Tax Loss Harvesting vs Roth Conversion: Which Should You Do First in 2026?
If you have $50,000 sitting in a taxable brokerage account and another $200,000 in a traditional IRA, you’ve probably been told to do both tax loss harvesting and a Roth conversion this year. The trouble is, doing them in the wrong order can quietly cost you thousands — and your CPA almost certainly hasn’t run the math on the sequencing.
The question of tax loss harvesting vs Roth conversion isn’t really an either/or. It’s a question of which one you should run first, in which year, and at which income level. The answer matters because the two strategies pull on the same tax brackets in opposite directions: harvesting losses lowers this year’s taxable income, while a Roth conversion deliberately raises it. Run them in the wrong sequence and you’re paying tax at a higher marginal rate than necessary.
I started running both strategies in my own portfolio a few years back, mostly out of curiosity about whether the much-praised approach actually moved the needle once you accounted for the brackets they collide with. The honest answer: yes, but the order matters more than the strategy choice itself. Below is the comparison framework I use, the table I built to sanity-check sequencing in any given year, and the five questions a reader asked me last month that turned out to be the same questions every DIY investor eventually runs into.
Tax loss harvesting vs Roth conversion: the 60-second comparison
Before we get into the weeds, here’s the quick mental model. Tax loss harvesting (TLH) is the practice of selling losing positions in a taxable account to bank a capital loss, which you can use to offset capital gains and up to $3,000 of ordinary income per year. Roth conversion is the act of moving money from a pre-tax account (traditional IRA, traditional 401(k) if your plan allows in-service rollovers) into a Roth IRA, paying ordinary income tax on the converted amount today in exchange for tax-free withdrawals later.
One is a tax deferral strategy that compresses your current bill. The other is a tax acceleration strategy that pre-pays your future bill. They are opposites in the year you execute them — and that’s why sequencing matters.
| Feature | Tax Loss Harvesting | Roth Conversion |
|---|---|---|
| Effect on this year’s tax bill | Lowers it | Raises it |
| Which account it touches | Taxable brokerage only | Pre-tax retirement (IRA, sometimes 401(k)) |
| Annual cap | No cap on offsetting gains; $3,000 against ordinary income | No federal limit (income or amount) |
| Wash-sale rule applies? | Yes — 30-day window each side | No |
| Best year to use it | High-income years with realized gains | Low-income years (sabbatical, between jobs, early retirement) |
| Time value | Defers tax; benefits compound | Pre-pays tax; benefits compound tax-free |
| State tax consideration | Lowers state taxable income too | Increases state taxable income (consider moves) |
| Reversible? | Once realized, the loss is locked in | No — Tax Cuts and Jobs Act killed recharacterization in 2018 |
That last row is the one most people forget. Before 2018 you could undo a Roth conversion if the market dropped after you ran it. Now, the IRS confirms recharacterizations are no longer allowed for Roth conversions. Once you convert, it’s permanent — which is why the sequencing question is more important than it used to be.
The pros and cons of tax loss harvesting vs Roth conversion in the same year
It is technically possible to do both strategies in the same calendar year. Most people probably shouldn’t, because the strategies cancel out the brackets they were each trying to optimize for. Here’s how I think about the trade-offs on each side.
Tax loss harvesting: where it wins and where it doesn’t
TLH genuinely shines when you have realized short-term capital gains (taxed at ordinary rates up to 37% federally) or you’re sitting in a high tax bracket. Vanguard’s 2023 research paper on TLH found that the strategy can add roughly 0.95% in annual after-tax alpha for high-income investors in volatile markets — meaningful, but not life-changing. The benefit scales with your marginal rate; at the 12% bracket, TLH is almost a wash once you factor in the lower future basis.
Where it falls apart: small portfolios with few lots, dividend-reinvesting accounts that constantly trigger wash sales, and years where you also want to claim the 0% long-term capital gains rate (single filers under $48,350 in taxable income for 2026 per IRS Topic 409). If you’re harvesting losses in a year you could have realized gains at 0%, you’re literally throwing away free tax savings. We dug into the math for sub-$50K accounts in our analysis of tax loss harvesting for small portfolios — the threshold where TLH starts paying for itself is usually higher than people assume.
Roth conversion: where it wins and where it doesn’t
Roth conversions are the single most powerful lever for people in unusually low-income years. The classic windows: the years between retirement and Social Security claiming, a sabbatical or career break, the year after a job loss, or — for high earners — any year where a business loss compresses ordinary income. In those windows you can convert tens of thousands of dollars at 12% or 22% federal rates that would otherwise have come out at 24%, 32%, or higher in your 70s when required minimum distributions begin.
They fail when you do them in a peak-earning year without thinking about IRMAA (the Medicare premium surcharge) or the net investment income tax (NIIT). A $50,000 conversion that pushes your modified AGI above the $206,000 single filer threshold can trigger Medicare Part B and D surcharges two years later, per the Social Security Administration’s IRMAA schedule. They also fail for people who can’t pay the conversion tax out of taxable assets — paying conversion tax from the IRA itself usually wipes out most of the benefit.
The other consideration most DIY investors miss: state residency. If you’re planning a move from California (13.3% top marginal rate) to Texas (no income tax), running the conversion after establishing the new residency saves real money. Our walkthrough of the backdoor Roth IRA covers the related pro-rata trap for anyone who has a mix of pre-tax and after-tax IRA dollars.
