The 'It's Not Gain Until I Sell' Fallacy
By VestedGrant Editorial · Updated · Tax year 2026 · 7 min read
Sanford Robinson Gifford, Cows in a Pond at Sunset, 1860 An unrealized gain is real wealth whose tax is deferred, not wealth that does not exist yet. Federal law taxes the gain only when the stock is sold, and for 2026 a long-term gain costs a high earner up to 23.8% (20% plus the 3.8% net investment income tax). The holding decision is therefore a trade between that tax cost and the risk of keeping one stock.
Facts checked November 21, 2026 · Figures for tax year 2026 · 9 sources
Key takeaways
- Federal tax law recognizes a capital gain only when the asset is sold, so an unrealized gain is a deferred tax, not a missing gain.
- For 2026 the top federal rate on a long-term gain is 23.8%: the 20% capital gains rate plus the 3.8% net investment income tax.
- In a hypothetical $2 million sale with $1 million of gain, holding instead of selling only wins after tax if the single stock keeps pace within about 0.7 percentage points a year of the diversified alternative.
- Basis steps up to fair market value at death under Section 1014, which makes holding until death a real tax strategy mainly for people with short horizons.
- Section 1031 like-kind exchanges apply only to real property, so they cannot defer gain on a stock sale.
On this page (show sections)
- Are unrealized gains real?
- Does holding to defer the tax beat selling now?
- How much underperformance cancels the deferral advantage?
- When does holding until death make tax sense?
- Why is “paper gains can evaporate” an argument for selling?
- What moves put the correct framing into practice?
- What changes when the gain is treated as stored wealth?
- Questions people ask next
- Sources
Consider a senior IC who holds 8,000 RSU shares trading at $275, which shows $2.2 million on the account screen. Asked about the position, the engineer says, “it’s not real until I sell it.” The framing sounds sensible: paper gains can evaporate, markets fluctuate, only cash in hand is certain. The framing is also financially incoherent and a common driver of under-diversification.
The fallacy has two components. First, treating an unrealized gain as “not real” makes it feel like there is nothing to lose by holding. Second, treating the sale as the moment the gain is created makes selling feel like “creating” a loss if the stock falls afterward. Both framings lead to the same result: never sell. The paper wealth is never converted to diversified wealth.
This article explains why the fallacy is wrong, what a correct framing looks like, and the mechanical moves that put the correct framing into practice. Dollar figures in the examples are hypothetical and use 2026 federal rates.
Are unrealized gains real?
Unrealized gains are real wealth; only the tax on them is deferred. Brokerage statements report them, net worth calculations count them, lenders accept them as collateral for securities-based loans and mortgages, and divorce settlements divide them.
Tax law treats the gain as unrealized for income tax purposes. As IRS Topic 409 puts it, a taxpayer has a capital gain when the asset is sold for more than its adjusted basis. That is a rule about when the tax is due, not a statement about whether the gain exists. In every economic sense the gain exists today; the tax bill is simply postponed.
A household balance sheet makes the point concrete. The brokerage account appears at current market value. Net worth is calculated on that value. Banks lend against that value. Decisions about buying a house or retiring depend on that value. Treating the same value as non-existent until it is sold is inconsistent with how the household actually uses the wealth.
Does holding to defer the tax beat selling now?
Tax deferral is worth something, but only if the concentrated stock keeps up with the diversified alternative. The usual justification for the fallacy runs: “If I sell, I pay tax. If I hold, I defer tax and earn more on the full amount.” A worked example shows how thin that advantage is.
Take a hypothetical $2 million position with a $1 million cost basis, so $1 million of long-term gain. Assume the gain is taxed at 23.8%: the 20% rate that applies for 2026 above $545,500 of taxable income for a single filer, plus the 3.8% net investment income tax that applies above $200,000 of modified AGI. Assume both paths earn 8% a year for 10 years and are fully sold at the end.
| Path | Tax today | Invested | Value after 10 years | Tax at year 10 | After-tax result |
|---|---|---|---|---|---|
| A: sell now, diversify | $238,000 | $1,762,000 | $3,804,026 | $486,002 | $3,318,024 |
| B: hold the single stock | $0 | $2,000,000 | $4,317,850 | $789,648 | $3,528,202 |
At equal returns, holding wins by about $210,000, because the deferred tax keeps compounding. That is the honest value of deferral.
How much underperformance cancels the deferral advantage?
