Case study
Selling a concentrated stock position: what diversification really costs
Maybe it's your employer's stock. Maybe it's a company you bought early that took off. Either way, a big slice of your net worth is now riding on one ticker, and selling means a big tax bill. What does diversifying actually cost?
This case is different from the others on this site. With a single stock, the question isn't only "which earns more?" It's also "what happens if this one company has a bad decade?" The math can't answer the second question, but it can tell you exactly what you're paying to stop worrying about it.
The setup
- $300,000 in one company's stock, bought for $60,000, a $240,000 long-term gain
- 20% all-in tax on the gain
- The alternative is a broad index fund expected to return 7% a year
- A 15-year horizon
Selling costs $48,000 in tax, leaving $252,000 to diversify.
If the stock is expected to do as well as the market
Suppose you think the stock's expected return is the same 7% as the index. Then switching just costs you the tax and the growth it would have earned:
After 15 years and all taxes, keeping the stock leaves about $674,000 and diversifying leaves about $607,000. If both earn exactly 7%, diversifying costs you about $67,500.
That's the price of the insurance. But it rests on a big "if": that one company is as likely to deliver 7% as the whole market.
Flip the question
A more useful way to frame it: how much would the stock have to return for keeping it to beat selling? In this example, the answer is 6.24% a year. If you expect the stock to earn less than about 6.24%, selling wins even after the tax bill.
| If the stock earns | Selling leaves you, after 15 years |
|---|---|
| 4% a year | +$162,394 |
| 5% a year | +$95,678 |
| 6% a year | +$19,447 |
| 7% a year (same as index) | −$67,547 |
| 9% a year | −$279,575 |
Put another way: keeping the stock is a bet that it will earn within about 0.76 points a year of the market, or better. That's the hurdle the stock has to clear, not the index.
Why "expected return" is doing a lot of work
The table assumes the stock earns a steady return. Single stocks don't behave that way. A landmark study by finance professor Hendrik Bessembinder looked at U.S. stocks from 1926 to 2016 and found that 58% of them had lifetime returns below one-month Treasury bills. Just over 4% of companies accounted for all of the stock market's net wealth creation above T-bills.
Your stock may well be one of the winners. It may already have been. But the range of outcomes for one company is far wider than for an index fund, and the downside cases (a 50% drop, a bankruptcy) aren't captured by any single "expected return." That risk is the real reason people diversify. The hurdle just tells you how much you're paying for the protection.
What changes the math
- Time horizon. Over 5 years the stock needs to earn at least 6.00% for keeping to win; over 30 years, 6.47%. The longer you hold, the less the upfront tax matters.
- Holding for heirs. If you'd hold until death and your heirs get a step-up in basis, the gain is never taxed on the keep path. Selling becomes more expensive: the stock only needs to earn about 5.76% for keeping to win.
Options people use instead of selling it all at once
These are approaches you'll often see discussed. Which, if any, fits depends on your situation, so they're worth raising with a tax professional:
- Selling in stages across tax years, which can keep more of the gain in a lower capital-gains bracket
- Selling lots with the highest cost basis first, if you bought at different times
- Giving appreciated shares to charity instead of cash, which avoids the gain entirely if you were going to give anyway
- Stopping new purchases of the stock, so the concentration shrinks as everything else grows
Try it with your numbers
Enter your position and the return you honestly expect from the stock, then compare it with the hurdle.
Run this example in the calculator
Figures assume a single lot, a flat 20% rate, steady annual returns, and both paths fully sold at the end of the period except in the step-up case. They don't capture the risk of a single stock, which is the main reason to diversify. These are illustrations of the arithmetic, not a recommendation to buy, sell or hold any security.