Photo: Daniel Borman via Flickr
Over the years, I have written several times about the concept of “optimal cash balances” in technology companies. This 2007 post considering Apple’s cash position provides a good overview. The analysis involves studying the variability of operating cash flows, projected R&D expenditures and quantifying the insurance cost of holding cash on hand. However, an even more fundamental issue has to do with public market perception and the investment community’s perspective of growth versus value. And one of the key determinants of which bucket a company is placed has to do with dividend policy.In general, a company with modest growth expectations and a significant dividend will be treated as a “value” stock, while a company with more aggressive growth prospects and pays little or no dividend will be deemed a “growth” stock. These labels fit pretty well when growth was more of a linear function, where the shift to value occurred when linear growth slowed and the business became a cash generator able to support the payment of large dividends. But the massively scalable, technology-enabled business models of today sometimes exhibit sharp non-linear growth together with massive cash generation that renders the demarcation among growth, value and dividend policy murky and confusing.
Consider the two ends of the continuum: oil producers and web-enabled technology companies at the steepest part of their growth ramp. Absent exploration activities, oil producers control a known amount of inventory, can estimate the rate at which this inventory is monetized and have a good handle on the costs of extraction and distribution. They are, in effect, depleting annuities that can be valued with a high degree of precision. Reinvestment is necessary to keep plant and equipment sufficiently current to achieve operating objectives, but is a small fraction of the cash flows generated from oil sales. The difference between the two is largely paid out as dividends. Stable dividends that will eventually decline along with production.
Now consider the rapidly scaling web tech company. It is growing by double-digit percentages monthly, and every penny of revenue and then some is being invested in infrastructure, sales, support and operations. Growth is stratospheric but the range of potential outcomes is vast. Volatility of future cash flows is extremely high, but the rate of growth makes investment in the business a massively NPV-positive project. No dividends are expected to be paid for years, but when the business achieves scale it is expected to be a massive cash generator with highly attractive margins.
Large technology companies often have features of both of these kinds of businesses as well as everything in between. Consider Microsoft. Is the OS business kind of like an oil well? Gushing profits and cash flow, but with poor prospects for growth? What about its Home & Entertainment business? Revenues growing rapidly but hemorrhaging cash. Google AdSense? Strong growth plus crazy cash flow. Android? Growing super fast but not yet moving the needle on cash flow. Apple iPad and iPhone? Super fast and kicking off cash. MacBook? Less rapid growth but generating cash. Bottom line: companies are made up of businesses with a wide variety of growth and cash flow generation characteristics. But the interesting thing is how the cash flows from high performers can be so significant that even though they are growing at a break-neck pace, they can still gush billions in free cash flow. This is what I think about when I imagine Facebook’s revenue engine at scale. Growth plus more cash than is necessary to support continued rapid growth. So how should the investing community deal with this new phenomenon?
Stop viewing dividends as an automatic signal that the best days of growth are in the past. This is a reflexive assumption that is increasingly being proven wrong. Take Apple and Google, for starters. Their businesses are growing rapidly with good prospects ahead, yet are generating so much free cash that its retention can only get management into trouble. Establish a healthy dividend that is super safe, and return additional non-essential cash to shareholders via stock buybacks. This way an analyst would value a business in two pieces: its dividend stream (a straight NPV calculation) and its operating business (using option maths to value the future probability-weighted cash flows and growth prospects). The operating business would be fully valued with an appropriate growth stock multiple, with the added benefit of its easy-to-value dividend stream. Investors should also favour this policy because of management’s discipline around capital investment, confident that the fanciful and wasteful projects undertaken by highly profitable companies afraid of losing their growth stock luster would be shelved.
If investors were really using their heads, they would look beyond dividend policy and to the true level of cash required to fuel rapid growth. As a shareholder, am I better off with Microsoft paying out an additional $100 billion in dividends or spending those dollars investing far afield and trying to reclaim their super-growth of the 1980s and 1990s? The answer is clear. Apple and Google are facing a similar question today. And for them the answer should be clear as well: resist senseless diversification and low-probability projects and pay out excess cash to shareholders. Improve transparency across business lines to give investors visibility into the revenue momentum of each business line. With transparency and sound financial policies premium market valuations will result. The issue isn’t whether they should adopt such policies, but whether they’ll have guts to do so.