Direct indexing sounds sophisticated. It is not. You buy the 300 or 500 stocks in an index individually instead of buying one ETF that holds them. That is the entire trick.
Why would a doctor do that? Because you are taxed like few other professionals. The data shows the real product being sold here is not stock picking. It is tax management. And for high-income physicians, taxes are the largest drag on long-term taxable compounding.
This is the only honest way to evaluate direct indexing: does the after-tax benefit beat the after-fee cost?
This content is for educational purposes only and is not financial advice, not legal advice, and not tax advice. Figures vary based on individual circumstances and market conditions, so you should consult a qualified professional before making decisions.
Direct indexing is simple in concept. Instead of owning VTI or SCHB, you own 400 individual stocks that replicate the exposure of that index. Your portfolio looks like the index, behaves like the index, but you have direct ownership of each loss and each gain.
For physicians, that detail matters more than it does for most investors. Three numbers define you.
First, marginal tax rate. An attending in the 35% or 37% federal bracket plus state tax plus NIIT lives in a very different tax world than a median household. The value of a harvested dollar of loss scales directly with that rate.
Second, taxable account size. I have worked with surgeons who keep $800k to $2M in taxable brokerages because 401(k), 403(b), and backdoor Roths are full. That is where direct indexing can live. It is useless inside a qualified retirement account. No taxes to harvest.
Third, time poverty. You cannot manually monitor 400 positions for wash sales while covering call. Automation is the pitch.
The data shows the core promise is narrow. Direct indexing does not systematically outperform the market on a pre-tax basis. The tracking engine is designed to match the index, not beat it. The only persistent source of excess return, if any, comes from after-tax alpha generated by realizing losses.
Not magic. Just tax accounting.
What the Data Shows About Tax-Loss Harvesting
The data shows tax-loss harvesting works best when two conditions meet: volatility and a high tax rate applied to realized gains.
In a flat up market, fewer losses appear. In a volatile or down market, more do. Studies of direct indexing historically show the first few years generate the most harvestable losses. After that, your cost basis trends down and the pool of loss opportunities shrinks unless you add new cash.
What drives the benefit? Four variables.
- Your marginal rate. A $10,000 realized loss offsetting short-term gains saves $3,700 at 37% federal. At 22%, that same loss saves $2,200. Same action, 68% more benefit for the higher bracket.
- What you get to offset. Losses first offset gains of the same type. Leftover net losses can offset $3,000 of ordinary income per year, with the rest carried forward. For a physician selling a practice asset, ESPP shares, or concentrated stock, having a bank of losses matters.
- Short-term vs. long-term. Short-term losses offsetting short-term gains are more valuable than long-term losses offsetting long-term gains. The rate spread is the source of the math.
- Reinvestment. The tax savings only compounds if you reinvest it. Otherwise you just deferred taxes.
Outcomes are not guaranteed. At all. The advantage is opportunistic. And it must be measured net of fees, bid-ask spreads, and tracking error against the benchmark you would have owned anyway.
That chart illustrates the point. The gross benefit range expands with the tax bracket. The data shows the high-bracket physician with a large taxable balance is the only profile where 1.0% to 2.0% pre-fee tax alpha is even plausible in a typical year. For a resident at 22%, the math rarely works. Fees eat it alive.
How Direct Indexing Works in Practice
Here is how the machine actually runs. No mystery.
Your provider buys 250-500 stocks to track, say, the S&P 500 or a broad US equity index. The portfolio aims for a tracking error under 1%. Not zero. Under.
Every day, the system scans unrealized losses. If NVDA is down 7% from your lot basis, it flags it. If that loss exceeds a threshold, it sells NVDA and immediately buys a substitute security that maintains similar factor and sector exposure, for example, another semiconductor name or an ETF proxy if allowed by your mandate.
You stay invested. You realize a loss. You preserve exposure.
Why automation matters: wash-sale rules. If you sell at a loss and buy the same or substantially identical security within 30 days before or after, the IRS disallows the loss. A good direct indexing engine navigates substitutions across hundreds of lots. A human doing this in Excel will make a mistake. I have seen it happen. A physician client sold and rebought the same stock in two accounts and blew up six months of harvested losses.
Rebalancing is the other piece. The index changes. You add cash. Dividends hit. The portfolio must drift back toward the target without recreating gains.
Complexity for tax alpha. That is the tradeoff. Simple ETF investing gives you low fees and near-zero tracking error. Direct indexing gives you more tax flexibility and more moving parts.
