With Microsoft (MSFT) closing at a 52-week high last week (~$47 per share), as well as a ten-year high, it’s interesting to remember that this is not the all-time high for the stock; in fact, it’s not even close. Let’s take a trip down memory lane and think about just how crazy valuations were during the tech bubble – and what that can teach us about sensible investing:
Microsoft
MSFT peaked in late 1999 / early 2000, at roughly $60 per share (split adjusted); at the time, the company had more than 10 billion diluted shares outstanding – good for a market capitalization in excess of $600 billion ($616.3B at its peak, and the world’s most valuable publicly traded company of all time – a record that was taken by Apple (AAPL) in August 2012). In the fiscal year ended June 30, 2000, Microsoft reported $23 billion in sales, for a price-to-sales multiple of 26X. By comparison, Salesforce.com (CRM) trades at a trailing price-to-sales multiple of 10X – and is currently at a revenue base that’s one-fifth as large as Microsoft’s was in fiscal 2000.
Moving down the line, Microsoft would report ~$11 billion in operating income in fiscal 2000; assuming a 30% effective tax rate (roughly what Microsoft was paying at the time after accounting for investment gains / losses and other one-off events), that leaves $7.7 billion in net income. Investors were paying more than 70x forward earnings for MSFT shares near the peak.
Using that as our starting point, what would be needed to justify buying / holding the stock?
An investor at the time looking out ten years may have assumed the terminal P/E would be 25x; to account for multiple contraction, earnings would need to have grown to $24 billion, or a compounded annual growth rate of ~8% over the fifteen years – and that’s to break even on the investment. As we can see with hindsight, both of those assumptions – particularly the multiple Mr. Market would be willing to pay for Microsoft common stock – have proven too optimistic.
What if you actually wanted to make some money on the initial purchase? Assuming a required annualized return of 7%, comparable to the yield on a 10-year treasury in late 1999, the market cap would need to increase to more than $1.6 trillion by the terminal year; assuming an (even more) optimistic 30x earnings multiple in the terminal period, earnings in 2014 would need to reach $53.3 billion – a 13.8% CAGR from the $7.7 billion (normalized) earned in fiscal 2000. That would require looking out fifteen years – at a time when most investors were likely hoping to double their money, or more, in the next fifteen months (remember that in the five years to December 31, 1999, Microsoft’s stock increased by more than 1400%).
Again, hindsight is revealing: the $53.3 billion figure required for this calculation was roughly double Microsoft’s reported operating income in fiscal 2014; while the reduction in shares out over the measurement period (~16%) makes this comparison a bit more palatable, the end result is still a very painful outcome for the optimistic equity holder from 2000.
A stock price that exceeded intrinsic value by a wide margin was borrowing returns from the future: in September 2014 – nearly fifteen years later – Microsoft common stock still trades ~20% lower than the price level on the last day of the 1990s. Including all dividends paid since the company initiated their payout in fiscal 2004 (including the special dividend), my math suggests that investors are still slightly in the red. Investors (if you can call them that) who were expecting a double in the next 15 months haven’t even got all their money back after 15 years.
A current example
I think sound investment decision-making, at its core, can be succinctly captured in a question:
“Are current opinions on the stock, as represented by the valuation, plausible?”
In periods like the late 1990s (and I would argue for a growing number of stocks currently), I think investors find themselves asking a slight variation of that question:
“Are current opinions on the stock, as represented by the valuation, possible?”
The first question requires thinking about numerous outcomes and their respective probabilities. Plausibility – a reasonable valuation – requires the consideration of all scenarios and their implications. The second question, on the other hand, is simply begging for a reason to believe; with the slightest possibility and creativity on the part of analysts, the party can march on indefinitely (there were surely analysts who still thought Microsoft was a buy at $600B+).
I think a decent current example is Tesla (TSLA): investors, assuming they demand 10% annualized returns, believe the company will be worth more than $90 billion by 2024; at 15% per annum, the terminal market cap jumps to more than $140 billion. If you ask people buying the stock today about their expectations for the coming decade, I would be very surprised to hear that they are targeting just 10% per annum.
Assuming a multiple of 20x in 2024 (compared to the current low teens multiple for Honda (HMC), Toyota (TM), General Motors (GM), Ford (F), and others), Tesla would need to earn ~$4.5 billion in net income (or more than the entire value Mr. Market attributed to the company two years ago), assuming the share count stops growing.
By comparison, the company had $2 billion in sales in the most recent fiscal year, with sales in the first six months of FY14 up 44%; assuming 10% net margins, which would be much higher than the reported margins for the four auto companies mentioned above, this would require $45 billion in sales a decade from now – a compounded annual growth rate of 36% in the coming decade. That’s not too far from the rates the company reported year to date; that will be harder to sustain off a $20 billion base than from $2 billion.
That possibility – a 10% annualized return – assumes things go pretty well over the coming decade, to say the least. What about if things don’t go as well as expected? What if a GM or a Ford, two companies that spent ~20x more (individually) on R&D as Tesla in the past year, are able to match the company’s innovativeness with time? What if someone like BYD can steal Tesla’s thunder? I’m not sure how to put those risks into probabilities, but I’m sure they’re much higher than zero. When Elon Musk says, “The stock price that we have is more than we have any right to deserve,” I think people should listen; with the stock price up another 50% since he said that last October (here), my conclusion is that they haven’t.
Conclusion
While the tech bubble may have passed fifteen years ago, many companies are still struggling to escape its long shadow (Cisco (CSCO), Oracle (ORCL), and Qualcomm (QCOM) are three names still below their all-time highs as well); investors in today’s high-flyers should study how things ultimately played out for their predecessors before becoming too complacent with those ever-climbing stock prices.
