The Disposition Effect: Why Investors Sell Winners and Hold On to Losers

The price you paid for an investment means nothing to the market, yet it often steers which holdings people sell. The disposition effect is that pull: the tendency of investors to sell investments that have risen since they bought them too soon and to hold on to investments that have fallen for too long. A 2014 investor bulletin from the Securities and Exchange Commission (SEC), the US market regulator, highlights a 2010 Library of Congress report that lists the disposition effect among the investing behaviors that can undermine investment performance. The bulletin adds that winners sold often keep outperforming the losers still held.¹

The hinge is the price you paid. Two holdings can look equally promising today, yet the one showing a gain feels easy to sell and the one showing a loss feels like money the market still owes you. One question cuts through it: if you held cash instead of this investment today, would you buy it at the current price? We offer it as a rule of thumb that no study has tested; the fixes that did work in experiments come further down. For the wider picture of how beliefs shape money choices, see what a money mindset is and what psychologists actually measure.

A selling habit, named by two economists in 1985

The disposition effect is a pattern in selling, not buying: given a choice, investors tend to cash in holdings that are up and keep holdings that are down, measured against what they paid. The economists Hersh Shefrin and Meir Statman gave it its name in 1985, in a Journal of Finance article titled after the habit itself.²

Shefrin and Statman started from a feature of Daniel Kahneman and Amos Tversky’s approach to choice under uncertainty: an aversion to realizing losses. They placed that aversion in a wider framework with four other ingredients: mental accounting, regret aversion, self-control and taxes.² Why a loss stings more than an equal gain pleases is explained in our piece on prospect theory and how people weigh wins against losses; the disposition effect is what that lopsided weighting looks like at the moment of selling.

Here is how it plays out, in an illustrative case. You own two shares bought on the same day: one is now up by about a third, the other down by about a third. A car repair means you need cash. Selling the first feels like banking a win. Selling the second feels like making a loss official, so you keep it “until it comes back”.

The market does not know what you paid. The price you paid is a fact about the past; the choice in front of you is which holding you would rather own from here.

How the disposition effect shows up in real trading records

Terrance Odean’s 1998 study of trading at a large US discount brokerage, the best-known test of the disposition effect, found investors far readier to sell winners than losers, and the winners they sold went on to beat the losers they kept.³

The study

Moderate evidence

Odean (1998): what 10,000 brokerage accounts sold, and what they kept

Across the year, a stock that was up was more than 50% more likely to be sold, from day to day, than a stock that was down. The pattern survived checks for portfolio rebalancing and for the higher trading costs of low-priced shares, and it reversed only in December, when tax-motivated selling of losers took over. Over the following year, the winners investors sold beat a market index by about 3.4 percentage points more than the losers they held.³

The practical meaning: the habit did not just feel cautious, it pointed money the wrong way, because the stocks people held on to in hope of a rebound tended to lag. Two caveats matter. These were one broker’s self-directed customers more than 30 years ago, and the comparison sets stock returns against an index rather than tracking each investor’s own result after costs; Odean does argue that in taxable accounts the habit also lowers after-tax returns.³

The pattern is not only American. A 2001 study that tracked the daily trades of individuals and institutions in Finland’s stock market also found investors reluctant to realize losses, alongside tax-loss selling.⁴ The takeaway for an ordinary portfolio is modest but real: if you notice that your sales are mostly of holdings in profit, it is worth asking whether that is a plan or a reflex.

Why a losing investment is so hard to sell

No single explanation of the disposition effect has won.⁵ The leading accounts point to the purchase price acting as a reference point, to a reluctance to admit a mistake, and to a belief that losers will bounce back, and each has some support from experiments or trading data.

The reference-point account comes from prospect theory. Measured from the purchase price, Odean explains, a stock that has risen sits where people tend to play safe and lock in a gain, while one that has fallen sits where people become willing to gamble on getting back to even. He also notes that the two motives blur: an investor who will not sell at a loss may persuade himself the stock will bounce back rather than admit he cannot accept the loss.³

Investors who sell winners and hold losers because they expect the losers to outperform the winners in the future are, on average, mistaken.

If the purchase price drives the habit, making it less visible should help, and in one lab test it did. In a 2014 Caltech experiment by Cary Frydman and Antonio Rangel, the disposition effect was about a quarter smaller when the trading screen stopped displaying what each stock had cost, though it did not disappear.⁵

1. Purchase price: the reference point studies usually measure gains and losses from
2. Above it, a winner: tempting to sell and lock in the gain
3. Below it, a loser: tempting to hold and wait to get back to even
4. From here: what each holding is likely to do next is what the decision is about; in Odean’s data, sold winners went on to beat held losers

A second account is about self-image: selling at a loss means admitting that buying was a mistake. In brokerage data and an experiment, Tom Chang, David Solomon and Mark Westerfield found the disposition effect in individual stocks but a reverse pattern in mutual funds, where investors realized losses more readily than gains. Their explanation is that delegating lets an investor blame the fund manager instead of themselves.⁶ Picture a bad month in your account: on that theory, the stock you chose yourself reads as a verdict on your judgment, while a fund’s fall can be pinned on whoever runs it.

