The past is prologue: Disposition effect

The disposition effect, our tendency to sell stocks held at a profit too soon and hold on to stocks held at a loss for too long, is one of the most powerful drivers of investment underperformance. As I have discussed in the past, value investors are particularly at risk of suffering from a stronger disposition effect.

I have previously written about a study that shows that market environment influences the extent to which we suffer from the disposition effect. If people are in a bear market, they tend to sell stocks held at a profit much faster than in a bull market.

A new study of 189,530 Chinese retail investors went a step further and could show that it is not just the environment these investors are in (bull or bear market) that influences the disposition effect, but also past investment experiences.

Using the boom-bust cycle in Chinese equities from 2013 to 2016 as the background, they showed that investors who had accumulated more losses in their investments in the past saw the disposition effect increase, while investors with more positive trading experiences in the past saw their disposition effect decline.

Take the chart below. The horizontal axis in each of the three graphs shows how many trades the investors have made in the past (left), how many loss-making trades they have realised (middle) and how many profitable trades they have realised (right). The important bit of the charts is the dashed grey line, which measures the size of the disposition effect.

The propensity to sell winners and losers depends on past experiences

Source: Yeong et al. (2025)

If you look at the right-hand graph, you see that as investors accumulate more profitable trades, they tend to let their winners run for longer, which reduces the disposition effect. Meanwhile, the middle graph shows that investors who accumulate more losing trades tend to stick with losing stocks even longer than in the past, which increases the size of the disposition effect. It’s as if these investors want to avoid the pain of selling at a loss more, the more they experienced that psychological pain in the past. The result, of course, is that these investors tend to do even worse over time.

If you combine all portfolios, the study produces one of the most impactful charts I have ever seen on the disposition effect. Why I have never seen the charts below before is beyond me, because if a picture says more than a thousand words, then the two pictures below say more than two thousand words.

On the right, you see the distribution of profit and losses of trades that have been realised (i.e. an investment has been sold). As you can see, it is almost perfectly symmetrical with the same number of winning and losing trades. Meanwhile, on the left, you can see the distribution of profits and losses of the stocks that are held in the portfolios and have not been sold (yet). Clearly, there are many, many more losing stocks in the portfolios of these investors than winning stocks. If you want to show to people what havoc the disposition effect wreaks in their portfolios, you couldn’t do any better than these two graphs.

The resulting distribution of wins and losses of stocks held in the portfolio and stocks sold

Source: Yeong et al. (2025)