Direct indexing to unwind concentrated employer stock

You have decided to unwind the $400,000 of your employer's stock that has quietly become half your net worth, and the obstacle is not nerve — it is the capital gains tax waiting on every lot you sell. What if the very act of building the diversified portfolio that replaces the position also manufactured the losses that pay the tax on selling it? Direct indexing turns the diversification you need and the tax bill you fear into two halves of the same machine.

The 30-second answer: Direct indexing means owning the individual stocks of an index inside a separately managed account instead of buying the index fund, so a manager can harvest losses on the names that are down while keeping your overall index exposure intact — and can exclude your employer and its sector entirely. Those harvested losses offset the capital gains you realize as you sell down your concentrated employer stock, so the diversification sleeve funds the tax on the unwind. It is most powerful for a large, gain-heavy position unwound over several years. The trade-offs are real: higher fees than an index fund (often 0.2% to 0.4%), more complexity, and a harvesting benefit that decays as the account appreciates.

This guide is a spoke off the diversification playbook: it takes one tool from that menu — direct indexing — and works it through in full for the specific job of unwinding a concentrated employer-stock position. It assumes you are a US resident filing a US return, and that you have already decided you want out of the concentration; the question here is purely how to do it tax-efficiently. We will build the mechanics from the ground up, run the math on a multi-year unwind, and be honest about the years when this tool is not worth its fee.

What direct indexing actually is

Most people own an index by buying a single fund — an S&P 500 ETF, say. You hand over your money, you get one security, and that security holds the 500 underlying stocks for you. It is cheap, simple, and tax-dumb. When the fund is up 8% on the year, you have an 8% gain you cannot touch without selling, even though perhaps 150 of the 500 underlying names are individually down. Those internal losses are invisible to you, because you own the wrapper, not the contents.

Direct indexing removes the wrapper. Inside a separately managed account (SMA), you hold the individual constituent stocks directly — not all 500, usually a representative sample of 150 to 400 names chosen to track the index closely. Your overall exposure is the same broad-market exposure you wanted. But now the contents are yours to manage at the individual-lot level. When one name falls below what you paid for it, the manager can sell that specific lot, book the loss, and immediately buy a different but highly correlated name so your index exposure barely moves. The account keeps tracking the index while quietly producing tax losses out of its own internal dispersion.

Two features make this the right tool for a concentrated employer-stock holder specifically. First, the harvested losses can be aimed at the gains you generate when you sell your employer stock. Second, the account is customisable: you can tell the manager to exclude your employer entirely, and usually its whole sector, so the new money genuinely diversifies away from your concentration instead of buying more of it through the index.

Worked example. Why the wrapper hides your losses

Picture a simplified index of just four equally weighted stocks, $25,000 in each, $100,000 total. Over a year the prices move apart, which is normal — dispersion happens even in a flat market.

Holding	Start value	End value	Gain/loss
Stock A	$25,000	$32,000	+$7,000
Stock B	$25,000	$27,000	+$2,000
Stock C	$25,000	$21,000	−$4,000
Stock D	$25,000	$23,000	−$2,000
Total	$100,000	$103,000	+$3,000

If you owned this through an index fund, you would see one number: up $3,000. There is nothing to harvest, because the fund is a single security sitting at a gain. If you owned the four stocks directly, you would see something different: stocks C and D are down a combined $6,000, and you can sell them to harvest $6,000 of losses while the account is still up $3,000 overall. The manager rebuys correlated replacements for C and D so your exposure is unchanged. The fund holder has a $3,000 gain and zero usable losses; the direct-index holder has the same $3,000 of economic gain and $6,000 of harvested losses to deploy against gains elsewhere. The dispersion was always there — only the wrapper decided whether you could use it.

Bottom line: Direct indexing replaces the index-fund wrapper with the individual stocks inside it, so a manager can harvest the losses created by normal dispersion among constituents and exclude your employer entirely, all while tracking the same index — the fund holder cannot touch those internal losses, the direct-index holder can.

How harvested losses fund the unwind

Here is the mechanism that makes this more than an academic curiosity for a concentrated holder. When you sell a lot of appreciated employer stock, you realize a capital gain, and that gain is taxable. Harvested capital losses offset capital gains dollar for dollar before any tax is calculated. So if your direct-indexing sleeve produces losses while you sell employer stock at a gain, the losses absorb the gains, and your tax bill shrinks accordingly.

