Negative News Precedes Drop in Stocks With Big Price Gains

Stocks that crash after price run-ups have more negative sentiment in earnings news during their run-up period than stocks that don’t crash.

On average, bubble stocks have two more negative news articles in the run-up period (the two years leading up to the stock price run-up) than non-bubble stocks. While this may seem like a small difference, it accounts for approximately 25% and 16%, respectively, of the total number of earnings news articles over two years for non-bubble and bubble stocks. There is no statistical difference in the number of positive earnings news items between the two types of stocks.

The negative sentiment holds predictive power up to two years in advance, allowing it to predict whether a stock run-up reflects a bubble and will be followed by a crash. On average, bubble stocks maintained a relatively stable return for around six months before crashing. The average drop was nearly 33% in terms of average return through two years after the price run-up. For non-bubble stocks, on average, there was a persistent return increase after the price run-up.

Furthermore, the study finds that past negative sentiment in earnings news predict a bubble’s future maximum drawdown: A greater number of past negative sentiment items in earnings news implies a higher maximum drawdown.

The bubble and non-bubble stocks used in this study were identified by finding stocks that experienced a price run-up in the past two years. The author defines a price run-up in three ways: a 100% or more raw return in the past two years; a 100% or more net market-adjusted return in the past two years; or a 50% or more raw return in the past five years.

Stocks were categorized as bubble stocks if they experienced a crash after their run-up, defined as a 40% or more drawdown in absolute terms beginning at any point in the two years after the first month in which the stock first experiences a run-up. If this didn’t apply, then they were categorized as non-bubble stocks.

Although the study focused its attention on medium- to large-size stocks, the author says the probability of a crash is even higher for small stocks that fall under the bubble stocks group.

Source: “Bubble Sentiment,” by Gen Li; Yale School of Management, International Center for Finance, November 2019.