Which should you do first? A decision framework
Here’s the rule I land on after running this exercise for myself and a half-dozen friends who asked: do tax loss harvesting in years you have high ordinary income, and do Roth conversions in years you have unusually low ordinary income. When in doubt, lean toward Roth conversions earlier, because the tax-free compounding window is the asset.
If you’re staring at a year where you might want to do both, the decision usually comes down to three questions:
- What’s your current marginal federal bracket? If you’re in the 24% bracket or higher, harvesting losses first to drop yourself toward the 22% threshold can save real money on the conversion. The 2026 single-filer brackets jump from 22% to 24% at $103,350 in taxable income (per IRS 2026 inflation adjustments). A few thousand in harvested losses can be the difference.
- Do you have meaningful realized gains this year? If you sold a rental, exercised options, or rebalanced a concentrated stock position, harvest losses first to offset those gains dollar-for-dollar. Then size the Roth conversion to fill whatever room is left in your target bracket.
- Will next year be lower-income? If so, defer the Roth conversion to next year and just do TLH this year. Stacking the conversion into a lower-income year is almost always worth more than running both in the same year.
Want to see how much more your Roth balance is worth over 20 years versus a taxable account?
A worked example: tax loss harvesting vs Roth conversion at $135,000 of income
Numbers always make the trade-off concrete. Picture a single filer in 2026 with $135,000 of W-2 wages, $25,000 in a taxable brokerage with $8,000 of unrealized losses, and $400,000 in a traditional IRA. They want to know whether to harvest the $8,000 loss, run a $20,000 Roth conversion, or both.
Without any moves, taxable income lands around $120,000 after the $15,000 standard deduction — squarely in the 24% bracket. A $20,000 Roth conversion taxes the full $20,000 at 24% = $4,800 of additional federal tax.
Now harvest the $8,000 loss first. You can only use $3,000 against ordinary income this year, so taxable income drops to $117,000. The remaining $5,000 of losses carries forward to future years (a real, durable asset). Then run the $20,000 conversion: the first $400 still hits at 24% (we just dipped below $103,350 by $3,000), but the rest is unavoidably in the 24% bracket. Net savings this year: the $3,000 deduction × 24% = $720.
Now reverse the sequence. Run the conversion first: $20,000 hits at 24% = $4,800. Then harvest the loss: $3,000 against ordinary income at 24% = $720 saved. Same answer, $720 net. The sequencing doesn’t matter in a single year if both moves are happening at the same marginal rate.
The sequencing only matters when one of the moves crosses a bracket. If our hypothetical filer had $103,500 of taxable income before any moves, harvesting first would slide them into the 22% bracket, then the conversion would fill back into the 22% bracket — saving 2% on $17,000 of conversion = $340 versus reversing the order. The decision tree is essentially: run whichever move crosses the most beneficial bracket boundary first.
When you should do neither
One thing that gets lost in personal finance content: there are real years where the right answer is to do nothing. If you are in the 0% long-term capital gains bracket (under $48,350 single / $96,700 married for 2026), you should be realizing gains, not harvesting losses. If your income is so low that a Roth conversion would still happen at 10% federal, you’re better off just contributing directly to a Roth IRA and leaving the conversion bucket alone. And if you don’t have cash outside the IRA to pay the conversion tax bill, almost every analysis I’ve seen says skip the conversion. For younger readers thinking about which account to fund, the direct contribution decision is often more impactful than any conversion strategy.
Frequently asked questions
Can I do tax loss harvesting and a Roth conversion in the same calendar year?
Yes. They live in different account types and are reported on different forms (Schedule D for harvested losses, Form 1099-R for the conversion). The question isn’t whether you can — it’s whether the marginal rates make sense. As shown above, doing both in a year where you’d cross a bracket boundary in the right direction is worth it; doing both in a year where you simply stay in the same bracket adds no value beyond doing either alone.
Does tax loss harvesting hurt a future Roth conversion?
Not directly. TLH lowers cost basis in your taxable account, which means more capital gains when you eventually sell. A Roth conversion is unrelated — it draws from a pre-tax retirement account. The only indirect link is that lower current-year ordinary income (thanks to the $3,000 TLH deduction) leaves slightly more room in the current bracket for a conversion.
Should I do the Roth conversion in December or earlier in the year?
Most planners suggest December because by then you know your actual income for the year and can size the conversion precisely to fill a bracket without overshooting. The trade-off is you give up 11 months of tax-free compounding inside the Roth. In the same way TLH benefits from monitoring throughout the year, Roth conversions executed earlier with a conservative size and then “topped up” in December often capture both precision and compounding.
What if the market drops right after my Roth conversion?
You can’t undo it anymore — the Tax Cuts and Jobs Act removed recharacterization in 2018. This is the single biggest argument for sizing conversions conservatively and spreading them across multiple years rather than doing one large conversion. You’ll still benefit from tax-free recovery inside the Roth, but you’ve paid tax on a value that no longer exists.
Does state tax change the answer between tax loss harvesting and Roth conversion?
Yes, significantly. High-tax states amplify both the cost of a conversion (you’ll pay state income tax on the converted amount) and the benefit of harvesting losses (the deduction reduces state taxable income too). If you live in California, New York, or New Jersey and you’re planning a move to a no-income-tax state in the next few years, defer the Roth conversion until after the move. The state tax difference on a $50,000 conversion can be $4,000–$6,000.
Photo by Kelly Sikkema on
Unsplash
