I’ve owned shares of Microsoft (NASDAQ:MSFT) since 2011. Currently, the stock accounts for more than 15% of my portfolio.
I have great respect for the company’s management team, primarily CEO Satya Nadella, and believe that the company has significant long-term growth opportunities ahead in large addressable markets, along with sustainable competitive advantages that will enable their continued success. In addition, the company generates prodigious amounts of free cash flow (more than $100 billion in cumulative free cash flow over the past three years, despite significant capital expenditures). It also maintains a fortress-like balance sheet, with more than $60 billion in net cash that provides support for capital returns and inorganic opportunities.
That’s a long way of saying that I continue to feel very confident about the company’s financial position, its management team and its long-term prospects.
However, as is always the case in investing, the quality of Microsoft’s business isn’t the only consideration. Like a bettor at the race track, the question we must answer is whether the odds currently being offered are attractive. This requires a discussion on both the business prospects and the valuation. If an investment is made in this company’s shares for the next decade, is there reason to believe that the results will turn out good?
In an attempt to answer this question, I took a look at forecasts from the analyst team at Goldman Sachs (NYSE:GS). As of their most recent update, they have modeled results for Microsoft through fiscal 2027 (eight years from now).
In summary, their model assumes that revenues will increase at a roughly 10% compounded annual growth rate, that operating income will increase at a roughly 13% CAGR (with margins climbing by more than 100 basis points per year on average) and that earnings per share will increase at a roughly 14% CAGR.
Here’s what that looks like on an annual basis for Microsoft’s earnings per share.
Now, as we think about those expectations, a discussion on base rates is warranted (major shout-out to Michael Mauboussin, whose research on this topic is unparalleled). This is essentially the idea that we can look at similar situations in the past to derive some insight on reasonable expectations for the future. As Mauboussin notes, this exercise allows an investor to improve the quality of his or her forecast, as well as to provide a reality check on the claims and assumptions of others.
As an example, here are the base rates for sales growth delivered by companies with more than $50 billion in inflation-adjusted revenues from 1950 to 2015.
As shown above in the far right column, less than 5% of companies that have started with an inflation adjusted base of $50 billion or more in revenues have ultimately delivered a CAGR of 10% or more. That says something about how difficult it can be to sustain outsized top-line growth for long periods of time once you’ve reached a certain scale.
Microsoft’s revenues last year were $126 billion. Anybody that accepts Goldman’s revenue assumptions is putting more weight on Microsoft’s recent results than they are on the historic experience of companies like it (at least in terms of size, which I think you can argue offers some value as a relevant reference class). At the current valuation, Mr. Market is expecting future results that are quite a bit better than what companies of a similar magnitude have ultimately delivered over the ensuing 5-10 years in past occurrences. Only time will tell if that was an accurate belief.
Let’s assume we think those forecasts are reasonable. At the current price of $187 per share, Microsoft trades at 13 – 14 times the 2027 EPS estimate from Goldman’s analysts. At different terminal multiples, here’s an expected price return CAGR from owning the stock for the next eight years (the current dividend is a quarterly payout of $0.51 per share, good for a yield of 1.1%).
As you can see, a terminal price-earnings multiple in the low-to-mid 20’s will result in a mid-single digit price CAGR for the stock over the next eight years. The terminal multiple would need to be around 30 times trailing earnings for the annualized return to climb past 10% per annum.
Here’s what the math looks like if the earnings per share that Microsoft reports in the terminal period (fiscal 2027) is ultimately 25% lower or higher than what Goldman’s analysts expect.
In the scenario where the results fall 25% short of Goldman’s expectations, which still implies a ten year earnings per share CAGR of roughly 12% from fiscal 2017 to fiscal 2027, you need a terminal multiple in the high-20’s (at least) to generate a decent absolute return. I define that as a mid-to-high single digit price CAGR, which sounds reasonable to me given the current interest rate environment.
If the results are ultimately 25% higher than Goldman expects, you can generate attractive absolute returns if the terminal multiple is anywhere around 20 times earnings (in that scenario, Microsoft’s trailing 10-year EPS CAGR would be roughly 18%).
<p class="canvas-atom canvas-text Mb(1.0em) Mb(0)–sm Mt(0.8em)–sm" type="text" content="Conclusion” data-reactid=”74″>Conclusion
The question, in my mind, is how to think about terminal multiples. The more I think about this topic, the more I essentially feel torn between two schools of thought.