The ordering rules matter, so be precise. Losses first offset gains of the same character — long-term losses against long-term gains, short-term against short-term. Then any net loss of one character offsets net gain of the other. Then up to $3,000 of net loss per year can offset ordinary income (wages, interest). Anything still left over is carried forward indefinitely to future years, never expiring. For a concentrated holder, the relevant flow is usually simple: long-term losses from the sleeve offset long-term gains from selling down the employer position you have held more than a year.

There is one asymmetry worth committing to memory. The wash-sale rule applies to losses but not to gains. A harvested loss is disallowed if you buy a substantially identical security within 30 days before or after the sale — which is exactly why the manager buys a correlated-but-not-identical replacement rather than rebuying the same stock. But there is no wash-sale rule on gains, so realizing gains to use up losses is never blocked. Note also that the wash-sale rule reaches across all your accounts, including a spouse's and your IRAs, so the sleeve has to be coordinated with everything else you hold.

Worked example. A volatile year that absorbs a $20,000 gain

Suppose you fund a direct-indexing account and instruct it to exclude your employer. The market has a volatile year — plenty of dispersion among constituents — and the manager harvests $20,000 of losses while keeping you tracking the index. In the same year you trim your employer stock and realize a $20,000 long-term capital gain. You are a high earner, so your long-term rate is 15% plus the 3.8% Net Investment Income Tax (NIIT) — 18.8% all in.

Line	Without direct indexing	With direct indexing
Long-term gain from selling employer stock	$20,000	$20,000
Harvested losses from the sleeve	$0	−$20,000
Net taxable long-term gain	$20,000	$0
Tax at 18.8%	$3,760	$0
Tax saved this year	—	$3,760

The harvested losses cancel the gain exactly, and the $3,760 of tax you would otherwise owe never comes due. You have diversified $20,000 out of your employer stock this year at zero federal tax cost on the gain, and the money you used to do it — the direct-indexing account — is the thing that produced the offsetting losses. This is the whole idea in one table: the diversification sleeve pays the tax on the unwind. Whether a single year actually delivers $20,000 of harvest depends on volatility and account size; a calmer year on the same account might yield far less.

Bottom line: Harvested losses offset realized gains dollar for dollar before tax, then up to $3,000 of ordinary income, then carry forward indefinitely — so a direct-indexing sleeve that produces losses while you sell employer stock at a gain can erase the tax on the unwind, with the wash-sale rule applying only to the losses and never to the gains.

Excluding your employer is the customisation that matters

A plain S&P 500 fund is not neutral for you. If you work at one of the large companies that dominate the index, that fund already holds a slice of your employer and a large slice of its sector. Pouring diversification money into it means buying more of exactly the exposure you are trying to escape. The concentration follows you into the supposedly diversified holding.

Direct indexing fixes this because the account is built stock by stock. You instruct the manager to exclude your employer entirely, and usually to under-weight or exclude its sector, while still tracking the broad index as closely as possible by redistributing that weight across the remaining names. The diversification you deploy is now genuinely diversifying. This is a structural advantage an index fund cannot offer at all — a fund is one security with fixed contents, take it or leave it.

Worked example. The hidden double-exposure in a plain index fund

Suppose you work at a large-cap technology company, your employer stock is $300,000, and you are deploying a fresh $300,000 of diversification money. Assume your employer is roughly 6% of the S&P 500 by weight and its sector is roughly 30%.

Approach	Employer stock you hold	New employer exposure from the $300k	Total in your employer
Buy a plain S&P 500 fund	$300,000	$18,000 (6% of $300,000)	$318,000
Direct index, employer excluded	$300,000	$0	$300,000

The plain fund quietly adds $18,000 more of your employer through the index weighting, and far more than that in its broader sector, so your "diversification" leaves you more concentrated in the single name than when you started. The excluded direct-index sleeve adds zero new employer exposure and tilts away from the sector, so every dollar deployed reduces your concentration as a share of net worth. On a position this size the $18,000 is not catastrophic, but it runs in precisely the wrong direction, and it compounds every year you keep adding to the fund. The exclusion is not a luxury feature; for a concentrated holder it is the point.

Bottom line: A plain index fund silently adds more of your employer and its sector through the index weighting, working against your diversification, while a direct-indexing account excludes the name entirely and redistributes the weight — so the new money genuinely reduces your concentration instead of quietly rebuilding it.

The multi-year unwind: running the engine across the gain

Direct indexing earns its keep over a multi-year unwind, not a single transaction. A large, gain-heavy concentrated position is rarely sold in one year — doing so can push the gain into the 20% long-term bracket and stack NIIT on the whole amount. You stage the sales across several years, and you run the direct-indexing sleeve alongside so that each year's harvest offsets that year's realized gain.