A return shortfall of roughly 0.7 percentage points a year cancels the deferral advantage in the example above. If the single stock earns about 7.29% a year while the diversified portfolio earns 8%, path B ends at the same $3,318,024 after tax as path A.
That gap is small. A single stock does not need to collapse for selling to come out ahead; it only needs to trail a diversified portfolio modestly over a decade. The single stock also carries a much wider range of outcomes than the 8% average suggests, which the table cannot show: the same position can double or halve while a broad index moves far less.
The comparison also ignores what the diversified path buys besides return. A household that sold path A’s stock no longer has its net worth, its paycheck and its unvested equity all tied to one employer. Deferral is worth something. Whether it is worth enough depends on how much the holder expects the single stock to lag, and on how much the household can afford to lose if the forecast is wrong.
When does holding until death make tax sense?
Holding until death makes tax sense mainly for people with a short horizon, because Section 1014 resets the basis of inherited property to its fair market value at the date of death. The gain that built up during the holder’s life is never subject to income tax. The property is still counted in the estate for estate tax, which for 2026 applies only to estates above the $15,000,000 basic exclusion amount.
Consider a hypothetical 75-year-old with a $5 million concentrated position and a $500,000 basis. Selling realizes $4.5 million of long-term gain; at 23.8% the federal tax is about $1,071,000. Holding until death leaves heirs the full $5 million with a $5 million basis, so they owe no income tax on the past gain.
The calculation changes for a younger holder. A 40-year-old waiting for a basis step-up would keep the position exposed to single-stock risk for decades, and the expected cost of concentration grows with every year of exposure. For long horizons, concentration risk tends to outweigh the step-up; for very short horizons, the step-up tends to dominate.
Why is “paper gains can evaporate” an argument for selling?
The risk that paper gains evaporate is an argument for selling, because the gain can only evaporate while the stock is held. The holder who says “it’s not gain until I sell” has correctly spotted the risk and then drawn the opposite conclusion.
A hypothetical 40% drawdown shows the stakes. On the $2.2 million position from the opening, a 40% fall takes the value to $1.32 million, an $880,000 loss. Single stocks can and do fall that far, sometimes in a single year, without the broad market doing the same.
A correct mental model reads: “I have $2.2 million of wealth held in a risky single-stock form. I can convert it to a less risky diversified form by selling and reinvesting. The cost of conversion is the tax on the realized gain. The benefit is lower concentration risk.” Under that frame, the decision becomes a trade-off between the present cost of the tax and the expected cost of concentration, which is exactly the trade-off the deferral table measures.
What moves put the correct framing into practice?
The correct framing shows up in a handful of mechanical habits that some holders adopt to take the emotion out of the decision.
-
Counting the position at full current value in net worth. The position is wealth, not a hypothetical. Tracking it at market value keeps the size of the exposure visible.
-
Asking the cash question. “If this amount were cash today, would I buy this stock with it?” A no answer is a diversification signal, because holding is economically the same as buying at today’s price.
-
Scheduling sales on a fixed rule. A recurring sale, such as a set share of each vest or a fixed amount each quarter, converts paper to diversified wealth regardless of short-term psychology. Employees subject to trading windows often use a Rule 10b5-1 plan; for employees who are not directors or officers, the SEC requires a 30-day cooling-off period before the first trade under a new plan.
-
Treating the tax as a conversion fee. If a $500,000 sale is almost entirely long-term gain, the federal cost at 23.8% is about $119,000. That fee buys a reduction in single-stock risk, much like an insurance premium.
-
Measuring concentration as a percentage. The share of net worth held in one stock, not the dollar amount, describes how much a fall in that stock would change the household’s plans.
What changes when the gain is treated as stored wealth?
Treating the gain as stored wealth turns the question from “should I realize the gain?” into “is this the right place to store this much wealth?” The holder in the opening example does not have $2.2 million of “unrealized gain.” The holder has $2.2 million of wealth that happens to be stored in a single stock.
Under the storage frame, the gain and loss language disappears. The choice is not between “selling” and “keeping the gain.” The choice is between two storage forms, a single stock or a diversified portfolio, and the capital gains tax is the fee for moving between them.
Most people answer the storage question quickly when it is asked this way: few would choose to put half of their net worth into one company’s stock if they were starting from cash. The only thing preventing the diversifying trade is a framing that treats the current storage form as privileged. The tax rules give the current form a deferral advantage, which the table above measures, but they do not make the gain any less real.