<p class="canvas-atom canvas-text Mb(1.0em) Mb(0)–sm Mt(0.8em)–sm" type="text" content="The first comes from Charlie Munger (Trades, Portfolio):” data-reactid=”76″>The first comes from Charlie Munger (Trades, Portfolio):
“Over the long term, it’s hard for a stock to earn a much better return than the business which underlies it earns. If the business earns 6% on capital over 40 years and you hold it for that 40 years, you’re not going to make much different than a 6% return – even if you originally buy it at a huge discount. Conversely, if a business earns 18% on capital over 20 or 30 years, even if you pay an expensive looking price, you’ll end up with a fine result.”
The idea here is that the truly great businesses are worth substantially more than an average business. When you think you’ve found one, do what you can to keep it. You shouldn’t sell because the multiple went up to 25x, 30x or even 35x. At present, Microsoft certainly seems to fit that description (in my opinion, it may be the epitome of that description).
<p class="canvas-atom canvas-text Mb(1.0em) Mb(0)–sm Mt(0.8em)–sm" type="text" content="The second view comes from Bill Nygren (Trades, Portfolio) of the Oakmark Funds:” data-reactid=”81″>The second view comes from Bill Nygren (Trades, Portfolio) of the Oakmark Funds:
“For almost all businesses, our crystal ball goes dark after seven years, so we assume all businesses trade at similar P/E’s after seven years. With an estimate of fair value seven years in the future, we can discount that back at an appropriate risk-adjusted discount rate to estimate today’s value. Whether that results in a near infinite or a negative P/E on current earnings is not of concern to us.”
His perspective is that it is difficult to comment on the quality of a business seven-plus years out. I agree: the idea that you can accurately forecast where most businesses will be a decade from now seems optimistic – and Microsoft’s own history over the past twenty years bears that out. The company was seemingly on top of the world at the turn of the century, then became a “dinosaur” that many believed was in a state of terminal decline, before returning to best-in-class status today.
That evolution doesn’t strike me as particularly odd; I do not find it unusual that any company, especially in the tech industry, will experience such swings over the course of a few decades. To me, the idea that you can predict what will happen to a company five-plus years from now – let alone 20 years from now – requires some hubris.
When I discussed this Nygren quote in a previous article, I wrote:
“Personally, I’m on the fence on this one – it seems a bit draconian, but I can see the merit of the approach Nygren has adopted. When I build models, I rarely stray from a level near the historic market multiple, especially when I’m applying that against an estimate of normalized profits. As with most things in investing, this is a balancing act that seems to be more art than science.”
I still think that’s a reasonable way to think about it. I think the “right” answer is a blend of Munger and Nygren.
A great business like Microsoft that appears to have substantial long-term growth prospects may deserve some premium to the average company in terms of what you’d be willing to pay for a dollar of earnings or free cash flow 5-10 years from now, but it’s tough for me to defend something meaningfully higher than average. The reason why is because it’s too hard to say that any company will maintain its greatness decades into the future. The opportunity to bet on a unique individual / group of people may be an exception, but that’s a topic for another day.
So, what’s the definition of “some premium?” Well, I don’t think it’s 100% higher than the market multiple, but I don’t think it’s 10% higher either (again, I mean this in terms of the multiple applied to an EPS estimate five to seven years down the road). If you accept the Goldman’s estimates outlined above (which I find plausible), apply a trailing terminal multiple of 20 – 30 times in fiscal 2027 and require an 8% price CAGR over the forecast period, the output is a fair value estimate of around $150 to $225 per share.
Given those numbers, especially in consideration of the position size in my portfolio, it will be tough to sit idly by if the stock continues to run. I love to own great businesses for the long-term, but the idea that one can disregard the price they’re being asked to pay is a bridge too far for me. There must be some limit. For that reason, if Microsoft continues to climb towards the high end of that range in the relatively near future, it’s likely I will trim my position.
Disclosure: Long MSFT
Read more here:
- Yelp: Fighting to Live Up to Expectations
- Fox: Searching for a Path Forward
- A Stellar Year for Chipotle – And High Hopes for 2020
Not a Premium Member of GuruFocus? Sign up for a free 7-day trial here.
<p class="canvas-atom canvas-text Mb(1.0em) Mb(0)–sm Mt(0.8em)–sm" type="text" content="This article first appeared on GuruFocus.
” data-reactid=”103″>This article first appeared on GuruFocus.
Add Comment