It speaks volumes that in terms of the peak dot-com era valuations Microsoft at 80x earnings was probably one of the more reasonable ones. There aren’t many examples of technology companies from the late Nineties that have delivered your money back in one piece, let alone generated a positive return.
For every Apple, Amazon and Google that have gone on to flourish over the past fifteen years there was a Pets.com or Webvan that spectacularly blew up and disappeared into nothingness along with shareholder’s cash. Even the mighty Cisco, a company that can generate billions of dollars in net income every year, is still only sitting at $29 a share compared to $80 at peak internet mania. That’s capital destruction of 60% with a stock that is capable of generating 20% of its revenue in after tax net profits.
The crazy valuations weren’t just limited to the technology sector either, as the mid to late Nineties bull market managed to send shares in good old Coca-Cola to almost 40x earnings. The whole era is probably the perfect example of Jeremy Siegel’s growth trap: the idea that investors are too often drawn to chasing growth versus market exactions of growth. It’s an observation he uses to rationalize how the best twenty performers between 1953 and 2003 came from relatively stable and non-disruptive sectors. Not only were they disproportionately represented by things like household goods, healthcare, tobacco and food & beverage stocks, but there was a distinct lack of stocks from the more innovative sectors of the economy.
Finance and technology for example are probably the two biggest industries that have emerged in developed economies over the last six or seven decades, and as portion of the economy those two sectors have probably grown the most over that time frame. The thing is that Siegel’s top twenty list didn’t contain a single technology or finance sector stock. The growth trap is his way of explaining that growth in stocks from these sectors was largely baked into their valuations, whereas investors tended to underestimate the earnings growth capability of established stocks in the more boring sectors.
Nowhere is Siegel’s theme more apparent in the market of the late 1990’s. This was a market where people were regularly paying triple digit multiples of annual earnings. Buying a share of Cisco for example meant you would be paying $175 for every $1 of actual net profit. For Microsoft the comparable figure was $80. Meanwhile you could quite happily pick up shares in Imperial Tobacco or British Amierican Tobacco at a 9% earnings yield with a 7% dividend. Never mind the insane rate of compounding that would generate over a decade or two – somewhere around 15% a year as it turned out – all the craze was the future growth of the internet and dot-com stocks.
Microsoft: Profit And Free Cash Flow
It’s actually a testament to how high quality Microsoft’s earnings are that it has eventually managed to shake off its post dot-com mega hangover. So far most of the stock studies have focused on companies with huge brand equity: stocks from the boring sectors like beverages and tobacco.
With Microsoft what you have is just an old fashioned monopoly. The fact is that using a computer in the Nineties and the early 2000’s meant you were almost guaranteed to be using their products. It could have been the iconic Windows operating system or Microsoft Office, it could have been a home computer or a work machine. It didn’t matter. The company had an iron grip and insurmountable market share of the profits in PC software.
It’s been a bit of a turbulent time for Microsoft recently as they are still dealing with the gradual shift towards mobile devices and away from desktop. Accounting for the associated goodwill and impairment charges relating to Phone Hardware and the acquisition of the Nokia phone business, 2015 earnings per share was $2.63. At the end of the first year in the most recent five year period in 2011 Microsoft generated earnings per share of $2.69. So it’s clear that there has been some adjustments going on with the core business as trends have shifted towards mobiles and the cloud.
Despite that the company remains a tremendous profit machine. Adjusted net income of $22bn on revenues of $93.5bn is pretty impressive during a rough patch. The balance sheet has accumulated an extra $30bn in net cash since 2010 which translates to about $3.70 per share.
Now compare that to a January 2010 share price of $30. In the following five years the company has accumulated 10% of it’s starting value in net cash on the balance sheet – that’s after paying out all the dividends, buybacks and capital spending required to maintain and enhance its operations. The core business is just a mega cash cow.
