Earnings EstimatesDownload printable PDF
One lesson many investors quickly learn is that the market is forward-looking. Security prices are dictated by expectations, and prices fluctuate as these expectations are affirmed or are proven to be unfounded. Some of the most frequently followed measures of market expectations are consensus earnings estimates. Stock Investor Pro provides a variety of earnings estimate data from I/B/E/S, which is part of Thomson Reuters. Users will find this data on the Estimates tab of the Stock Notebook within Stock Investor Pro. In this installment of Stock Investor News, we provide a discussion of this data, as well as how it may be used in the screening process.
Looking at Consensus Estimates
The Estimates tab for Yahoo! Inc. (YHOO) shows a high number of analysts providing quarterly and annual estimates. Roughly 45% of the companies in the Stock Investor database are tracked by at least one analyst.
Looking at the figure, we see that Yahoo!’s consensus earnings estimate for the current fiscal quarter ending June 2012 is $0.22 per share. For the current fiscal year ending December 2012, analysts are expecting Yahoo! to earn $0.95 per share. The current consensus estimate of growth in earnings per share over the next three to five years is 12.5% on an annualized basis.
Examining the range of estimates, as indicated by the high and low estimates for each period, provides an indication of the degree of consensus between analysts. A wide range of estimates would point to great disagreement among analysts, indicating greater uncertainty, and a greater chance for an earnings surprise (both positive and negative). For the current fiscal year, analysts tracking Yahoo! see earnings coming in at between $1.18 per share and $0.68 per share.
Perhaps a more meaningful measure of uncertainty among analysts is the standard deviation of estimates. Standard deviation provides a statistical measure of the dispersion of the earnings estimates. Assuming that earnings estimates are normally distributed, there is a 68.3% probability that the actual earnings will be within plus or minus one standard deviation of the consensus estimate. Continuing with this assumption, there is a 95.4% probability that the actual earnings will be within plus or minus two standard deviations, and a 99.7% probability that the actual earnings will be within plus or minus three standard deviations. Therefore, given Yahoo!’s mean earnings estimate for the current fiscal year of $0.95 per share and a standard deviation of $0.11, there is a 95.4% probability that Yahoo! will announce fiscal-year 2012 earnings between $0.73 per share and $1.17 per share (the mean estimate of $0.95 per share plus or minus two times the standard deviation of $0.11).
The number of estimates provides a feel for the depth of coverage for a company. Twenty-two analysts are providing an estimate for Yahoo!’s current fiscal quarter ending June 2012. As mentioned earlier, Yahoo! has a high number of analysts providing consensus earnings estimates. Currently, on average, six analysts provide consensus estimates for those companies in the SI Pro database with any analyst following. When using earnings estimates in stock analysis, the first rule to keep in mind is that the current price usually reflects the consensus earnings estimate. Therefore, in pursuing a screening strategy using earnings estimates it is best to focus on surprises and revisions, rather than raw estimates.
Earnings surprises occur whenever a company reports actual earnings that differ from the consensus estimate. During the earnings reporting season—which for most companies follows the end of their fiscal quarters in March, June, September and December—financial newspapers and websites provide daily reports on earnings announcements. Firms with significant earnings surprises are often highlighted in the media.
A positive earnings surprise occurs when actual announced earnings are above the consensus per share estimate. Negative earnings surprises take place when announced earnings are below the earnings expectations. Firms with significant positive earnings surprises tend to show above-average price performance, while those with negative surprises tend to have below-average price performance.
Measuring Earnings Surprises
To say that a company missed or exceeded their consensus estimate for the quarter does not necessarily capture the true significance of such an event. There are several ways of measuring the significance of an earnings surprise. One of them is the percentage surprise.
The Estimates tab for Google Inc. (GOOG), one of Yahoo!’s main competitors, shows that on April 12, 2012, the company announced that it had earned $10.08 per share in its first quarter. This exceeded the consensus estimate for the period of $9.65 per share—a $0.43 per share positive earnings surprise. Referring again to the figure, we see that this resulted in a 4.4% surprise. To calculate the percentage surprise, use the following formula:
Surprise % = (Actual EPS – EPS Estimate) ÷ EPS Estimate
Google Surprise % = ($10.08 – $9.65) ÷ $9.65
Google Surprise % = 0.044, or 4.4%
Another method of measuring the magnitude or significance of earnings surprises is with the standardized unexpected earnings (SUE) score. SUE measures the earnings surprise in terms of its number of standard deviations above or below the consensus earnings estimate. An earnings surprise is considered more significant the farther it is outside the statistical range of estimates expected at the time of the announcement. As stated earlier, assuming a normal distribution of earnings estimates, 68.3% of actual earnings will be within one standard deviation of the consensus estimate, 95.4% will be within two standard deviations, and 99.7% will be within three standard deviations. The absolute value of SUE measures the degree of unexpected earnings. There is no earnings surprise when the SUE score equals zero; the actual earnings per share is in line with the consensus earnings estimate.
The standard deviation of Google’s earnings estimates for its second quarter was $0.38. Given this, along with Google’s $0.43 positive earnings surprise, we arrive at a statistically significant SUE score of 1.1:
SUE = (Actual EPS – EPS Estimate) ÷ Standard Deviation of Earnings Estimates
Google SUE = ($10.08 – $9.65) ÷ $0.38
Google SUE = 1.1
Note that SUE scores in Stock Investor Pro are absolute values, meaning they will never be negative.
Those stocks with higher SUE scores tend to exhibit stronger price reactions to earnings surprises than stocks with relatively low SUE scores. This is not always the case, however, as Google shares gained less than 1% the day following the company’s aftermarket earnings announcement.
Impact of Earnings Surprises
Changes in stock price resulting from a positive or negative earnings surprise can be felt immediately, but the surprise can also have a long-term effect. While it may be difficult for individuals to buy on the initial surprise event, studies indicate that the surprise effect can persist for as long as a year after the announcement.
Therefore, it may not be prudent to buy a stock that has declined following a negative earnings surprise thinking it is now “attractively priced.” There is a good chance that the stock will continue to underpeform the market for some time. Alternatively, it may not be too late to buy a stock that has seen its price jump following a positive earnings surprise.
Not surprisingly, the stock prices of larger firms tend to adjust to surprises faster than those of small firms. This is because more analysts and portfolio managers, who can buy and sell on news quickly, track larger firms.
Firms with an earnings surprise one quarter—either positive or negative—tend to have additional surprises in subsequent quarters. This so-called “cockroach effect” implies that where there is one earnings surprise, more are likely to follow. When a surprise signals a change in company fundamentals, management and analysts alike are typically slow to realize the strength and persistence of the change. As a result, company guidance and analyst estimates do not accurately reflect this change in fundamentals.