How Good is Guidance?
- 1. HOW GOOD IS GUIDANCE? _________________________________________________________________________________ FUNDAMENTAL RESEARCH JUNE 21, 2012 HOW GOOD IS GUIDANCE? WHO MAKES THE BEST PREDICTIONS, ANALYSTS OR COMPANIES THEMSELVES, AND WHAT DOES THAT MEAN FOR STOCK PRICES? HIGHLIGHTS Companies are more accurate than analysts at estimating their future earnings per share results. However, after companies issue guidance, analysts adjust their estimates, ultimately creating estimates that are more accurate than those issued by the company. Companies issuing positive guidance saw the greatest increase in estimate accuracy post-guidance, suggesting that positive guidance is more “surprising” to analysts than negative or in-line guidance. The excess stock returns associated with company issued guidance are split between the date that guidance was issued and the date the company reported earnings results. On the guidance date, companies giving positive guidance see an excess stock return of 2.6%, while those guiding lower lose 1.3%, and those that guide in-line gain 0.7%. Companies that offer guidance – positive or negative – see a minimal additional positive excess return if they beat the final analyst estimate, but significant negative excess returns if they miss or match estimates, suggesting that the market expects both groups of companies to beat estimates. Investors should be aware of these expectations when assessing the risk of their positions as companies report earnings. Strategies of guiding lower to enable a company to beat estimates or guiding higher when the company’s fundamentals don’t support it to take advantage of the stock price effect on guidance day would not be effective, as both have negative returns on average. DEFINITIONS Initial Consensus: The consensus analyst earnings per share estimate at the time that the company issued guidance. Company Issued Guidance: A public statement provided by the company estimating earnings per share for the coming period. Final Consensus: The consensus analyst earnings per share estimate at the time that the company reports quarterly earnings per share results. Actual: Quarterly earnings per share results reported by the company. Positive Guidance: Company issued guidance projecting higher EPS when rounded to two decimal places than the consensus analyst estimate at the time guidance is issued.
- 2. HOW GOOD IS GUIDANCE? JUNE 21, 2012 Negative Guidance: Company issued guidance projecting lower EPS when rounded to two decimal places than the consensus analyst estimate at the time guidance is issued. In-line Guidance: Company issued guidance projecting the same EPS when rounded to two decimal places as the consensus analyst estimate at the time guidance is issued. METHODOLOGY To evaluate the accuracy of company issued guidance and the effects of guidance on stock prices, we examined every instance of quarterly earnings per share guidance given by S&P 500 companies over the past eight quarters, a total of 972 guidance announcements. For each guidance announcement, we compiled the consensus analyst estimate at the time guidance was given, the guidance estimate given by the company, the final analyst estimate prior to the earnings announcement, and the actual earnings per share reported by the company. This allowed the determination of relative accuracy of the analyst and company earnings estimates. Stock returns were compiled for the applicable dates to measure the effect on stock prices when companies announced their guidance and when they reported actual earnings results. This analysis is limited to instances in which companies provided quarterly earnings per share estimates with a specific dollar value or range of values. In cases where a range was given, we used the midpoint of the range. All values were rounded to the cent for the purposes of comparing actual results with estimates and guidance. WHY DO COMPANIES ISSUE GUIDANCE? The primary reason that corporations issue guidance is to frame expectations for their upcoming financial results. Companies differ on the scope and detail of their guidance, with some companies limiting themselves to qualitative descriptions of their earnings trend, and others providing detailed estimates for revenue, earnings, margins, and more. Only a subset of companies chooses to issue guidance. Over the past two years, on average 122 of the 500 companies in the S&P 500 index have issued quarterly earnings guidance. Many of these companies routinely issue guidance, while others only do so as needed. Of the 187 companies that issued guidance during the period examined, 44% of them did so in all eight quarters. On the other hand, 34% of the companies in the study only offered guidance in one or two quarters. Companies that provide earnings guidance are more likely to give guidance that is below the consensus analyst estimate than above it. During our study period, 59% of the guidance estimates were negative, compared with 32% positive. There are a couple of potential explanations for this. One is that some companies only provide guidance if there is negative material nonpublic information that the company wants to make investors aware of. However, when looking at companies that gave guidance while reporting their previous quarter’s earnings, we see that they give negative guidance at a higher rate than those companies that give guidance at a later point in the quarter. Another explanation for the negative bias in guidance is that companies tend to be conservative when forecasting their results to avoid promising earnings that they will not be able to deliver. This explanation is more likely to be the reason for more of the companies giving negative guidance, especially considering the effects on stock prices of missing earnings estimates, discussed below.
