Capital losses from a tax-loss harvest are not interchangeable. Under the Internal Revenue Code's capital gain and loss netting rules, a short-term capital loss and a long-term capital loss follow a specific sequence of application — and that sequence has concrete implications for which harvested loss produces a larger after-tax benefit per dollar. The mechanics are straightforward once the netting order is clear, but the distinction is frequently overlooked in discussions that treat "harvested losses" as a single undifferentiated resource.
This article covers how the IRC §1222 netting rules work, why short-term losses may carry a higher after-tax value per dollar than long-term losses in many situations, and what that implies for harvest priority decisions. For the foundational mechanics of how tax-loss harvesting works, TLH 101 covers the basics.
How does IRC §1222 determine which capital losses apply to which gains?
How does IRC §1222 determine the order in which capital losses offset capital gains, and why does the netting sequence matter?
Under IRC §1222, short-term capital losses are first netted against short-term capital gains, and long-term capital losses against long-term capital gains — producing a net short-term figure and a net long-term figure that are then combined, with any remaining net loss crossing over to offset a net gain in the other character bucket.
The four-step netting sequence under §1222 proceeds as follows:
- Short-term netting: All short-term capital gains minus all short-term capital losses produces a net short-term capital gain (NSTCG) or net short-term capital loss (NSTCL) for the year.
- Long-term netting: All long-term capital gains minus all long-term capital losses produces a net long-term capital gain (NLTCG) or net long-term capital loss (NLTCL).
- Same-sign combination: When both net positions are gains, they stack — each is taxed at its applicable rate. When both are losses, they are combined for the §1211(b) deduction against ordinary income.
- Cross-character offset: When one net position is a gain and the other is a loss, the net loss offsets the net gain in the other bucket — this cross-bucket step is where loss character becomes visible in dollar terms.
The cross-character offset in Step 4 is the crux of the issue. A net short-term capital loss crossing over to reduce a long-term capital gain reduces a gain that might otherwise be taxed at potentially preferential rates. A net long-term capital loss crossing over to reduce a short-term capital gain reduces a gain potentially taxed at higher ordinary income rates. The direction of the offset affects the per-dollar value of the result. (Source: IRC §1222; IRS Publication 550, Chapter 4.)
Why may a dollar of short-term capital loss be worth more than a dollar of long-term loss?
Why can a dollar of short-term capital loss potentially produce a larger tax benefit than a dollar of long-term capital loss?
Short-term capital gains are taxed at ordinary income rates — potentially up to 37% under IRC §1(h) for higher-income taxpayers — while long-term capital gains face preferential rates that may be as low as 0% or potentially up to 20%, so a short-term loss applied to a short-term gain may eliminate a larger per-dollar tax liability than the same loss amount applied to a long-term gain.
The rate differential means that the same nominal loss amount can carry a different after-tax value depending on what it offsets. Consider a simplified illustration: an investor realizes an approximately $10,000 short-term capital gain and an approximately $10,000 long-term capital gain in the same year. If the investor is in the highest bracket, the short-term gain may be taxed at potentially up to 37% and the long-term gain at potentially up to 20%. An approximately $10,000 short-term capital loss applied to the short-term gain may eliminate approximately $3,700 of potential tax; an approximately $10,000 long-term capital loss applied to the long-term gain may eliminate approximately $2,000 — from the same nominal loss amount in each case.
This is the core reason tax-aware lot selection systems that weight losses by applicable tax rate typically prefer short-term lots over long-term lots carrying equal dollar losses. The lot selection methods article covers how HIFO and rate-weighted scoring approaches reflect this priority in practice.
How does the §1211(b) ordinary income deduction interact with capital loss character?
How does the §1211(b) annual limit on using capital losses against ordinary income depend on whether the losses are short-term or long-term?
When net capital losses exceed net capital gains for the year, IRC §1211(b) allows individual taxpayers to deduct up to approximately $3,000 of the net capital loss against ordinary income annually — and per IRS guidance, any short-term component of the net loss is applied first before drawing on the long-term component.
The approximately $3,000 annual ordinary income deduction ceiling applies to both single and joint filers. When the net capital loss exceeds approximately $3,000, the excess carries forward indefinitely under IRC §1212, with character preserved. The ordering rule — short-term applied before long-term — means that in a year where total net capital losses exceed current-year gains by more than approximately $3,000, investors with significant short-term losses may offset up to approximately $3,000 of potentially higher-rate ordinary income before any long-term component is used at all.
The character-preservation rule under IRC §1212 is significant for multi-year planning. A short-term loss that carries forward retains its short-term character in future years — arriving in the next year as a short-term loss and entering the §1222 netting process accordingly. For a detailed look at how carryforward mechanics work across years, including how character survives the carry, see the capital loss carryforward article.
What does loss character priority mean for choosing which positions to harvest?
How does understanding loss character help inform decisions about which positions to prioritize when harvesting?
When both short-term and long-term unrealized losses are available, many tax-aware investors prioritize harvesting short-term positions first in years when they also have short-term gains — because offsetting a short-term gain directly may produce a higher per-dollar tax benefit than the same loss amount applied after a cross-character offset step.
Several practical considerations shape harvest priority decisions:
- Gain composition matters. An investor with primarily short-term gains that year has more potential benefit from prioritizing short-term losses, since those losses can offset the highest-rate gains directly. An investor with primarily long-term gains may still benefit from short-term losses when they cross over, but the direct rate advantage is less pronounced.
- The carry value of short-term losses. Investors who expect to realize short-term gains in future years — from stock compensation, frequent rebalancing, or other strategies — may find short-term carryforward losses more valuable than long-term losses for the same reason: they enter the netting process in the higher-rate bucket in future years.
- Near-term holding period. Lots approaching the one-year mark for long-term treatment create a specific case. Harvesting them before they cross converts a near-long-term position into a fresh short-term lot, changing both the character of the immediate harvest and the character of future gains on the replacement. Whether that conversion is net-positive depends on the investor's expected gain composition in that year and future years.
The zero-tax exit strategy article looks at how a multi-year loss bank can be deployed strategically to offset gains at potentially minimal federal tax cost — and why the character composition of that loss reserve affects what it may ultimately be worth.
How does HarvestEngine incorporate loss character into harvest decisions?
How does HarvestEngine's optimizer use capital loss character when scoring candidate lots for a potential harvest?
HarvestEngine's default lot-selection strategy scores each candidate lot by multiplying its unrealized loss by the applicable tax rate for that lot's character — short-term lots receive a higher effective rate weight than long-term lots with equal dollar losses — so short-term candidates are generally ranked ahead of comparable long-term losses when both are available.
The scoring formula is implemented in the optimizer module (src/tlh/optimizer/long_only.py). For each candidate lot, the score is approximately: unrealized loss × effective_tax_rate(term), where the effective rate for short-term positions reflects applicable ordinary income rates under IRC §1(h) and the rate for long-term positions reflects applicable preferential rates. Two lots each carrying an approximately equal unrealized loss receive different scores if one is short-term and the other is long-term — and the higher-scored lot is selected first.
The engine also applies a holding-period protection filter: lots within a configurable number of days of the one-year long-term threshold are typically not harvested unless the unrealized loss is large enough to justify resetting the holding period on the replacement. This avoids inadvertently converting a nearly-long-term position into a fresh short-term lot, which would change the character of future activity on that position. For the full IRS code context behind these mechanics — §1(h), §1222, §1211, and §1212 — the IRS code TLH cheat sheet has all of them in one reference.
Read this next with lot selection methods, capital loss carryforward, the zero-tax exit strategy, tax alpha explained, and the IRS code TLH cheat sheet.