Forecasting Future Earnings Is The Key To Successful Stock Investing: Analyst Estimates Are A Starting Point

Includes: PCLN, SO, SYK, UTX
by: Chuck Carnevale

Looking for good dividend-paying stocks to add to your retirement portfolio? You might want to try trusting analysts' estimates to aid in your selection process.

Putting Analysts' Estimates Into Perspective

I believe that estimating future earnings growth is a major key to successful long-term stock investing. If you're a true investor, then you are actually investing in the business. Consequently, the success of the business that you invest in is going to be the primary determinant of how much money you can expect to earn on that investment. Stated more directly, when you invest in a stock you are buying its future earnings potential.

To my way of thinking, the only logical reason I would ever want to own any stock (business) is because I believe that the company is a profitable enterprise. But not just in the past, or the present, but most importantly, in the future. After all, and as I previously stated, future profits will be the source of any long-term return expectations I might have. Moreover, the growth of those profits will be a primary contributor to the total annualized return I can expect the investment to produce on my behalf.

Therefore, I believe as investors, we cannot escape the obligation to forecast future earnings, because our results depend on it. Furthermore, we should not guess, nor should we simply play hunches. Instead, we must attempt to calculate reasonable probabilities based on all the factual information that we can assemble. Then we should apply analytical methods based upon our earnings driven rationale that provide us reasons to believe that the relationships producing earnings growth will persist in the future. In other words, we must strive to forecast future earnings as accurately as we possibly can. On the other hand, we should simultaneously realize that perfection is not to be expected.

As an aside, there are many who criticize or even claim that we should eschew utilizing forward earnings forecasts when trying to determine fair value, or even when trying to decide what stock to own. I find these positions rather bizarre. I cannot think of any logical reason why anyone would invest in a business, unless they had a reasonable expectation of that business's ability to generate future profits. Since I am confident that both capital appreciation and dividend income will be a function of the company's future earnings power, estimating future earnings must be the essential element of long-term success.

The Selection Dilemma

But here is the dilemma. With all the thousands of companies to choose from, how can I forecast future earnings accurately enough in order to pick the stocks that might best meet my goals and objectives? I believe the obvious answer is by initially relying upon the consensus estimates of leading analysts following a given company. These estimates are readily available and provided by earnings estimate aggregators such as Standard & Poor's Capital IQ, Zacks, Thomson Reuters IBES (Institutional Brokers Estimator Service) and others. Moreover, earnings estimates can be found on most major financial websites and blogs.

Admittedly, consensus estimates may not be perfect; in fact, I would almost guarantee that many estimates will be wrong. On the other hand, I also believe that consensus estimates are generally accurate enough to be of value, especially for stock screening purposes. And more importantly, the closer to current time the estimate that you're relying on is (this year or next year's estimate), the more likely it is that they will be accurate enough to be of value.

At this point, I am sure many readers skeptical of analysts' estimates will cite numerous studies by academics suggesting that analysts' estimates are not accurate enough to be of value. Perhaps the most famous study on the accuracy of earnings estimates is the McKinsey study. Frankly, I have read the McKinsey study, and the majority of other similar studies, and found that they provide little evidence to deter me from relying on, and benefiting from analysts' estimates.

For example, one prominent study suggests that analysts' estimates are often off by factors of 12% or more. Later in this report, I will illustrate why I believe that margins of error at that magnitude are not really relevant deterrents against relying on analysts' estimates. This is especially true if forward earnings estimates are utilized and viewed correctly.

For more color on the relevance and importance of incorporating forward earnings estimates into your investing toolbox, I refer readers to the following article written by one of my favorite financial writers Jeff Miller. The article was written on October of 2010, but the principles that Jeff articulates are timeless in my opinion. Jeff also wrote a prior article that I feel is worth reviewing.

Consensus Earnings Estimates Accuracy

Perhaps most importantly, we must ask ourselves the question: just how accurate do analysts' estimates need to be to be of real value? I believe the answer to this important question is - within a reasonable range of probability. Since forecasting is all about the future, and much of the future is an unknown, we must accept the fact that estimates will contain a level of imprecision. Therefore, we should expect discrepancies to manifest when our forecast eventually turns to actual reality (reported earnings).

Furthermore, I believe it would be naïve to expect analysts' estimates, or even consensus estimates, to be perfect. There are a lot of unknown variables in the future that could affect the ultimate results. Furthermore, many companies, especially multinationals with numerous divisions and diverse businesses, are complex enterprises with a lot of moving parts. Therefore, and once again, I believe that the best that a rational person can hope for is that estimates fall within a reasonable range of probability and accuracy.