Questions people ask next
What if I really do believe my company's stock will outperform?
A belief that one stock will outperform is a forecast, and holding on that basis is a deliberate bet rather than inertia. The honest version of that bet names a size: what share of net worth the holder is willing to keep in one stock if the forecast is wrong. The tax deferral does not change the size of the bet; it only delays when the gain is taxed.
Doesn't selling and reinvesting just trigger tax every time I rebalance?
Selling concentrated stock triggers capital gains tax once, and the reinvested proceeds start with a new cost basis equal to the purchase price. Later rebalancing of a diversified portfolio usually realizes smaller gains, because each holding has appreciated less on its own than a single stock that has grown for years. The first sale is the expensive one.
Can a 1031 exchange or a QSBS rollover defer the gain on my company stock?
A Section 1031 like-kind exchange applies only to real property, so it cannot defer gain on stock. Section 1045 defers gain only on qualified small business stock held more than six months and reinvested in other qualified small business stock within 60 days of the sale. Neither is a general way to defer tax on public company shares.
How does an exchange fund defer tax on a concentrated position?
An exchange fund is a partnership that accepts contributed stock and gives the contributor a share of a diversified pool. Section 721 lets the contribution happen without recognizing gain, provided the partnership is not treated as an investment company. Section 704 triggers gain if contributed property is distributed to another partner within 7 years, which is why these funds typically lock investors in for that long.
Is there a position size where holding is clearly fine?
No federal rule sets a safe concentration level. The useful measure is the position as a percentage of liquid net worth, because the same dollar loss means very different things to a household with $3 million and one with $30 million. A small position can wait for a convenient rebalancing moment; a dominant one carries more risk with every month of delay.
Sources
- Rev. Proc. 2025-32irs.gov
Supporting text
Heads of Household $66,200 $579,600 All Other Individuals $49,450 $545,500
- IRS Topic no. 559, Net investment income taxirs.gov
Supporting text
A 3.8 percent net investment income tax (NIIT) applies to individuals, estates, and trusts that have net investment income above applicable threshold amounts.
- 17 CFR 240.10b5-1 (eCFR)ecfr.gov
Supporting text
no purchases or sales occur until the expiration of a cooling-off period that is 30 days after the adoption of the contract, instruction or plan;
- 26 U.S.C. 1045(a)uscode.house.gov
Supporting text
the cost of any qualified small business stock purchased by the taxpayer during the 60-day period beginning on the date of such sale,
- IRS Topic no. 409, Capital gains and lossesirs.gov
Supporting text
You have a capital gain if you sell the asset for more than your adjusted basis. You have a capital loss if you sell the asset for less than your adjusted basis.
- 26 U.S.C. § 1014, Basis of property acquired from a decedentuscode.house.gov
Supporting text
the fair market value of the property at the date of the decedent's death
- 26 U.S.C. § 1031, Exchange of real property held for productive use or investmentuscode.house.gov
Supporting text
No gain or loss shall be recognized on the exchange of real property held for productive use in a trade or business or for investment
- 26 U.S.C. § 721, Nonrecognition of gain or loss on contributionuscode.house.gov
Supporting text
Subsection (a) shall not apply to gain realized on a transfer of property to a partnership which would be treated as an investment company
- 26 U.S.C. § 704, Partner's distributive shareuscode.house.gov
Supporting text
is distributed (directly or indirectly) by the partnership (other than to the contributing partner) within 7 years of being contributed
Facts checked November 21, 2026 · Tax year 2026
Find a fee-only advisor for your equity
Four questions. You see up to three fee-only advisors listed on Wellspent whose records include equity compensation, right here. Free, no obligation.
- What happens next: your matches appear on this page straight away. Advisors are not sent your details: you open a record and contact the advisor yourself.
- What is shared: your email stays with VestedGrant and is not passed to advisors. The advisor directory (Wellspent) receives your state and topic; an AI model ranks the matches from your answers, without your email. Details in the privacy policy.
Finding fee-only advisors for your answers…
Three advisors picked for your situation
Each is a fee-only advisor listed on Wellspent whose record includes equity compensation. Open a record to see how they work and to contact them directly. No referral fees either way.
No advisor matched those details yet
Try again with "Other / outside the US" as the state, or browse equity-compensation advisors on Wellspent.