Portfolio Strategy Education 9 min read April 2026

Does Direct Indexing
Actually Beat ETFs?

An honest, data-driven analysis of when direct indexing outperforms ETFs, when it doesn't, and the specific conditions required for the tax benefit to justify the cost.

Embark Funds

Embark Funds Research

Investor Education Series · April 2026

01

The Honest Answer

Direct indexing beats ETFs after tax — not before tax

The answer depends entirely on what you're measuring. On a pre-tax basis, direct indexing does not beat ETFs — it slightly underperforms due to higher fees and tracking error. On an after-tax basis, direct indexing typically outperforms ETFs by 1–2% annually for investors in the top tax brackets.

"Direct indexing doesn't generate alpha from better stock picking. It generates alpha from better tax management. The investment returns are the same — the tax bill is different."

This distinction matters because many investors evaluate direct indexing by comparing pre-tax portfolio returns to the S&P 500. By that measure, direct indexing will always lag slightly (by the fee amount plus tracking error). The correct comparison is after-tax returns — and that's where direct indexing's advantage shows up.

02

Pre-Tax Performance

Why direct indexing slightly underperforms before taxes

Before tax adjustments, direct indexing portfolios typically trail their benchmark index by 0.3–0.7% annually. This shortfall comes from three sources:

1. Management fees (0.09–0.40%): Direct indexing costs more than a VOO ETF (0.03%). The fee gap of 0.06–0.37% is a direct drag on pre-tax returns.

2. Tracking error (0.5–2.0%): Tax-loss harvesting substitutions, stock exclusions, and rebalancing cause the portfolio to deviate from the index. Some of this deviation helps (the substitute outperforms), some hurts (the substitute underperforms). On average, tracking error is a wash to slightly negative pre-tax.

3. Cash drag: Some platforms hold small cash buffers for trading and fee payments. In a rising market, this uninvested cash creates a minor drag.

If you judge direct indexing solely on pre-tax performance, it will look worse than an ETF. This is expected and by design — the strategy trades pre-tax precision for after-tax optimization.

03

After-Tax Performance

Where direct indexing creates measurable outperformance

After-tax, direct indexing typically outperforms ETFs by 0.8–2.0% annually, depending on tax bracket, market conditions, and portfolio age. Here's the evidence:

Parametric Portfolio Associates (2020 study): Analyzed 10 years of direct indexing portfolios and found median after-tax outperformance of 1.08% annually over equivalent ETF strategies. Top-bracket investors in high-tax states saw benefits exceeding 1.5%.

Sosner, Krasner, & Pafka (Journal of Portfolio Management, 2020): Found that direct indexing generated 1.0–1.5% annual tax alpha for top-bracket investors over a 20-year simulation period. The benefit was most significant in volatile markets.

Wealthfront published data: Reports median tax-loss harvesting of $15,000–$20,000 annually per $1M in client portfolios — equivalent to 1.5–2.0% of portfolio value in harvested losses.

10-Year Comparison: $1M Portfolio in Top Tax Bracket

S&P 500 ETF (After-Tax)

$2,060,000

After-tax value at year 10

8% average annual return minus 0.03% fee, minus dividend tax drag of ~0.3%. No loss harvesting. Capital gains taxed at 23.8% upon sale.

Direct Indexing (After-Tax)

$2,230,000

After-tax value at year 10

Same 8% pre-tax return minus 0.35% fee, plus 1.2% average annual tax alpha. Harvested losses offset gains throughout the period. $170K after-tax improvement.

The Fee Question Resolved: On $1M, the incremental fee of ~$3,500/year ($1M × 0.35%) is more than offset by ~$12,000/year in tax savings ($1M × 1.2% tax alpha). The net annual benefit after fees is approximately $8,500.

Free Weekly Insights

Get This Analysis Every Week

Join investors at Google, Amazon & Apple who stay ahead with Embark’s weekly insights on concentrated stock strategies.

No spam. Unsubscribe anytime.

04

When DI Doesn't Win

Five scenarios where ETFs are the better choice

Low Tax Bracket (<24%)

At the 12–22% bracket, each harvested loss is worth less. A $10,000 loss saves $1,200–$2,200 instead of $3,700 at the 37% bracket. The fee premium of direct indexing may eat most or all of the benefit.

Tax-Advantaged Accounts

In 401(k)s, IRAs, and Roth IRAs, there's nothing to harvest. Gains are already tax-deferred or tax-free. Direct indexing in these accounts costs 0.09–0.40% for zero additional benefit. Use VOO at 0.03%.

Short Time Horizon (<3 years)

Tax alpha compounds over time. In 1–2 years, the total harvested losses may not exceed the cumulative fee difference. The break-even horizon is typically 2–3 years for top-bracket investors.

