basis on your 1099-B and reports the realized loss per lot to the IRS, but tax saved is unchanged. If you’re optimizing WHICH lots to sell to minimize gain instead of maximize loss, that’s a separate calculation (HIFO vs LIFO vs FIFO); the upcoming Crypto Tax Lot Optimizer covers that.
Does this apply to crypto?
Yes for the loss side: crypto sold at a loss in a taxable account generates a capital loss subject to the same Schedule D waterfall. The wash-sale rule, however, has historically not applied to crypto because the IRS classifies it as property, not a security (the 2022 Build Back Better proposal would have changed that, but it didn’t pass). So you can sell BTC at a loss at 10:30 AM, rebuy BTC at 10:31 AM, and claim the full loss, at least until Congress closes the gap, which most year-end tax-planning newsletters expect.
Is the 7 % reinvestment assumption realistic?
It's the conservative midpoint of long-run S&P 500 real returns (about 7 % nominal after a 2 % inflation drag against the 9-10 % nominal historical mean). The calculator surfaces the 10-yr compounded value because the headline 'tax saved' number understates the strategy — a $1,000 tax saving today, reinvested in the same broad-market ETF you held the loss in, becomes ~$1,967 after 10 years. That's the real return on harvesting. If you want a more conservative number, divide the 10-yr value by 2; if more aggressive, multiply by 1.4 for the historical 10-12 % nominal range.