The harvest is largest in the early years, when the sleeve is young and many lots sit near or below their purchase price, and it decays as the account appreciates and fewer names trade at a loss. That decay pattern happens to line up well with a staged unwind: you get the most loss-harvesting help in years one and two, exactly when you are doing the heaviest selling.

Worked example. Unwinding $120,000 of gain over four years

You hold employer stock with $120,000 of embedded long-term gain to unwind, and you fund a direct-indexing sleeve at the start. Your all-in long-term rate is 18.8% (15% plus NIIT). You stage $30,000 of gain per year for four years, and the sleeve harvests a declining amount each year as it appreciates: $12,000, $9,000, $6,000, then $3,000.

Year	Gain realized	Harvested losses	Net taxable gain	Tax at 18.8%	Tax without sleeve
1	$30,000	$12,000	$18,000	$3,384	$5,640
2	$30,000	$9,000	$21,000	$3,948	$5,640
3	$30,000	$6,000	$24,000	$4,512	$5,640
4	$30,000	$3,000	$27,000	$5,076	$5,640
Total	$120,000	$30,000	$90,000	$16,920	$22,560

Across the four years the sleeve harvests $30,000 of losses, which reduce the taxable gain from $120,000 to $90,000. Tax falls from $22,560 to $16,920 — a saving of $5,640, roughly one full year's worth of the original tax bill, deferred or erased by the harvesting engine. Notice the harvest shrinking year over year: that is the decay, and it is why this tool wants to be deployed early and against a large gain. The $5,640 saving is real money, but be honest about scale — on a $120,000 gain it is a meaningful trim, not a transformation. The sleeve's job is to take the edge off the tax, not to make the gain disappear.

Bottom line: Stage a large embedded gain across several years and run the direct-indexing sleeve alongside it, so each year's harvest offsets that year's realized gain — the harvest is biggest early and decays as the account appreciates, which lines up with doing your heaviest selling in years one and two and meaningfully trims, though does not erase, the total tax.

The fees, the decay, and when it is not worth it

This is the section the marketing materials skip. Direct indexing is not free, and there are years and situations where it actively costs you money. Being clear-eyed about this is what separates using the tool from being sold it.

The fee is the first cost. A direct-indexing SMA typically charges around 0.2% to 0.4% per year in management fees, against 0.03% to 0.10% for a broad index ETF — call it an extra 0.2 to 0.3 percentage points annually on the whole account balance. On a $250,000 sleeve, a 0.3% premium over a cheap ETF is $750 a year, paid whether or not the harvesting did anything for you that year. The second cost is decay: as the account appreciates, the fraction of lots sitting below their purchase price falls, so the annual harvest shrinks toward zero. A sleeve that harvested 4% to 5% of its value in year one might harvest 1% or less by year five. The third cost is complexity and tracking error — you hold a sample of the index, not the whole thing, so your return will diverge slightly from the benchmark in either direction.

The decision rule is simple: direct indexing is worth it when the after-tax value of the harvest exceeds the extra fee. That is true with room to spare in the early years of a large, gain-heavy unwind. It stops being true on a small, mature, fully appreciated account with no gains left to offset.

Worked example. The year it is not worth the fee

Suppose your unwind is essentially done. Your direct-indexing sleeve is now $200,000, it has appreciated for several years, and almost every lot is in the money. This year is a strong up market, so the manager finds only $2,000 of losses to harvest. You also have no large concentrated gains left to offset — you finished selling the employer stock last year — so the $2,000 of losses can only soak up $2,000 of ordinary income (within the $3,000 annual cap), saving tax at your 35% marginal ordinary rate.

Line	Amount
Harvested losses this year	$2,000
Tax saved (against ordinary income at 35%)	$700
Extra fee vs a cheap index ETF (0.3% of $200,000)	$600
Net benefit this year	$100

The harvest is barely covering its own fee. And this is a relatively generous year — the $700 assumes you can use all $2,000 against ordinary income at a high marginal rate. In a year with no harvestable losses at all, the math is purely −$600: you pay the premium and get nothing for it. Once the unwind is complete and the account has appreciated, the honest move is often to stop paying for the harvesting engine and transition the money into a plain low-cost index fund. The tool was built for the unwind; it is not automatically a permanent holding.

Bottom line: Direct indexing charges roughly 0.2% to 0.4% versus 0.03% to 0.10% for an ETF, and its harvest decays as the account appreciates — so on a small, mature, fully appreciated sleeve with no gains left to offset, the harvested-loss value can fall below the extra fee, at which point the tool costs you money and you should move to a plain index fund.

