The disposition effect: why losing investors keep getting worse

The disposition effect — the habit of selling winners too early and clinging to losers too long — is one of the costliest biases in investing. New research reveals something uncomfortable: losing money doesn't cure it. It feeds it.

There's an old proverb: once bitten, twice shy. The idea is that a painful experience leaves a mark, and the mark makes you cautious next time. It's a reassuring story. When it comes to investing, it's also largely wrong.

The disposition effect was named by Hersh Shefrin and Meir Statman in 1985. 13 years later, Terrance Odean produced the first large-scale proof. Tracking 10,000 US brokerage accounts, he found investors were consistently more likely to realise a gain than a loss, even when the loser was objectively the worse stock to hold.

The bias isn't exotic. It's been documented across markets, asset classes and investor types, from US retail traders to emerging-market professionals. If you've ever held a sinking share because selling it felt like 'making the loss real' — or cashed in a rising one because you didn't want to give back the profit — you've lived it. You just may not have named it.

The scar tissue of an early loss changes how you respond to the next one. That, it turns out, is the problem.

The doom loop: how losses make the disposition effect worse

A study of nearly 190,000 Chinese investors has found that past losses amplify the disposition effect by around 10 per cent, while past gains reduce it by roughly the same amount. The asymmetry produces a self-reinforcing spiral: the people losing money keep making the mistake that loses them more.

The paper, 'Navigating through fear and greed: The experience-driven disposition effect' (Yeung et al., 2025), tracked 189,530 retail investors through the 2013–2016 Chinese equity boom and bust — more than 40 million trades in all. It was presented at the American Finance Association meetings in 2026. The authors sit at serious institutions: Jessica Wachter is at Wharton, Michael Kahana at Penn.

The baseline figure matters. On average, investors were 2.67 percentage points more likely to realise a gain than a loss — the standard disposition effect, updated for a modern retail sample. What the researchers did next was the twist. They sorted investors by their cumulative trading experience. A two-standard-deviation increase in past significant losses amplified the bias by about 10 per cent. A two-standard-deviation increase in past significant gains reduced it by about the same margin.

So the investor who most needs to change their behaviour is the one least likely to. Lose enough, and you sell winners faster, cling to losers longer and churn the portfolio in the exact pattern that produced the losses in the first place. That's the doom loop. Each loss adds another layer of scar tissue. The investor doesn't get tougher. They get stiffer.

Two subgroups get hit hardest: younger investors (under 36) and those with less wealth. That matters now, because millions of people opened their first trading accounts during the pandemic and met their first real drawdowns in 2022. The model predicts those losses will have made them worse traders, not better.

Joachim Klement, who covered the research on his Investing blog in March 2026, flagged an observation from the paper's own data: when you split investors' holdings between the stocks they've already sold and the stocks they're still holding, the realised trades look fairly symmetrical — but the stocks still on the books are overwhelmingly losers. The loop shows up right there in the portfolio. The winners get marked as wins and closed. The losers stay open, hoping.

One caveat, stated plainly: this is a working paper. It has not yet cleared peer review. The data comes from Chinese retail investors, who may trade differently from their UK or US counterparts. The research team is strong and the mechanism is consistent with earlier memory-based theories, but the findings should be read as credible rather than final.

Why losing investors can't learn their way out

The doom loop is self-sealing because the brain edits the evidence. Investors over-remember their gains and suppress the memory of their losses, which means the raw material for learning is distorted before analysis even starts.

Maya Strahilevitz, Terrance Odean and Brad Barber showed this directly in a 2011 study. When investors were asked to recall their past trading, gains came back more readily and more accurately than losses. The books are cooked before the audit begins.

The Yeung team's contribution is a mechanism for that editing. Large losses, they argue, leave sticky memory traces that distort the next decision. This is different from the prospect-theory explanation, in which loss aversion is a fixed psychological trait. Under the memory model, the bias is learned and deepened rather than hard-wired and stable.

Financial education, often proposed as the fix, runs into the same wall. Literacy correlates with better average behaviour, as Lusardi and Mitchell have documented. But at the moment of decision — when the screen flashes red and the urge to hold or the urge to sell is acute — emotional override wins. Knowing about a bias is not the same as escaping it.