Small Portfolio (<$50K taxable)

On a $50K portfolio, even 2% tax alpha is only $1,000/year. After the fee premium ($175–$185/year at 0.35% vs 0.03%), the net benefit is $815–$825 — meaningful but modest. Below $25K, the math rarely works.

The bottom line: If you're in a tax bracket below 24%, hold primarily tax-advantaged accounts, have a short time horizon, or have under $50K in taxable investments, ETFs are the better choice. Direct indexing's fee premium isn't justified without the tax alpha to offset it.

05

Tracking Error Reality

How much does your portfolio actually deviate from the index?

Tracking error is the standard deviation of the difference between your portfolio's return and the benchmark's return. For direct indexing, typical annualized tracking error ranges from 0.5% to 2.0% — meaning in any given year, your portfolio may return 0.5–2.0% more or less than the S&P 500 (before tax adjustments).

Sources of tracking error in direct indexing:
Tax-loss harvesting substitutions — replacing Coca-Cola with PepsiCo creates sector alignment but not identical performance
Stock exclusions — removing NVIDIA from the S&P 500 means your portfolio misses NVIDIA's specific return (positive or negative)
Rebalancing timing — daily monitoring reduces this, but it's never zero
Fractional share rounding — small portfolios can't precisely match index weights
Cash holdings — uninvested cash during market moves creates tracking deviation

Is tracking error a problem? For most investors, 0.5–1.5% tracking error is acceptable when the after-tax benefit exceeds it. The key question is: does your after-tax return exceed the benchmark's after-tax return? If yes, the tracking error is the cost of that outperformance — and it's worth paying.

For investors who exclude their employer's stock (e.g., removing a 4–7% position), tracking error will be higher in years when that stock outperforms the rest of the index. This is intentional — you're trading tracking precision for income generation from your holdings.

06

Market Conditions

Direct indexing performs best in volatile markets

Tax-loss harvesting thrives on dispersion — when individual stocks within the index have widely varying returns. The more stocks that decline (even while the index rises), the more harvesting opportunities exist.

High-volatility years (best for DI): In 2022 (S&P 500 down 19.4%), nearly every stock in the index was harvestable at some point. Direct indexing portfolios generated massive losses that carried forward to offset future gains. In 2020 (sharp Q1 crash followed by recovery), the March selloff created a one-time harvesting bonanza.

Low-dispersion bull markets (worst for DI): When the index rises steadily with low volatility and most stocks participate, fewer individual stocks decline, and harvesting opportunities diminish. The "Magnificent Seven" concentration of 2023–2024 actually created dispersion (7 stocks massively up, many others flat or down), which benefited direct indexing.

Over a full market cycle (10+ years): The tax alpha averages out to the 1–2% range regardless of individual year conditions, because volatile years generate banked losses that carry forward into calm years. This is why a long time horizon matters.

07

FAQ

Performance questions about direct indexing vs ETFs

Does direct indexing outperform the S&P 500?

On a pre-tax basis, no — direct indexing slightly trails the S&P 500 by the fee amount (0.09–0.40%) plus tracking error. On an after-tax basis, yes — for investors in the 32%+ tax bracket with $100K+ in taxable accounts, direct indexing typically outperforms ETFs by 0.8–2.0% annually through tax-loss harvesting. The outperformance is entirely from tax management, not stock selection.

How much tracking error should I expect with direct indexing?

Typical annualized tracking error is 0.5–2.0% for a well-managed direct indexing portfolio. Excluding large-cap stocks (like your employer's stock) increases tracking error. Applying aggressive factor tilts increases it further. Most investors accept 1–1.5% tracking error in exchange for the after-tax benefit.

Is the tax benefit of direct indexing permanent?

Tax-loss harvesting defers taxes rather than eliminating them — your cost basis decreases with each harvest. However, with a stepped-up basis at death (under current law), unrealized gains are eliminated for heirs. And the time value of deferred taxes, plus the ability to offset short-term gains (37%) with long-term losses (23.8%), makes the benefit substantial even though it's technically a deferral.

Should I switch from ETFs to direct indexing?

If you have $100K+ in taxable accounts, are in the 32%+ bracket, and have a 5+ year horizon — yes. The transition can be managed tax-efficiently by transferring existing holdings in-kind (avoiding capital gains on the switch). For investors with concentrated stock alongside their diversified portfolio, pairing direct indexing with Embark's §721 SPV creates a comprehensive tax-optimized structure.

The Complete Strategy

Direct Indexing Wins on Tax Efficiency. Embark Wins on Concentrated Stock.

Direct indexing generates 1–2% after-tax alpha on diversified holdings. Embark's §721 SPV defers 100% of the gain on concentrated positions and generates income. Together, they form a complete tax-optimized portfolio.