Moreover, I have always believed in running the numbers out to their logical conclusions. In other words, let's calculate and think through what an earnings miss might actually mean in numerical terms. For example, a very common miss that the media seems to take great glee and relish in reporting, is a plus or minus by a penny. In other words, XYZ Company reported earnings today that missed analysts' expectations by one cent. We've all seen the stock price of the company reporting such a miss often fall by ridiculous percentages of 5%, 10%, or more.

However, let's do a simple calculation and apply some basic logic to what a one penny miss might actually mean. If we assume that the normal P/E ratio of the average company is 15, then a company that delivered one penny less than expected should only have a market value that is $.15 less than what our original estimate indicated. For a very fast growing company, let's say one growing by 20%, a one penny miss might require a discount of $.20 (i.e. P/E 20 x one penny). My point being; that a one penny earnings miss does not really amount to very much in the long-term scheme of things. Moreover, consider that this same math applies whether the earnings miss is plus a penny, or minus a penny.

As an aside, based on my own, albeit anecdotal experience, there tend to be more earnings misses on the plus side than there are on the negative side. I believe this is because a significant portion of the ultimate estimate that an analyst makes is significantly based on guidance from the company itself. I believe that prudent (smart) management teams are more likely to guide lower and therefore exceed expectations, than the other way around. This presents the argument that relying on the forward earnings estimates of analysts may be a very conservative way to base stock investing decisions on.

What If Calculations - Running The Numbers To Their Logical Conclusions

Next I will run through a few sample companies with varying historical growth rates in order to illustrate the magnitude, or lack thereof, of what an earnings miss truly represents. The companies I'm going to utilize in this example are only offered because they represent low growth, moderate growth and fast growth examples. To be clear, I am not recommending or building a case for investment, or against investment, in any of these samples. Instead, I am utilizing them to provide some mathematical realities of what earnings estimates are actually all about.

Stryker Corp. (NYSE:SYK)

A good example of a company with a history of beating consensus earnings estimates is Stryker Corp. The following graphics are provided courtesy of MSN Money reporting earnings estimates on Stryker provided to them by Zacks.

I believe the most important benefit that earnings estimates provide is a perspective of whether a company is going to grow in the future or not, as well as an idea of what magnitude that growth may fall into. Although it would be wonderful to know precisely what that growth would be, it's more important, as I will illustrate later, that earnings actually grow rather than shrink.

According to the consensus of approximately 16 to 18 analysts reporting to Zacks, the estimated earnings growth rate for Stryker over the next 5 years is 8.5%. Later, I will present the consensus of 29 analysts reporting to Standard & Poor's Capital IQ reflecting their expectation that Stryker will grow earnings over the next 5 years at 9.1% per annum.

My goal is to illustrate that the 7% magnitude of difference (8.5% versus 9.1%) is really not material. In other words, Stryker would be an attractive investment if either one of those milestones were achieved. Obviously, Stryker is a better investment at 9.1% growth than at 8.5% growth. However, both of these 5-year growth rates would be acceptable.

At this juncture, the reader might recall that I hypothesized that near earnings are likely to be more accurate, and perhaps more meaningful, than long-term estimates like the five-year growth rate estimates presented above. Therefore, this next graph, again provided courtesy of MSN Money based on the consensus reported by Zacks, provides closer estimates. I am focusing on fiscal year-ends, December 2013 and December 2014 of $4.33 and $4.70 respectively. However, estimates are also provided for the next couple of quarters on the table under the graph.

Next I'm going to turn to the Estimated Earnings and Return Calculator courtesy of F.A.S.T. Graphs™ to provide a second opinion, as well as the opportunity to run a few varying what-if scenarios. The consensus of 29 analysts reporting to Standard & Poor's Capital IQ expects Stryker to grow earnings at a rate of 9.1% over the next five years.

Furthermore, they expect fiscal 2013 earnings of $4.33 (identical to Zacks), and $4.71 for fiscal 2014 earnings, or one penny more than Zacks. Therefore, regarding the next two years, there is a high consensus among the two earnings aggregators, Zacks and Standard & Poor's Capital IQ. However, remember there is a 7% magnitude difference (9.1% is 7% higher than 8.5%) between the two five-year estimated earnings growth rates.

When running these numbers to their logical conclusion through the Estimated Earnings and Return Calculator, calculates an expected five-year total annualized return, including dividends, of 8.6 % per annum.

However, what if Zacks' analysts were correct, and Stryker Corp only grew at 8.5%, and if earnings in fiscal 2014 were only $4.70 instead of $4.71? The five-year estimated total annual return only dropped to 8.2% from 8.6%.

Now when you consider that both of these numbers are more likely than not to be moderately imprecise, you should start getting a perspective on why estimates do not need to be exact. However, they do need to be within a reasonable range of accuracy, and frankly, they usually are.