- 3. HOW GOOD IS GUIDANCE? JUNE 21, 2012 IS GUIDANCE ACCURATE? Given the far greater amount of information that a company’s management team has about their operations than any outside analyst could have, it seems logical that company issued guidance should be more accurate than analyst estimates. Overall, this is the case – on an absolute basis, the average initial consensus differs from the actual earnings by 17.9%. Company issued guidance, on the other hand, differs from actual results by 9.8%, a significant improvement. When companies give earnings guidance, the analysts covering the company usually revise their estimates towards the guidance that was given. Guidance is not taken as a given, though, as analysts interpret the guidance and add their own analysis, and continue to revise their estimates. The final analyst consensus estimate, incorporating company issued guidance and information gained since the guidance announcement, is the most accurate, differing from actual results on average by 8.0%. EXHIBIT 1. ABSOLUTE ESTIMATE ERROR Positive Guidance Negative Guidance In-line Guidance All Guidance Initial Consensus 17.9% 9.9% 6.9% 12.2% Guidance 9.8% 10.1% 6.8% 9.8% Final Consensus 8.4% 8.1% 6.1% 8.0% 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% Source: Thomson ONE As shown above in Exhibit 1, the relative accuracy of company issued guidance varies depending on the type of guidance given. When companies give positive guidance, they tend to be significantly more accurate than initial analyst consensus, lowering absolute error from 17.9% to 9.8%. Again, we see the effect of additional analyst revisions, as the final consensus estimate differs from the actual EPS results by 8.4%.
- 4. HOW GOOD IS GUIDANCE? JUNE 21, 2012 In the case of companies issuing guidance in line with expectations, the error doesn’t change significantly (small difference due to rounding), and the estimates exceed actual earnings by 6.9%. The error in these instances is lower than for companies issuing positive or negative guidance. These companies also see additional accuracy as a result of further analyst revisions, but to a lesser extent. Accuracy increases by 0.8%, giving the final consensus estimate accuracy of 6.1%. When companies give negative guidance, they are slightly less accurate on an absolute basis, differing from actual earnings by 10.1%, compared to the 9.9% difference between initial analyst estimates and actual earnings. Final consensus estimates for negatively guiding companies are again more accurate, with error of 8.1%. When viewed on a simple average basis, as shown in Exhibit 2, below, it becomes apparent that, even though the accuracy is similar on an absolute basis, companies giving negative guidance are significantly less accurate than initial analyst estimates because their inaccuracy is in the opposite direction and is more extreme. In other words, the initial analyst estimates are higher than the eventual earnings reports by 2.7% on average. The guidance given by these companies, however, is lower than actual earnings by 7.6%, implying that when a company decides to give negative guidance, they are very conservative so as to avoid the possibility of missing a lowered target. EXHIBIT 2. AVERAGE ESTIMATE ERROR Initial Consensus, 2.7% Guidance, -7.6% Final Consensus, -5.2% -10.0% -8.0% -6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Source: Thomson ONE WHAT HAPPENS TO STOCK PRICES? To evaluate the effect of giving guidance on stock prices, we analyzed the excess returns of the companies that gave guidance during the study period on the day they announced their guidance (or the following day for companies that announced guidance after the market close). For this
- 5. HOW GOOD IS GUIDANCE? JUNE 21, 2012 analysis, we define excess return as the return of the stock minus the return of the S&P 500 index on the same day. This reduces market effects on price, allowing measurement of the individual company’s performance. When companies in our sample gave positive guidance, their stock priced increased by an average of 2.6% during the first trading session. See Exhibit 3, below. This outweighs the 1.3% loss that the average company saw after giving negative guidance. This disparity suggests that positive guidance is more “surprising” to the market, possibly due to negative guidance often following negative events that are already publicly known. There was also a slight positive effect when companies confirmed analyst estimates, as market participants felt more certainty about the coming earnings. EXHIBIT 3. STOCK PRICE EFFECTS ON GUIDANCE ANNOUNCEMENT DATE Excess stock return on day after guidance announcement Positive Guidance 2.6% Negative Guidance -1.3% In-line Guidance 0.7% Source: Thomson ONE When companies report their actual earnings results, there is an additional stock price effect, which varies depending on the nature of the previously issued guidance. As shown below in Exhibit 4, companies that gave positive guidance are rewarded a modest 0.4% excess return if they beat the final consensus estimate. This is in addition to the positive excess returns achieved when they announced the positive guidance