Who Benefits Most: Physician-Specific Use Cases
The data shows fit is binary. You either have the taxable balance and tax rate to justify it, or you do not.
Most likely to benefit:
- Peak-earning attendings with $500k+ taxable. You are already paying 35%+ federal, you have annual restricted stock sales from a health system, and you consistently realize gains. Direct indexing gives you a buffer.
- Dual-physician households. Two high incomes, large joint taxable account, high annual cash additions that create fresh loss lots. New cash is fuel for harvesting.
- Doctors near a liquidity event. Selling a private practice interest, surgical center shares, or appreciated company stock. A harvested loss bank reduces that one-time tax hit.
Poor fit:
- Residents and fellows. Low current taxable income, small taxable balances, most savings in Roth and 403(b). The gross benefit is pennies. The fees are dollars.
- Physicians with 90%+ assets in retirement accounts. Nothing to harvest. Do not pay for a feature you cannot use.
- Investors who hate tax forms. Direct indexing can generate a 100+ page 1099-B with hundreds of trades.
The decision lens is quantitative. Larger taxable balance multiplied by higher marginal rate equals higher expected after-tax value. Smaller balance or lower rate equals lower value. No exceptions.
Costs, Risks, and Hidden Frictions
I will be direct. The costs are underestimated and the benefits are oversold in marketing decks.
Fee drag. Direct indexing platforms typically charge 0.20% to 0.40% on top of underlying trading costs. Some advisors bundle it into a 0.80% to 1.00% AUM fee. That fee is certain. The tax benefit is uncertain. You pay the fee in up markets and down markets. You get the tax benefit mostly in volatile or down markets.
Tracking error. You will not perfectly track the index. Sometimes you outperform, sometimes you underperform by 0.30% to 1.50% in a given year due to substitution choices. Over ten years that compounds. The data shows tracking error is the single largest risk to net benefit claims.
Deferred gain buildup. Every loss you harvest today lowers your cost basis. You own the same market exposure but with a lower basis. If you ever liquidate that taxable account, you will have a larger gain than if you had just held the ETF. Tax-loss harvesting is generally tax deferral, not permanent tax elimination, unless you get a step-up at death or donate appreciated shares.
Operational risks. Wash sales across accounts, misreporting, account transfers that force realization, and the sheer complexity of tracking hundreds of lots. For a busy physician, the question is whether you want this complexity or whether a simpler approach of broad ETFs plus annual tactical harvesting does 80% of the job at 10% of the hassle.
Look at that breakdown. The data shows a 2.2% gross benefit story can become 1.45% net after realistic frictions. Still positive. But not 2.2%. And in low-volatility years, that gross might be 0.60% and the net might be 0.10%. Fees do not care about your market environment.
Action Steps for Doctors: A Data-Driven Decision Framework
Do not start with the product. Start with your own numbers.
Use this sequence. It takes 15 minutes and prevents expensive mistakes.
- Assess taxable assets. Below $250k in taxable? Stop. Broad ETFs win on simplicity. Above $500k and adding cash annually? Continue.
- Confirm marginal tax rate. Below 32% federal? Benefit is thin. At 35% to 37% plus state and NIIT? Benefit is materially larger.
- Define time horizon. Under 10 years until you need the taxable account? Limited time to compound deferred taxes. Over 15 to 20 years? More time for reinvested tax savings to work.
- Compare the alternative. What does low-cost ETF indexing plus one manual tax-loss harvest per year in your taxable account get you? Many physicians capture 40% to 60% of the benefit with none of the fees.
- Demand net numbers. Ask any direct indexing provider for after-fee, after-tax projections under three scenarios: low volatility, average volatility, high volatility. Not marketing charts. Actual modeled numbers with fees and tracking error included. If they cannot provide it, walk away.
The right comparison is never pre-tax return. The data shows the only comparison that matters is net after-tax, after-fee return versus a simple ETF portfolio you would hold anyway.
Direct indexing is not good or bad. It is a tool. For high-income physicians with sizable taxable accounts, high realized gains, and a long horizon, the data shows it can add meaningful after-tax value. For everyone else, it is expensive complexity disguised as sophistication.
Measure net. And insist on data.
Key Takeaways
- The data shows the main value of direct indexing is tax-loss harvesting, not market-beating stock selection.
- Physicians with high taxable income and sizable brokerage balances are the most likely to see meaningful after-tax value.
- The right comparison is not gross return; it is net after-tax return after fees, tracking error, and implementation friction.