How direct indexing compares to the other tools

Direct indexing is one option on the diversification menu, and it is not always the right one. Here is how it sits against the alternatives for a concentrated employer-stock holder, each of which has its own full guide.

Tool	What it does	Triggers tax now?	Best for
Sell at vest, buy index fund	Diversifies immediately, no embedded gain on fresh vests	Only ordinary income at vest	Fresh vests, simplest case
Direct indexing	Harvests losses to offset gains, excludes employer	Yes, but offset by harvest	Large gain-heavy position, multi-year unwind
Completion portfolio	Shapes the rest of your holdings to balance a kept position	No	Holders who want to keep the stock
Exchange fund	Pools the position into a diversified partnership, tax-deferred	No (deferred ~7 years)	Very large, very low-basis, long-held stakes
Charitable gifting / DAF	Donates appreciated shares, deduction plus no gain	No (gain avoided)	Charitably inclined holders

Direct indexing's distinctive feature is that it is the only tool that actively manufactures tax losses to pay for the unwind. A completion portfolio does not reduce your single-name risk; an exchange fund defers tax rather than offsetting it and locks you up for years; charitable gifting only helps to the extent you were giving anyway. Direct indexing is the tool when you genuinely want out of the position, you have real embedded gains, and you are unwinding over several years. The tools are not exclusive — you might run a direct-indexing sleeve for the bulk of the unwind and gift a slice of the most-appreciated shares to a donor-advised fund in the same year.

Bottom line: Direct indexing is the only diversification tool that manufactures tax losses to offset the gains on selling your employer stock, which makes it the natural choice for a large gain-heavy position unwound over years — it pairs well with charitable gifting and is a different job from completion portfolios, which keep the position, and exchange funds, which defer rather than offset.

Common mistakes

• Buying a plain index fund without excluding your employer. The fund quietly rebuilds your concentration through the index weight; the whole point of the sleeve is to exclude the name.
• Triggering the wash-sale rule across accounts. A loss harvested in the sleeve is disallowed if you (or your spouse, or your IRA) buy the same security within 30 days — coordinate the sleeve with everything else you hold.
• Expecting a constant harvest. The harvest is biggest early and decays as the account appreciates; budgeting for year-one harvest rates in year five will disappoint you.
• Keeping the sleeve running after the unwind is done. Once the gains are gone and the account has appreciated, the fee can exceed the harvest — transition to a cheap index fund.
• Using a tiny account. On a small sleeve the fixed-ish fee swamps a modest harvest; the tool wants scale and a large offsetting gain to be worth the complexity.
• Forgetting NIIT in the math. Above $200,000 single / $250,000 married MAGI, add 3.8% to your capital gains rate when you model the value of each harvested dollar.

The closing read

Direct indexing is a precision tool for a specific job: unwinding a large, gain-heavy employer-stock position over several years without the tax bill swallowing the diversification. It works because it turns the diversification sleeve into a loss-harvesting engine — the same money you deploy to replace your employer stock manufactures the losses that pay the tax on selling it, and it excludes the employer so the new money genuinely diversifies. The honest caveats are that the fee is real, the harvest decays as the account grows, and on a small or fully appreciated sleeve the tool can cost more than it saves. Use it when the position is big, the embedded gain is large, and the unwind spans years — and have the discipline to step down to a plain index fund once the engine has run out of fuel. The tax tail should never wag the diversification dog, but when you can make the tail pay for the dog, you take it.

Cross-references

• What to do with vested RSUs: the diversification playbook
• Tax-loss harvesting and the wash-sale rule: a complete guide
• The completion portfolio: building around employer stock
• Exchange funds and swap funds for concentrated stock
• How much employer stock is too much?
• Rule 10b5-1 plans for selling company stock
• Donor-advised funds for equity compensation
• The RSU tax-saving playbook: harvesting, NUA, and DAFs
• US residents with US RSUs: the complete guide
• State tax optimization for US RSU holders
• Should you sell RSUs at vest or hold?

Critical disclaimer: this article reflects US federal tax rules and capital-markets practice as of June 2026 and is general information, not personalised advice. Capital gains rates, NIIT thresholds, wash-sale rules, and the fees and availability of direct-indexing accounts can change and depend on your specific facts, state of residence, account size, and basis. Direct indexing involves tracking error and is not suitable for every investor. This does not substitute for advice from a licensed CPA or CFP. Nothing here is a recommendation to buy or sell any specific security.