initially. If these companies miss the final consensus estimate, on the other hand, they lose 3.1%, more than erasing the initial benefit seen when they reported positive guidance in the first place. Positive guiding companies that match final estimates lose 2.1% on average. Given the modest impact for beating estimates and the negative effects for missing or matching, this data suggests that investors expect positive guiding companies to continue the upward momentum of earnings expectations. Companies that issued negative guidance see a somewhat similar pattern of excess stock returns when reporting earnings, although probably for different reasons. Like the positive guiding companies, companies giving negative guidance see modest returns for beating final estimates and significant losses for missing them. They are also hit with losses for matching expectations. This is on top of the 1.3% average loss they experience when giving the negative guidance initially. This market expectation that these companies should beat the final analyst consensus makes sense in light of the accuracy data discussed above. As these companies give guidance that is significantly lower than prevailing estimates at the time, they actually overshoot to the downside, producing estimates that are 7.6% below the actual estimate. This conservative guidance has, in effect, given the companies a healthy cushion for their earnings results. Even though analysts revise their estimates to a less conservative level, investors still expect them to beat estimates, suspecting that they are conservative. Companies that give guidance that is in-line with initial analyst consensus appear show a different pattern when it comes to stock returns upon reporting earnings. In these cases, companies see a significant return when beating estimates and a significant loss when missing. In contrast to companies that guide higher or lower than consensus, companies guiding in-line with estimates actually succeed in setting accurate expectations for the market, given the minor change in stock price when they match the final analyst consensus estimate.
- 6. HOW GOOD IS GUIDANCE? JUNE 21, 2012 EXHIBIT 4. STOCK PRICE EFFECTS ON EARNINGS REPORT DATE Excess stock return on earnings report date Beat Miss Match Positive Guidance 0.4% -3.1% -2.1% Negative Guidance 0.2% -2.7% -1.0% In-line Guidance 1.9% -3.8% 0.2% Source: Thomson ONE Having discussed the separate stock price effects at the time of guidance and when earnings are reported, it is instructive to combine them and draw conclusions. Overall, shareholders are best served when company management offers accurate guidance. In the case of positive guidance, it should come as no surprise that companies that guide higher and then surpass the final estimate see their stock prices benefit the most, at 3.0%, as seen below in Exhibit 5. When making the decision to guide higher, however, they should be confident they will match or beat their more optimistic guidance, as the penalty for missing exceeds the initial benefit from announcing positive guidance. Likewise, companies should avoid giving negative guidance unless it is absolutely necessary. Giving low expectations in hopes of setting an easy bar to clear is not effective, as the modest gains from beating these low estimates do not make up for the initial losses from the guidance announcement. On the other hand, negative guidance is sometimes appropriate when the company truly believes that analyst expectations are too high. This is exemplified by the cases where companies gave guidance in-line with estimates only to fail to meet them. The stocks of these companies endured more negative excess returns than those of the companies who gave negative guidance and ended up matching final estimates.
- 7. HOW GOOD IS GUIDANCE? JUNE 21, 2012 EXHIBIT 5. TOTAL STOCK PRICE EFFECTS (GUIDANCE DATE AND EARNINGS REPORT DATE) Total Excess Return Beat Miss Match Positive Guidance 3.0% -0.5% 0.5% Negative Guidance -1.1% -4.0% -2.3% In-line Guidance 2.6% -3.1% 0.8% Source: Thomson ONE CONCLUSION Company issued guidance provides valuable information to analysts and market participants by providing insight into company management’s outlook for upcoming earnings results. As this information is generally more accurate than analyst estimates at the time, it helps analysts to refine their estimate. When analysts overlay their analysis onto the guidance, they are able to come up with estimates that are ultimately more accurate than those issued by the company. There are significant effects on stock returns when companies issue guidance and when they report earnings results. By releasing new information into the market, companies alter the market expectations. In the case of positive and negative guidance alike, the market seems to expect companies to beat estimates. However, investors take in-line guiding companies at their word, as evidenced by the small market reaction for these companies that match estimate, compared with the significant positive and negative reactions when they beat and miss, respectively. AUTHOR: Greg Harrison Corporate Earnings Research Analyst gregory.harrison@thomsonreuters.com