Southern Company (NYSE:SO)

Next, let's run a slow growth example, Southern Company, through similar what-if scenarios. I will utilize the same graphs from the same sources in all following examples as I did with my first example. Zacks forecasts that Southern Company will grow earnings at 5% per annum rate over the next five years.

Zacks also reports that consensus' expectations of earnings per share for fiscal 2013 and 2014 are $2.75 and $2.92 respectively.

According to 15 analysts reporting to Standard & Poor's Capital IQ the following F.A.S.T. Graphs™ Estimated Earnings and Return Calculator expresses the same 5% five-year earnings growth rate but only $2.90 earnings per share for fiscal 2014. This calculates out to an expected five-year annualized total return, including dividends, of 6.8%.

In contrast, utilizing the override feature of the Estimated Earnings and Return Calculator, we discovered that calculations based on Zacks' estimates have very little effect on the estimated total annualized rate of return. In this example, the only differences are the $2.90 earnings per share estimate for 2014 by Capital IQ versus $2.92 earnings per share for 2014 by Zacks. In other words, the difference is immaterial.

The remaining examples, United Technologies Corp. and Inc. run the same comparisons as previously presented based on estimates provided by Standard & Poor's capital IQ versus estimates provided by Zacks. The Estimated Earnings and Return Calculator graphs reflect the minor difference that these variations in the estimates produced regarding future potential rates of return.

United Technologies (NYSE:UTX)

The following graphs, courtesy of MSN Money, reflect the consensus' earnings estimates from Zacks.

The first Estimated Earnings and Return Calculator graphs below reflects the consensus' estimates provided by Standard & Poor's Capital IQ. The next graph represents an override utilizing the consensus' estimates provided courtesy of MSN Money with Zacks' data.

Standard & Poor's Capital Estimates

MSN Money Consensus' Estimates By Zacks Inc. (NASDAQ:PCLN)

The following graphs, courtesy of MSN Money, reflect the consensus earnings estimates from Zacks.

The first Estimated Earnings and Return Calculator graphs below reflects the consensus' estimates provided by Standard & Poor's Capital IQ. The next graph represents an override utilizing the consensus' estimates provided courtesy of MSN Money with Zacks' data.

Standard & Poor's Capital Estimates

MSN Money Consensus' Estimates By Zacks

For additional color on earnings estimates, the Wall Street Journal provides an interesting analysis regarding the accuracy of earnings estimates on 150 major companies. Follow this link, click the prompt by company and you can look at any or all of the 150 companies and their earnings estimates versus actual.

Summary and Conclusions

Regarding forward earnings estimates, there are a few simple points that I would like to elaborate on. First of all, forecasting future earnings is certainly simpler and more reliable than trying to forecast stock prices or stock markets. A quote from one of my favorite financial legends, Marty Whitman, Chairman of the Board, Third Avenue Value Fund, speaks to this point:

I remain impressed with how much easier it is for us, and everybody else who has modicum of training, to determine what a business is worth, and what the dynamics of the business might be, compared with estimating the prices at which a non-arbitrage security will sell in near-term markets.

Second, I believe that analysts' estimates tend to be under-estimated more often than over-estimated. Furthermore, I believe the reason for this is that analysts derive a majority of their estimate based on guidance from the management of the companies they are providing estimates for. Common sense says that companies are more prone to under-estimate their guidance so they can beat consensus and see their stock price rise, than to miss estimates and see their price fall.

Additionally, estimates need only fall within a reasonable range of accuracy in order to be of great value regarding making long-term investment decisions. As I illustrated with my examples above, reasonable deviations will not really have a great enough impact that is strong enough to alter an investment decision. In this context, estimates should be used as guides, and I suggest that investors always have, and calculate a best case, moderate case and worst case scenario.

But perhaps most importantly of all, investors should consider all of the available estimates that analysts provide. However, their greater emphasis should be placed upon near-term estimates over the longer-term estimates. And, it is imperative that estimates are continuously monitored and updated. There is never a substitute for comprehensive due diligence.

But with all the above said, analysts' estimates provide an important metric that investors can utilize in order to make sound and smart long-term stock investing decisions. They will rarely be perfect, but in most cases they will be accurate enough to be a useful barometer that investors can rely upon to make reasonable long-term stock investment decisions.

Disclosure: Long UTX & SYK at the time of writing.

Disclaimer: The opinions in this document are for informational and educational purposes only and should not be construed as a recommendation to buy or sell the stocks mentioned or to solicit transactions or clients. Past performance of the companies discussed may not continue and the companies may not achieve the earnings growth as predicted. The information in this document is believed to be accurate, but under no circumstances should a person act upon the information contained within. We do not recommend that anyone act upon any investment information without first consulting an investment advisor as to the suitability of such investments for his specific situation.

Disclosure: I am long UTX, SYK. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.