There's a UK angle worth pausing on. The Financial Conduct Authority's Financial Lives 2022 survey found 12.9 million UK adults with low financial resilience — about 24 per cent of the adult population. Many of those same people opened trading accounts during the pandemic and met real losses in 2022. The FCA's own randomised experiment with more than 9,000 consumers found that digital engagement practices — push notifications, points and prize draws, trader leaderboards — increased both trading frequency and risk-taking, with disproportionate effects on those with lower financial literacy. If the Yeung findings hold, those prompts aren't just encouraging more trades. They're feeding the loop.

The brain doesn't just form scar tissue. It edits the medical record. You remember the recoveries but not the burns.

The cost of the doom loop

The disposition effect doesn't only distort which shares you hold. It churns the whole portfolio — and the bill for that churn has been measured.

Barber and Odean's 2000 paper, elegantly titled 'Trading is hazardous to your wealth', looked at more than 66,000 US households. The least active quintile of traders earned 18.5 per cent a year. The most active earned 11.4 per cent. A gap of 7.1 percentage points, driven by transaction costs, spreads and poor timing — not by bad luck.

Connect that gap to the Yeung findings and the picture sharpens. The disposition effect increases trading activity (selling winners, reshuffling positions) while the losers sit stuck in the account. The more losses you accumulate, the worse that pattern gets. On a £500,000 portfolio, a 7.1 percentage point annual drag is roughly £35,000 a year in foregone growth. That's not a statistical abstraction. That's a holiday, a car, a year of a grandchild's school fees.

The original Barber-Odean data is about trading activity generally, not the disposition effect in isolation. But the doom loop is a machine for producing exactly the kind of activity their study priced.

How to break the doom loop

You can't think your way out of the disposition effect. You can design your way out. The trick is to remove the triggers that set it off and the decisions that sustain it.

Start with the anchor. Cary Frydman and Antonio Rangel showed in a 2014 lab experiment that when the trading interface hides the original purchase price, the disposition effect shrinks by around 25 per cent. The purchase price is the emotional anchor; the moment you see it, the question becomes 'am I up or down on this one?' rather than 'is this still a good holding?'. Some platforms let you hide your cost basis. If yours doesn't, that's a design flaw, not a feature.

Next, automate the exit. Fischbacher, Hoffmann and Schudy (2017) found that pre-committed stop-loss and take-gain orders significantly reduce the disposition effect. The crucial part of their finding was the control condition: simply reminding people of their intended selling plan did nothing. Only the binding order worked. Automation strips the decision out of the moment of acute emotional pressure, which is the only moment the bias operates.

And then there's the simplest escape, the one this publication has argued for since it started: own the market, not individual stocks. If you hold a global tracker, there's no sell/hold decision on 'this loser' because there is no 'this loser'. Systematic rebalancing replaces emotional trading. The purchase-price anchor disappears. The doom loop has nothing to bite into. Getting rich slowly works partly because it denies the bias a target.

You can't undo the scarring. But you can stop reaching for the stove.

The real lesson experience teaches

The most valuable investing lesson isn't what to buy or sell. It's recognising when to take yourself out of the decision entirely.

Once bitten, twice shy turns out to be backwards for investors. Getting bitten, the evidence now suggests, makes you more likely to bite yourself. The disposition effect, the doom loop, the scar tissue — they all end up in the same place. The problem isn't a shortage of information or experience. The problem is that experience, when it involves losses, reinforces exactly the wrong instincts.

The investors who do best over a lifetime are not the ones who learned to trade better. They are the ones who learned to trade less. That's not a counsel of despair. It's a release.

Resources

Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of individual investors. The Journal of Finance, 55(2), 773–806.

Fischbacher, U., Hoffmann, G., & Schudy, S. (2017). The causal effect of stop-loss and take-gain orders on the disposition effect. The Review of Financial Studies, 30(6), 2110–2129.

Frydman, C., & Rangel, A. (2014). Debiasing the disposition effect by reducing the saliency of information about a stock's purchase price. Journal of Economic Behavior & Organization, 107, 541–552.

Odean, T. (1998). Are investors reluctant to realize their losses? The Journal of Finance, 53(5), 1775–1798.

Shefrin, H., & Statman, M. (1985). The disposition to sell winners too early and ride losers too long: Theory and evidence. The Journal of Finance, 40(3), 777–790.

Strahilevitz, M. A., Odean, T., & Barber, B. M. (2011). Once burned, twice shy: How naive learning, counterfactuals, and regret affect the repurchase of stocks previously sold. Journal of Marketing Research, 48(S1), S102–S120.

Yeung, T. L., Liu, R., Wachter, J. A., Kahana, M. J., & Zhang, Y. (2025). Navigating through fear and greed: The experience-driven disposition effect. Working paper.