Investors Sell Winning Stocks Too Early and Hold Losers Too Long. There's a Name and Real Data Behind It
6 min read
Loading...
6 min read
One stock is up 30% and the other is down 30%. You need cash. Which one do you sell?
Most people sell the winner. It feels good to lock in a gain. Selling the loser feels like admitting a mistake.
That pull has a name: the disposition effect. It is the habit of selling winning stocks too early and holding losing stocks too long. This post shows what the research found, what it can cost, and one check you can run on your own portfolio.
The disposition effect is a pattern in how people trade. They sell stocks that went up more readily than stocks that went down.
Two economists, Hersh Shefrin and Meir Statman, gave it this name in a 1985 paper in the Journal of Finance. The title says it all: "The Disposition to Sell Winners Too Early and Ride Losers Too Long". They tied it to prospect theory, the idea that people feel losses more strongly than equal gains. (Disposition effect, Wikipedia; I could not open the 1985 paper, so this is a secondary summary.)
In plain words, there are two feelings at work:
Shefrin has described it in an interview in just these terms: investors hold losers to avoid regret and sell winners to feel pride. (University of Manitoba news, search summary only.)
Theory is one thing. Did real investors do it?
Economist Terrance Odean checked in a 1998 paper, "Are Investors Reluctant to Realize Their Losses?" He studied the complete trading records of about 10,000 accounts at a large US discount brokerage, from 1987 to 1993.
His main result: outside December, investors sold about 14.8% of their winning positions but only about 9.8% of their losing positions. That is roughly 1.5 times as often. (Wikipedia summary of the paper; I did not open the original, so treat the exact percentages as secondhand.)
December was different. In that month the gap flipped, probably because people sold losers to cut their US tax bills. The same summary gives 10.8% of winners and 12.8% of losers realized in December.
Here is the part that matters for your wallet.
If winners are the better stocks, selling them early should cost you. Odean checked this too. Several secondary sources report that the winners investors sold went on to beat the losers they kept by about 3.4 percentage points over the next year. (Evidence Investor and other summaries; I could not confirm the 3.4% figure from the paper itself, so read it as "roughly three points", not a precise number.)
This does not mean every winner keeps winning. It means that, on average, the pattern did not pay.
This is NetWort arithmetic with made-up numbers, not a real stock.
You own two stocks, Rp10 juta each, so Rp20 juta in total.
| Stock | Bought at | Now worth | Change |
|---|---|---|---|
| A | Rp10 juta | Rp13 juta | +30% |
| B | Rp10 juta | Rp7 juta | -30% |
You need Rp5 juta in cash. Selling A feels natural. You keep B, saying "I'll sell when it gets back to Rp10 juta".
Now look at what B needs. A stock that falls 30% must rise about 43% to get back to even (Rp7 juta to Rp10 juta). Our post on the maths of recovering from a loss explains why the climb is always steeper than the drop.
Next, take the research average as an illustration. If the stock you sold beats the one you kept by 3.4 points in a year, the gap on Rp13 juta is about Rp442.000. That is one trade, one year. Repeat it across many trades and many years, and it adds up.
Partly. Studies on the Indonesia Stock Exchange (IDX) give mixed results.
One study of IDX data from 2004 to 2017 found that stock splits weaken the disposition effect among individual investors. Another found no disposition effect among LQ45 stocks during the 2013 crisis. (Review of Behavioral Finance, 2021 and a Petra Christian University journal paper; both seen as abstracts only.) So the effect is not guaranteed in every market or period. It is a tendency to watch for, not a law.
One US-specific driver does not apply here. Odean's December flip came from tax. On the IDX, the final income tax on a stock sale is 0.1% of the sale value, and it applies whether you made a profit or a loss. (IKPI explainer on stock tax, summary only; check the DJP site for the current rule.) Our guide to the final tax on IDX stock sales covers it. So tax is not a good reason to keep a loser.
None of this means you should sell every loser or hold every winner. Sometimes selling a winner is right, such as when it has grown too large a share of your portfolio. The point is to make the choice on purpose.
Open the NetWort Dashboard and look at your Portfolio Health score and your biggest positions. Sort them by gain and loss. Ask whether you are holding your worst performers for a reason, or only because selling would make the loss real.