Tardy earnings drag on stock prices

Amazon's earnings generated a bad market reaction, but it could have been worse.

Stock in the online retailer fell 9 percent in after-hours trading last Thursday after the company reported net income of $1 a share, missing estimates of $1.56. But it likely could have been worse if the company had delayed reporting its earnings.

Research from California State University, San Marcos shows that companies see their stock price fall farther and faster if they delay delivering negative earnings surprise announcements toward the end of earnings season. Likewise, companies beating estimates with surprise earnings see greater return when they report at the beginning of earnings season.

So if you have bad news to tell your investors, it's best to get ahead of the game and come out to tell them. Delaying the inevitable just makes you look shady and won't do them any favors in the long run.

That means if Amazon had held its earnings announcement for later in earnings season, investors may have penalized the company by dumping stock, driving the price down by even more.

"Market reaction is more favorable to earnings news announced at the beginning of earnings season compared with those announced at the end of earnings season," wrote Qi Sun, the paper's author and a professor of finance. The paper was published in 2015 in the International Journal of Business and Finance Research.

Prior research has shown that a company's earnings release is highly predictable, so it shouldn't be a surprise when the numbers come out. But there are complications: For one thing, the clustering of earnings announcements ( "earnings season") and business news coverage increase the information available to investors. On the other hand, a particular company can find its announcement lost in the noise.

Stocks with earnings that beat a median of analysts' estimates released at the beginning of earnings season see an average gain of 3.1 percent over a three-day window following earnings and 2.9 percent in the following 60 days. That total of 6 percent exceeds the total price increase of 2.8 percent for similar companies that report at the end of earnings season.

‹

• Phil and Iowa: They both stink at forecasting
• Regulators can't stop dark pools, up 22%

›

On the other hand, companies reporting negative surprise earnings early in the season see a far smaller drop in their share price than those that wait until the end of the season. Companies reporting late in the season see their stocks drop an average of 4.1 percent in the 60 days after negative surprise earnings, compared with 1.1 percent for companies reporting early.

There may be an industry effect as well. There's evidence that the post-announcement price drift for companies reporting at the end of earnings season can be affected by the earnings of companies in the same industry reported early in the season.

So companies in industries that are seen as ascendant could have their stock price inflated before their earnings come out, based on other companies in the industry. On the other hand, if energy companies have been sinking for a month, investors might not penalize negative earnings announcements made at the end of a season.

Extreme good earnings are clustered toward the beginning of earnings season, with 40 percent being released in the beginning and only 15 percent at the end. Extreme bad earnings, on the other hand, are more evenly distributed through the earnings season, with 28 percent reported at the beginning and 21 percent at the end. Nineteen percent are released after the season is over.

Researchers have long looked at corporate strategy for scheduling earnings announcements.

Managers issue news releases at the end of the day to give traders the night to sleep on the news; they keep earnings calls short when the news is bad and call selectively on positive analysts; they even time earnings announcements for Fridays, when investors and business journalists are packing up for the weekend. (Though whether or not the "Friday night dump" works is a matter of some debate.)

Note: The paper's research is based on 420,900 earnings announcements between 1985 and 2003 by companies listed on the New York Stock Exchange, American Stock Exchange and Nasdaq. "Earnings surprise" is defined as the difference between the I/B/E/S reported earnings-per-share and a median of analysts forecasts.