The disposition effect: why investors sell winners too early and hold losers too long
The disposition effect is the tendency of investors to sell assets that have gained in value too quickly and to hold assets that have lost value for too long. Shefrin and Statman (1985) identified and named the pattern; Odean (1998) confirmed it empirically in a large sample of real investor accounts—finding that the stocks investors sold subsequently outperformed the stocks they retained.
What the disposition effect is
The term was coined by Hersh Shefrin and Meir Statman in their 1985 paper The Disposition to Sell Winners Too Early and Ride Losers Too Long: Theory and Evidence, Journal of Finance. They drew on Prospect Theory (Kahneman and Tversky, 1979) and mental accounting to explain why investors systematically deviate from the financially optimal strategy of selling losers and holding winners.
The mechanism is rooted in the asymmetric shape of the Prospect Theory value function. In the gain domain, the function is concave—the marginal pleasure of an additional gain diminishes. In the loss domain, it is convex and steep—losses near the reference point are particularly painful to realise. Selling a winner moves away from the reference point in the gain domain, which feels less rewarding than it should. Selling a loser crystallises a loss, which feels more painful than the financial impact warrants. The rational strategy—let winners run, cut losses—runs directly against the emotional logic the value function describes.
Mental accounting reinforces the effect. Investors mentally segregate each position into its own account with the purchase price as its reference point. This means positions are evaluated individually rather than as part of a portfolio, making it psychologically easier to hold a losing position in isolation than to acknowledge its contribution to overall performance.
How it manifests in investing
The most direct consequence is a systematic bias in portfolio composition over time. An investor subject to the disposition effect progressively accumulates positions weighted toward underperforming assets: winners are sold while they are still productive; losers are retained long after the evidence for holding has deteriorated. The portfolio's composition drifts toward its weakest holdings.
The tax consequences compound the financial cost. Selling winners early realises taxable gains while the position is still performing. Holding losers prevents the realisation of tax losses that could offset gains elsewhere in the portfolio. The disposition effect systematically produces the worst possible tax sequence: early gain realisation and deferred loss realisation.
A third expression is the return-to-reference-point fixation: investors hold a losing position with the intention of selling once it recovers to the original purchase price, treating the entry price as a meaningful target even though it contains no information about the asset's future expected return. The entry price is an irrelevant historical fact; treating it as a threshold for exit is a direct consequence of the reference-point structure of the value function.
The cost
Odean (1998) examined 10,000 trading accounts at a US discount broker over 1987–1993 and found that the stocks investors sold outperformed those they retained by an average of 3.4 percentage points over the following year. The finding was robust across different market conditions and holding periods, and directly contrary to what investors expected when they made their decisions. Grinblatt and Keloharju (2001) confirmed the pattern in Finnish equity markets, demonstrating that the disposition effect is not limited to inexperienced individual investors—it is present across a wide range of investor types, including institutions.
What helps
Rules-based selling criteria remove the reference point from the exit decision. A momentum strategy exits positions when their trend reverses, regardless of whether the position is above or below its entry price. A volatility-triggered rule exits when a position's behaviour changes in a specified way, again without reference to the original purchase price. In both cases, the decision is grounded in evidence about expected future return, not in the investor's position on the value function.
Systematic rebalancing to target weights has the same effect at the portfolio level: it sells allocations that have grown above target and buys allocations that have fallen below—mechanically implementing the opposite of what the disposition effect would produce. The discipline is structural, not motivational.
The disposition effect in pfolio
pfolio's systematic rebalancing directly counteracts the disposition effect at the portfolio level. Monthly rebalancing buys assets that have declined in allocation weight and sells assets that have grown above target—the exact opposite of what the disposition effect would produce. At the position level, momentum signals exit assets when their trend reverses, based on price performance over the measurement window rather than the position's gain or loss relative to entry price. The entry price is not an input to any calculation in the methodology. See how we build portfolios for a full description of the signal and rebalancing logic.
Related articles
Disclaimer
Get started now