Since their underlying business is so incredibly capital light free cash flow generation is remarkably strong as well. On average Microsoft are able to bring in about $25bn in free cash flow every year, representing margin of about 30% of annual revenues, and about 25% on total invested capital. If this was Hershey or Coca-Cola you can pretty much guarantee that it would be trading at a perpetual premium to the wider market.
In the case of the big tech stocks though you have an ingrained level of uncertainty. It’s why you can pick Apple up for a 9% earnings yield, a company that last year made $53bn in after tax net profit on $233bn worth of sales. Likewise with IBM and Cisco trading for earnings multiples of 11x and 12.5x, with dividend yields of 3.30% and 3% respectively. They’re expected to struggle from this point on with the constant ever lingering threat of technological disruption. IBM has already had to reinvent itself a few times over the last century for example.
Microsoft Stock Returns Since The Dot Com
Picking up Microsoft stock in 2010 meant you were getting a pretty good deal. In return for a price of $30.50 a share you would be getting $2.12 in net income, a $0.52 dividend and $3.30 per share in net cash on the balance sheet.
Paying just over $14 for every $1 of after tax profit for such an incredible profit machine, one with a near monopoly on personal computing, is good old fashioned blue chip value investing. It doesn’t matter if the stock goes nowhere for a few years because there is a good expectation that paying that kind of price will look like a great deal in even five or ten years time. As it turns out the stock still went sideways for three years, although you would have collected about $2.30 in dividends for the wait.
Now compare that to the peak during the dot-com mania. At the start of 2000 Microsoft was trading at a split adjusted price of nearly $59 a share. For the entire year the company would go on to earn $0.85 a share in net income. The next time Microsoft stock would see a closing price of anywhere near that 60x forward earnings level would be nearly sixteen years later when the shares were trading at about $55 a share.
Between then and now the company have grown their earnings at an annual rate of 7.30% compounded, yet the madness of paying such a huge premium to own a share of the firm’s profits will have wiped out any capital appreciation between those two dates. That’s despite the earnings growth, despite the fact that Microsoft have retired 20% of the float in that time frame and despite the fact that the quality of the company’s earnings has remained largely intact. The ultimate example of business doing great, stock doing badly – all down to an extreme example of Siegel’s growth trap.
Aside from seeing your capital split in half during much of the 2000s as the value multiple began to contract, you would also have received dividend payments from fiscal year 2003. Over these last twelve years they would have come to $11.67 in total per share.
From the $59 you paid in 2000 you spent three years in the red before Microsoft began distributing a share of the profits for which they had no other productive use. As of the end of this week those payments would have returned $11.67 back into your pocket (including a very nice special dividend of $3 paid out in fiscal year 2005).
Adding the cumulative dividends to your stagnant share price action gives a total return of just 0.5% a year. It’s lousy, but that $11.69 in dividends is actually a strong sign of how well the underlying business has performed since then. Many names from that era don’t even exist any longer.
The ironic part is that the scars from that bubble meant that dead money label stuck with stock at precisely the time it was getting cheap again. You open up the share price chart and you see that it continuously goes nowhere; at $27 you could buy it in 2007 for the same price that you could in 2001. The difference is that you were only paying $19 for a buck of Microsoft profit rather than the $31 you were paying in 2001.
The label still sticks for the next years, going sideways, before taking off in 2013. You collect $3.70 in dividends while you wait – about 15% of your capital outlay. This is the main point about long-term investing: you can’t predict when the price will recover, and when it will begin to resemble the future prospects of the business again. In the case of Microsoft it took fifteen years for the stock to catch up to all the growth that had been built in during the bubble years. Even if you bought the stock just six months after the peak you would have paid 50% less for the shares. Same Microsoft, same business, same profits and the same growth. The difference though is equal to 5% annual returns versus 0.5% returns. On a $10,000 investment it would have added up to an extra $5,000.
It’s hard to say what the future for the shares look like. What will desktop and mobile computing look like in five or ten years time? It’s a disruptive sector to invest in. But paying 18x earnings in 2016 is likely to leave you a lot better off than 80x earnings did back in 2000.