On September 19th of 2016, a Wells Fargo (NYSE:WFC) analyst suggested that Maryland-based Aerospace, Defense and Technology company Lockheed Martin (NYSE:LMT) could generate $20 in earnings-per-share by 2020. The analyst pointed to organic growth, reduced share count, higher U.S. defense spending, F-35 growth and pension income as tailwinds for the company.
In addition, the analyst suggested that shares could be worth $300 in the coming year. The reasoning was logical enough. Here's how that particular number was calculated.
If Lockheed is to earn $20 by 2020, you can discount that value back in time. Using the company's cost of capital (they use 7.8%) this means that your discount rate would be about 1.25 (1.078^3). In other words, the present value of that $20 in 2020 EPS would be about $16 by 2017. Based on the consensus earnings multiple (around 19) that equates to a future share price just above $300.
It's logical, and comes about in a nice, clean fashion. Of course the investment world isn't quite so tidy. Just because you suppose something could happen does not automatically make it so. In the above example, any number of things could be drastically different as a reality. Earnings may not be exactly $20 - instead it could be $20.10 or $19.90, $18 or $22, $16 or $24, or… well you get the idea. Earnings estimates are simply that: guesses about the future that are overwhelmingly likely to be wrong.
Next up you have the discount rate to use. Seven point eight percent might be reasonable, and granted you need some sort of proxy, but here again there's a rational argument to be made for 7% or 9% or something a bit different. And these seemingly small changes can have large impacts.
Finally, the earnings multiple is perhaps the most finicky of the three aspects. Even if you get the future earnings exactly spot on, there's no mechanism that requires shares to trade at $300 instead of $280 or $320.
In other words, be cautious about expectations down to the exact dollar. It's absolutely necessary to create a baseline. However, without simultaneously acknowledging the wide range of potential error, you can set yourself up for future disappointment.
With that in mind, let's think about three separate scenarios for Lockheed Martin's future. In the first case, we'll suppose that the analyst gets the future earnings exactly correct: $20 in EPS by 2020.
Shares of Lockheed Martin presently trade with an earnings multiple north of 20. Should this ratio continue, $20 in EPS would imply a future price of about $413. In addition, you would collect dividends along the way. Based on the current $1.65 quarterly dividend, Lockheed's payout ratio sits around 55%. Should this ratio continue you would also anticipate receiving $37 or so in cash dividends per share.
Collectively that equates to a total anticipated value of about $450. As compared to the current price near $241, this would imply an average compound total gain of over 15% per annum. Clearly this (and anything better) would be a rather impressive result.
Then again, it's conceivable that the current valuation multiple is so high as a result of the solid anticipated growth prospects. For the second scenario, let's scale things back a bit.
Twenty dollars in EPS by 2020 is certainly possible, but I've also seen much more subdued expectations. Should Lockheed "only" earn say $17.50 by 2020, your results are naturally going to be lower. For one thing, if you keep the payout ratio constant, you might only anticipate receiving $33.50 in per share dividends in the coming four years.
Over the past two decades Lockheed has traded with an average earnings multiple of about 16. In the last decade this number has been closer to 13. Only in the last couple of years has something above 16 been a common occurrence. Should shares of Lockheed trade at say 16 times earnings, this would imply a future price of about $280.
That's a total value of about $313.50 including dividends, or an average compound gain of roughly 6.4% per annum. This would still build wealth, but a more conservative view provides a drastically different outcome.
Finally, we can look at a third scenario showing even more caution. For instance, should Lockheed "only" generate $15 in per share earnings by 2020 (still well above today's mark) and trade at say 14 times earnings, suddenly you're looking at the potential for nonexistent gains (even including dividends).
Of course this is just three of an infinite number of possibilities. Here's a look at a much wider range, showing 80 different alternatives:
The top row shows earnings-per-share by 2020 for Lockheed Martin. The left-hand column details various P/E ratios ranging from 11 to 20. The intersection of each metric represents the annualized rate of return (based on 4.25 years) that these assumptions would represent. The dividend is included and grows in line with earnings in this hypothetical example.
The three yellow highlighted cells roughly indicate the three scenarios that I outlined above. Moving from the lower right-hand corner to the middle-left, you have the robust scenario followed by the two more cautious approaches.
This sort of table allows you to complete various scenario analyses immediately. If you suspect that Lockheed can earn say $16 per share by 2020 and ought to trade at perhaps 17 times earnings, you can instantly see that these suppositions would translate to an assumption of 6% annual returns; keeping in mind that this is merely a baseline.
In viewing the table, a couple of things jump out to me. Notably, the values are somewhat skewed to the downside. This was done intentionally.
If the analyst's presumptions come to fruition and Lockheed earns $20 per share by 2020 (or even $19) and later trades at 18 or 20 times earnings, you're looking at the prospect for double-digit returns. It's just the way the math works out. And naturally if the results were a bit better - say earning $21 per share or trading at 22 times earnings - your anticipated returns would be even greater.
In this light, I don't find it useful to extrapolate to the sky. I could have included say $25 in earnings and a P/E of 30, but this doesn't add much insight - we already know the returns would be solid prior to getting to that mark.
The second thing to note is the wide range of values represented. An analyst is going to give you a very specific, down to the dollar, expectation but in reality you have to be cognizant of the many possibilities. In other words, I find it prudent to both come up with a baseline and recognize the great degree of possible fluctuations.
In short, if $20 in EPS for Lockheed Martin comes about, that's great and could very well mean solid returns for today's owner. If the company comes up a bit short of that expectation, something in the mid-single digits as a potential return assumption appears more prudent. And if the company only improves slightly, the returns are apt to be unimpressive. The takeaway is to not accept a given number at face value and instead create a range of scenarios that would better help you deal with the future fluctuations that are sure to come about.
Disclosure: I am/we are long WFC.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.