How should investors behave after a large price shock – either positive or negative - occurs in a stock they hold? Are such price shocks likely to be an overreaction? An underreaction? Neither?
As a practical matter, investors are commonly confronted with a price shock, and they must decide what action, if any, to take with their stock holdings following that event. A price shock is defined in this context as a discrete price change greater than 10%, either positive or negative over the period of one week of trading.
Unfortunately, investors have precious little guidance on this matter from either analysts or economists. Most economists and investment researchers today focus on smart beta/factor models and the active vs. passive debate (see last week’s column), while analysts generally focus on trying to pick specific stocks that will do well in the future. Following a price shock, very few commentators are willing to put their necks on the line to take a view on a firm. Empirical models used by researchers have poor predictive power when explaining the returns associated with a single individual stock.
Investors are really on their own in this arena then. I recently did some consulting work developing a pricing model for use around price shocks and was surprised how little help investors have in the field.
Price shocks are complex though. In particular, shocks have markedly different implications for a firm based not only on their direction but also on their magnitude. Further, short and long-term returns following a shock may be very different and some price shocks are likely to have valuation implications for competitors of the firm experiencing the shock.
To examine how investors should respond to price shocks, I looked shocks based on the magnitude of the price change in a security over a ten trading-day period, with shocks broken into categories for each 10% change (i.e. a price change of 10% to 20% is treated as distinct from a price change of 20% to 30%).
The results show that large and small price shocks lead to different future average returns. Small price shocks (“10% to 20%”, and “20% to 30%”) are continued on average, and large price shocks (“30% to 40%” and “More than 40%”) are reversed.
In other words, stocks with middle-of-the-road price shocks of anywhere from 10 to 30% tend to be driven by momentum effects with further moves in the future in the same direction. Price shocks greater than 30% instead appear to succumb to value factors and a subsequent partial reversal.
These results hold when examining short, intermediate, and long term time horizons. (For factor and smart beta fans out there - these results are based on abnormal returns for each time horizon, and hold after controlling for risk factors present in the Carhart (1997) four-factor model.)
Importantly, even over a four year time horizon, the full magnitude of the price shock is not reversed after taking into account market returns. While a partial price reversal does undo a portion of the impact on the stock price resulting from the shock, roughly one-quarter of the shock is permanent for the average firm. The dichotomy of future stocks returns in the context of the magnitude of the price shock itself is novel in the literature and these results offer a possible explanation for some of the contradiction regarding post-shock stock returns found by past researchers.
Given these results, I went a little deeper to see if the stock price changes are associated with changes in underlying economics at the firm. Here the results were similar - price shocks that are continued, rather than reversed, lead to changes in revenues and earnings per share which are in the same direction as the shock itself. Finally, the Sharpe Ratios of firms experiencing 10% to 20% positive (negative) price shock are higher (lower) following the shock after controlling for other risk factors.
Overall then, the results suggest that investors looking at a price shock need to decide based on magnitude and direction of the shock if they can withstand future volatility and how they want to position themselves going forward.
Mike McDonald is a PhD in finance and a university professor in the subject at Fairfield University in Connecticut. He also runs a consulting company doing work on quantitative investing, big data, and machine learning for a variety of financial firms, asset managers, institutional investors, and government regulators. Prior to getting his PhD, Mike worked for a major Wall Street bank and one of the top hedge funds. Comments, questions, and concerns are always welcome – email Mike at M.McDonald@MorningInvestmentsCT.com or visit his firm’s website at www.MorningInvestmentsCT.com