What’s the Best Time to Announce Earnings?

Access article

Earnings announcements made in pre-trading hours lead to greater stock-price volatility than do those made in post-closing hours.

Companies are united in their belief that it’s best not to announce quarterly earnings results during trading hours. About 95% of publicly held companies subscribe to that policy, according to new research.

Among those that do, there’s a fairly even split between earnings announced after the closing bell (i.e., before midnight, chosen by 53% of such companies) and those announced before the opening bell (after midnight, chosen by the rest).

But the timing decision has implications with respect to the market’s reaction to the announcement. That is, earnings releases made before trading opens are associated with greater volatility in stock prices than are those made after trading closes, according to a paper by Northwestern University professor Matt Lyle and three colleagues.

Remarkably, the increased volatility lasts for five days after the announcement, according to the research, which looks at about 70,000 quarterly earnings announcements with precise timestamps from 2006 to 2014. The data was provided by Wall Street Horizon, an earnings announcement tracker.

Specifically, on average the stock of a company that announces earnings in the pre-opening hours is approximately 3.2% more volatile per day during the five-day period, compared to its historical average, than is the stock of a company that announces earnings in the post-closing hours.

Additionally, the team looked at the specific hour when announcements are made. They found that those made close to either the opening or closing bell are less volatile than those made further away from those hours. For example, an announcement made at 6:00 a.m. Eastern time tends to be more volatile than one made at 9:00. The researchers hypothesize that the difference is explained by investors needing time to process the information before factoring it into trading activities.

“All else equal, as investors process information and revise their beliefs, they engage in trade which moves prices,” the paper says. The authors add that the differences in volatility that they document are “highly predictable” and persist even when controlling for many other factors associated with volatility, including company size, profitability, earnings surprises, volume, spreads, and historical volatility.

The researchers further note that option-trading strategies based on buying in the morning and selling at night can produce significant returns. Many CFOs, however, no doubt prefer a lower degree of volatility — an outcome that merely choosing a different hour for disclosing results could help achieve, the research suggests.

Lyle’s co-authors are Christopher Rigsby, a doctoral candidate at Northwestern’s Kellogg School of Management; Andrew Stephan, a professor at the University of Colorado; and Indiana University professor Teri Lombardi Yohn.