Huntsman Post

Accounting professor investigates the best time to disclose bad news

Editor’s note: Dr. Jeffrey Doyle’s research has been quoted in major national publications such as the “New York Times,” the “Wall Street Journal” and “Businessweek.” Last year he was named Professor of the Year for USU and Teacher of the Year for the Jon M. Huntsman School of Business. He is the George S. Eccles Chair in Capital Markets Research in the School of Accountancy. He agreed to share some of his most recent research with “The Huntsman Post.”

By Dr. Jeffrey Doyle

A common belief among investors is that firms tend to release bad news at times when market participants may not be fully paying attention, thus limiting any potential negative reaction. For example, if a firm has bad-earnings news to report, conventional wisdom would suggest that opportunistic managers would prefer to release the information after the markets are closed for the day and/or on Fridays, when there is less media and investor attention. In fact, research done in the 1970s produced evidence consistent with this notion.

However, given the evolution to a 24/7 financial media environment featuring dedicated financial channels such as CNBC over the past 30 years, I, along with my coauthor, Professor Matthew Magilke from the University of Utah, questioned the viability of a strategy to “hide” bad news and the associated implied assumption that the market is less informationally efficient at certain times.

In order to explore this subject, we gathered information on more than 50,000 quarterly earnings announcements from 2000 to 2005. We then eliminated those firms that never changed the timing of their announcements during the six years, reasoning that these firms were clearly not strategically picking a time and day to make an announcement based on the news they would be sharing. Among the remaining “switching” firms, we then compared announcements made before the market opened to those announced after the market was closed. If the opportunistic manager theory was true, the “smoking gun” would be finding worse earnings news being reported for those announcements released after the market was closed. Similarly, we compared announcements released on Monday through Thursday to those released on Fridays, looking to see if there was any evidence bad news was being saved for a Friday release. We found the following results:

  • We discovered no evidence that earnings released after the market was closed or on Fridays contained worse news, on average, than announcements made at other times, contradicting the conventional wisdom that managers opportunistically time the release of bad news.
  • We found that, rather than releasing bad news on Fridays, firms tend to avoid the day altogether for earnings announcements. The volume of Friday announcements was 82 percent lower than the average for Monday through Thursday announcements. It is as if managers now fear the perception of trying to hide their bad news.
  • We found some evidence that, for those firms with bad news that do switch to a Friday announcement, the market seems to penalize them for doing so, which would seem to back up our theory that firms are seeking to avoid Friday announcements altogether.
  • We then explored other determinants of the timing decision, including the more benign hypothesis that managers release earnings after the market closes and/or on Fridays to give the market more time to digest more complex information. Consistent with this idea, we found some evidence that more complex firms tend to announce earnings after the market closes. We also found that these announcements are associated with greater trading volume, possibly indicating a successful dissemination strategy. Firms that tend to announce after the market closes also tend to be smaller, have more analysts, have a higher percentage of institutional shareholders and be located in the western time zones.

In summary, rather than being a result of managers trying to “fool” the market by hiding bad news, our results are indicative of more benign managerial motives and institutional factors. Hence, our findings are consistent with efficient capital markets that are effective at monitoring new information, regardless of the time of the announcement.

This research was published in the January 2009 issue of “The Accounting Review,” one of the top three accounting journals in the world, as “The Timing of Earnings Announcements: An Examination of the Strategic Disclosure Hypothesis.” Please feel free to email Professor Doyle at jeffrey.doyle@usu.edu with any comments or for a copy of the complete article.

Click here to return to The Huntsman Post newsletter.