Up since you boughtDown since you boughtA stock you pickedA fund someone else runs

A stock you picked, Up since you bought: Sold readilythe gain feels like a win

A stock you picked, Down since you bought: Held on toselling means admitting a mistake

A fund someone else runs, Up since you bought: Kept longerthe pattern reverses

A fund someone else runs, Down since you bought: Dropped readilythe manager takes the blame

Not everyone accepts that investors simply hate realizing losses. Itzhak Ben-David and David Hirshleifer found that individual investors were more likely to sell big winners and big losers than small ones, with little sign of a jump in selling at break-even. They conclude that a simple preference for selling at a gain does not drive the pattern, and that changing beliefs about a stock could explain it.⁷

Whatever the mix, the explanations share a weak spot: the decision to sell is made in the moment, with the purchase price on screen and your judgment on trial. The fixes worth trying therefore work on that moment, either by settling the decision before it arrives or by taking the price off the screen.

What reduces the disposition effect in experiments

The best evidence on fixes comes from lab experiments, and two approaches have worked: committing in advance to automatic sales, and hiding the purchase price. In a 2017 experiment by Urs Fischbacher, Gerson Hoffmann and Simeon Schudy, student investors who could set automatic sell limits in advance showed a smaller disposition effect, while reminders of their own plans did not help.⁸

The authors’ reading is that a loss is easier to accept before it exists. Once the price has fallen, the calm plan gets renegotiated; an automatic sale removes the renegotiation. Their caveat applies to you too: this was a small lab market with students and simple assets, not a test of stop-loss orders in real markets.⁸

Outside the lab, the nearest equivalent is a broker’s automatic sell order, such as a stop order. That is a description, not a recommendation: the SEC’s Investor.gov (US, checked 2026) warns that a short-term price swing can trigger such a sale, and the price you get may differ from the one you set.⁹ A sell rule you only write down is closer to the reminder condition, which did not help. We found no randomized test with real investors’ money, so what carries over is the timing: think a fall through while the loss is still hypothetical.

In a 2006 study of trading records from a major discount brokerage, Ravi Dhar and Ning Zhu found that wealthier investors, those in professional occupations and those who traded more often showed a smaller disposition effect. The study shows an association; it does not show that experience cures the habit.¹⁰

One caution applies to the lab fixes above. A meta-analysismeta-analysis: A study that combines the results of earlier studies on the same question into one overall estimate. Pooling makes the estimate more precise, but it cannot repair the studies it pools: a meta-analysis of surveys is still survey evidence.Full entry in the glossary of disposition effect experiments by Stephen Cheung, a working paper revised in 2025 and not peer reviewed, found stronger signs of selective reporting for tests of fixes than for the basic pattern, so read the size of any lab benefit with some caution.¹¹

Fund investors are not immune, though the pattern can run the other way. In Chang’s team’s brokerage data, mutual fund investors were readier to sell funds at a loss than at a gain.⁶ The study did not look at index funds on their own, so it says nothing specific about people who hold low-cost index funds. If you hold funds of any kind, a bad spell calls for the same fresh-cash question as a falling stock, whichever way your instinct pulls.

When keeping a losing investment can make sense

Sometimes the reasons behind the disposition effect are sound: rebalancing a portfolio, avoiding high trading costs and taxes can all justify selling a winner or keeping a loser. Odean tested the first two, and they did not explain the pattern he found. For taxable accounts under US tax rules, he adds, the tax logic usually points the other way: defer gains, realize losses.³

Tax rules show why. US rules, as the Internal Revenue Service describes them, let capital losses offset capital gains, let a limited amount of any excess loss reduce other income, and let the rest carry forward to later years (as of 2026).¹² The UK government’s guidance says a reported loss on a chargeable asset is deducted from gains made in the same tax year, and unused losses can be carried forward (as of 2026).¹³ Rules differ elsewhere, so check your own tax authority’s guidance.

Myth

It isn't a real loss until you sell.

Fact

The drop in value has already happened. Selling records it, which can matter for tax, and frees the money for something else; it does not change what the holding is worth today.

Consider a worked example with no numbers attached. Someone holds a fund bought for retirement that has fallen, and a single stock that has doubled and now makes up most of the account. Rebalancing may argue for selling some of the winner; a plan built around the fund may argue for keeping the loser. Both are reasons that would still apply if the purchase prices were erased.

Then put the fresh-cash question from the top of this page to each holding in turn.

When a sale needs personal advice

Whether to sell a particular holding depends on your taxes, goals and the rest of your portfolio, which general research cannot weigh for you. The lab fixes above describe what reduced a bias in experiments; they are not a trading strategy, and the right choice for you may differ.

• Before a large sale, or one with tax consequences, a regulated financial adviser or a tax professional can go through your figures with you; find out how they are paid before you start. Investor.gov, the SEC’s site for US investors, advises checking the background and license of any financial professional.¹⁴ The UK’s Financial Conduct Authority notes that almost all firms offering financial services must be authorised or registered by it, and shows how to check.¹⁵ Outside the US and UK, look up your own country’s financial regulator.
• At a routine review of your investments, look at which holdings you sold over the past year and whether the reasons would survive without the purchase price. Beginners setting up a first plan can start with how investing works and the steps regulators suggest.

The bottom line

What you paid for an investment is history to the market, and the disposition effect is what happens when that history still picks the sale. Records of real investors show the habit, and the losers people kept tended to lag the winners they sold. The fixes with the best lab support act before the moment of decision: automatic sales committed to in advance, and less attention to what you paid. Neither has been tested with real portfolios, and real automatic orders carry their own risks.