When NOT to Harvest Losses: Three Cases Where the Default Move Is Wrong

Tax-loss harvesting gets treated like a free lunch. The market drops, you swap the position, you capture the loss, done. Most of the time that logic holds. But in my work with high earners, I've seen the "always harvest" reflex cost clients real money in three specific situations. Each one has a clean explanation, and knowing them will sharpen how you think about every future down-market decision.

Why the Default "Harvest Everything" Rule Falls Apart

A harvested loss saves you money only if you were going to owe taxes on the offsetting gain or income in the first place. That sounds obvious. It becomes a lot less obvious when you look at what the tax code actually says about long-term capital gains (LTCG) rates, charitable giving rules, and transaction economics.

The parent pillar, Tax-Loss Harvesting: How to Turn a Down Market Into Real Tax Savings, covers the full mechanics, the wash-sale rule, and how harvesting stacks with Roth conversions. This page is narrower: three cases where you should close the spreadsheet and do nothing.

The Three Situations Where Harvesting Is the Wrong Move

Case 1: You're in the 0% long-term capital gains bracket and you expect to stay there.

In 2025, the 0% LTCG rate applies to taxable income up to roughly $48,350 for single filers and $96,700 for married filing jointly. If your income sits below those thresholds, a realized long-term gain costs you nothing in federal tax. Harvesting a loss to offset a gain you were going to owe zero on produces zero benefit. You also reset the cost basis lower, which means a larger taxable gain in the future when you're likely in a higher bracket. That's not a wash. That's a self-inflicted future tax bill.

HENRYs (high earners, not rich yet) occasionally land in this bracket during a transition year, a sabbatical, a gap between jobs, or the early years of a business. The instinct is still to harvest because harvesting "feels productive." Resist it.

Case 2: You plan to gift the position to a donor-advised fund (DAF) or a qualified charity within the next year.

When you contribute an appreciated security directly to a DAF or a 501(c)(3), two things happen simultaneously: you avoid paying capital gains tax on the embedded appreciation, and you take a charitable deduction for the full fair market value. The IRS, in effect, lets you gift the pre-tax asset.

A position that is currently underwater doesn't get that treatment because there's no embedded gain to preserve. The right sequence with a losing position is to sell it, capture the loss on your own return, and then donate the cash proceeds. But if you were already planning to send that holding to charity regardless of its performance, harvesting the loss first and then donating cash is only marginally better than doing nothing, and the difference shrinks fast when you factor in transaction timing and the wash-sale window you now have to manage.

The cleaner rule: any position you intend to donate in the next twelve months should be evaluated as a potential charitable gift first, a harvest candidate second. If it has a loss, sell it and give cash. If it has a gain, give the shares. Don't let a short-term paper loss flip the analysis before you've thought through the charitable angle.

Case 3: The position is small and transaction costs, time, and complexity eat the benefit.

This one is underappreciated. A $4,000 loss in the 24% bracket saves you $960 in federal tax. That sounds worthwhile until you account for: bid-ask spread on the exit and re-entry, potential capital gains on the replacement position later, advisor time for monitoring the wash-sale clock, and state tax complexity if you're in a state that doesn't conform to federal capital loss treatment.

For small positions, especially thinly traded securities or positions held in accounts with transaction fees, the net benefit can slip below a few hundred dollars. I'm not saying $300 isn't real money. I'm saying it's not worth introducing tracking errors, wash-sale risk on a subsequent purchase, or complexity into an otherwise clean portfolio for $300.

A useful rough threshold: if the tax savings don't exceed the sum of transaction costs plus one hour of actual review time, the harvest is probably not worth doing.

When This Logic Applies vs. When It Doesn't

This framework applies specifically to losses you're considering harvesting for the first time in a position you might otherwise hold, donate, or simply leave alone. It does not apply when you have large, offsetting realized gains in the same year that you need to neutralize. In that scenario, even a marginal harvest is probably worth the friction.

The broader point: harvesting is a tool inside the Soil layer of a tax-aware plan, where asset location and tax architecture decisions compound over decades. A harvested loss that resets your basis at the wrong time or eliminates a future charitable deduction opportunity isn't tax planning. It's tax activity. Those are different things.

How This Connects to Tax-Loss Harvesting

The parent pillar at /strategies/tax-loss-harvesting covers the full case for why systematic harvesting creates real long-run value. The cases above are the exceptions that make the rule more precise. Good tax planning is knowing both sides.

Frequently Asked Questions

What if I'm in the 0% bracket now but expect to move to a higher bracket next year?

Then the calculus shifts. If you'll owe taxes on future gains, harvesting losses now to carry forward makes sense. The question is always about when you'll actually use the loss, not just whether you can generate one.

Can I still harvest a loss if I plan to donate to a DAF more than a year out?

Yes. A twelve-month window is a useful rule of thumb, not a legal threshold. The further out the planned donation, the more the harvested loss stands on its own merits. Just make sure you're not double-counting: harvesting the loss and then expecting to avoid gains tax on the same position later by donating it.

Does the "small position" threshold vary by income?

It does. The higher your marginal rate, the larger the tax benefit per dollar of loss, so the friction threshold shifts. A $4,000 loss is worth more at 37% than at 24%. Adjust accordingly rather than using a fixed dollar cutoff.

Is wash-sale risk different in a DAF context?

Yes. A DAF is a separate legal entity, so buying a substantially identical security inside your DAF after harvesting it in a taxable account can trigger the wash-sale rule. This is one more reason to think through the donation strategy before executing a harvest.

My robo-advisor harvests losses automatically. Should I turn that off?

Not necessarily, but you should understand what it's doing. Automated harvesting doesn't know about your planned charitable giving, your bracket situation in a transition year, or your small-position economics. Review the logic annually against your actual circumstances.

What to Do Next

1. Check your current-year bracket before harvesting anything. Confirm whether realized long-term gains this year actually cost you federal tax.
2. Identify any holdings you're likely to donate in the next twelve months and flag them as charitable candidates before treating them as harvest candidates.
3. For positions with small embedded losses, run the numbers: estimated tax savings minus transaction costs minus the value of your time. If the result is under a few hundred dollars, let it go.
4. For the full framework on how tax-loss harvesting works when it does make sense, including how to combine it with Roth conversions for larger combined benefit, read the parent pillar at /strategies/tax-loss-harvesting.

The information provided is for educational purposes only and does not constitute investment, legal, or tax advice. Tax law changes frequently — verify current rules before acting. Consult with qualified professionals for guidance specific to your situation.