Direct Indexing as a Concentration Offset: Harvesting Losses Against Employer-Stock Gains

A post-IPO employee holds $4.2 million of employer stock with $3.1 million of embedded long-term gain. Selling it all in one year produces $737,800 of federal tax at 23.8% plus whatever the state adds. Selling it over five years spreads the income but loses investment time for the proceeds. Both approaches leave significant tax on the table.

Direct indexing offers a third path. The employee moves $2 million of existing diversified assets into a direct-indexing separately managed account that tracks the S&P 500 via direct ownership of the underlying stocks. Over time, individual positions in the account go down even when the index goes up. The account harvests those losses. In a typical year, a direct-indexing SMA in an equity-bull-market environment produces realized losses equal to 1.5% to 3% of account value; in volatile or down years, losses can exceed 5%.

Those realized losses go on Schedule D against the employer-stock gains. The employee now sells employer stock at a pace that matches the loss production of the direct-indexing sleeve, diversifying the concentrated position without the full tax drag.

How direct indexing produces losses when the index is flat or up

An index like the S&P 500 is a blended average. Even in years when the index rises 15%, roughly 30-40% of the individual constituents have loss positions at some point during the year. A direct-indexing account that holds those constituents individually can sell the losers to harvest the losses while holding the winners.

The account is not trying to beat the index. It is trying to track the index after fees while generating as many realized losses as possible. Every time a position falls below its cost basis by a meaningful amount (typically 5-10%), the manager sells, buys a similar but not substantially identical replacement (often a sector ETF or a different stock in the same sector) for 31 days to avoid the wash-sale rule, then either buys back the original position or keeps the replacement depending on portfolio drift.

The replacement cycle matters. IRC §1091 wash-sale rules disallow a loss if substantially identical stock is purchased within 30 days before or after the sale. Direct-indexing managers typically use sector ETF replacements or correlated-but-distinct individual stocks and respect the 31-day window.

Loss yield: what to expect

Loss yield varies by market regime. In the benchmarks direct-indexing sponsors publish:

• Normal rising-market years: 1.5% to 3% of account value in realized losses.
• Flat or moderately negative years: 3% to 5%.
• Bear markets and volatility spikes: 5% to 10%+.
• Late-cycle bull markets: lower, often under 1.5% because fewer positions are below cost basis.

Loss yield also decays as the account ages. A new account has many positions still near their purchase price, producing many harvesting opportunities. A five-year-old account has many positions deeply appreciated and no longer close to cost basis, so fewer losses are available to harvest.

For the concentrated-stock offset strategy, this matters. The loss engine is strongest in years one through three, then tapers. Plans that sell the concentrated position over a multi-year window should match the pace to the loss availability, front-loading sales in the higher-loss-yield years.

Offsetting employer-stock gains

Long-term capital losses from direct indexing offset long-term capital gains from employer stock directly on Schedule D. Short-term losses (from positions held less than a year) offset short-term gains first, then long-term gains if short-term is exhausted.

Mechanically: realize $90,000 of long-term losses in the direct-indexing account, realize $90,000 of long-term gain on employer stock, and the net capital gain is zero for the year. The wash-sale rule does not apply across accounts in this way because the positions are different securities.

Unused losses carry forward indefinitely under IRC §1212. Losses above the current year’s gains can be used in future years against future gains.

The capital-loss limitation ($3,000 against ordinary income per year for individuals) still applies, but when the strategy is working the account rarely has excess losses after matching to employer-stock gains. The goal is to size the gain-realization to roughly match the loss production.

Comparison: direct indexing vs ETF and mutual fund alternatives

Vehicle	Typical annual loss yield	Transparency	Fees	Offset against employer-stock gains
Direct-indexing SMA	1.5-5% of account value	Full, position-level	0.20-0.45%	Yes, direct offset
Broad-market ETF (VTI, SPY)	0% (fund absorbs)	None	0.03-0.09%	No (can only realize on sale of ETF)
Mutual fund	0% (can distribute gains to you)	None	0.20-1.00%	No, and can add gains
Individual stock picking	Variable	Full	Trading costs	Yes if timed, unpredictable

The structural advantage of direct indexing over ETFs is that ETF investors cannot realize losses on internal positions; they can only realize gains or losses on the ETF wrapper itself. Direct indexing breaks open the wrapper and makes every underlying position a separate tax lot.

Fees and break-even analysis

Direct-indexing fees run 0.20% to 0.45% per year, compared to 0.03% to 0.10% on broad-market ETFs. The fee drag is roughly 0.15% to 0.40% per year.

Break-even against the ETF alternative: if the employee saves 23.8% federal plus 13.3% state (37.1% combined) on $60,000 of loss harvested from a $2M account, that’s $22,260 of tax saved. The incremental fee over the ETF is about $6,000 per year ($2M × 0.30%). Net benefit roughly $16,000 per year in a typical year, or 0.8% of the account value.

For accounts in low-tax states (Florida, Texas, Washington for income), the federal-only savings are 23.8% on $60,000 = $14,280, with $6,000 of incremental fees. Net $8,280, or 0.4%. Still positive but narrower.

The strategy is most compelling for high-tax-state residents with large embedded gains in employer stock and meaningful outside diversification assets to convert to direct indexing.

Risks and failure modes

Tracking error. Direct-indexing accounts track the index imperfectly. Over long periods tracking should be close, but in any given year the account can under- or over-perform by 0.5% to 1.5%.

Complexity at tax time. Every position sale is a separate tax lot. A direct-indexing account can produce 100+ transactions per year. The manager provides a Form 1099-B and supporting detail, but the preparer needs to be comfortable with the volume.

Loss exhaustion. After five to ten years, the account’s embedded gains are large and harvest opportunities are few. At that point the account becomes a concentrated position of its own (in the S&P 500 mostly). Some investors transition at that stage to an ETF of the same index or gift the positions to charity.

Wash-sale cross-account risk. If the employee owns the same stocks individually in another account (including a 401(k)), wash-sale rules can disallow direct-indexing losses. Managers typically ask for a list of outside holdings to avoid conflicts.

Frequently asked

Can I open a direct-indexing account in an IRA? Yes, but there is no tax benefit. Losses in a retirement account are not realized for tax purposes. Direct indexing is a taxable-account strategy.

What is the minimum account size? $100,000 to $250,000 for most retail-friendly sponsors (Schwab, Fidelity, Vanguard via acquisitions), $1M+ for premium sponsors (Parametric, Aperio, Eaton Vance).

Can I fund a direct-indexing account with my employer stock? Some sponsors accept concentrated stock as funding and build the diversified position around it. This is sometimes called a “completion portfolio” approach.

Does the strategy work against short-term gains on RSUs I just sold? Yes. Short-term losses offset short-term gains first, then long-term. The account produces both types depending on holding period.

How long do I need to stay in direct indexing to justify the setup? Break-even at typical fees and tax rates is roughly two to three years of meaningful loss production. Plan for a minimum five-year horizon.

Next step

Model your gain-realization schedule for employer stock over the next three to five years. For each year, identify how much gain you want to realize. Ask a direct-indexing sponsor for a historical loss-yield analysis on an account sized to match that gain pace. If the numbers work, fund the account at least 60 days before the first employer-stock sale so the loss engine has time to produce positions available for harvest.