Dividends are just a bribe to get you interested in slow growing companies who can’t be bothered to reinvest their earnings in something useful.”
—Andy Kessler, “I Still Hate Dividends, Professor Siegel,” 7 January 2008
However erroneous, the kind of sentiment expressed in the above quote has long been standard fare for growth- and momentum-oriented investors, who often ply their trade in the land of information technology. Dividends? Dividends? Those are for boring old food companies and electric utilities. Tech is about growth!
Of course, the bottom line for any investor—or any company—is not merely the rate of growth, but the total return realised by shareholders. In the aftermath of the late 1990s tech stock bubble, the industry’s giants entered a period of increasing maturity, with cyclical tendencies exposed and falling rates of revenue growth. Many retained highly profitable, well-established franchises capable of churning out incredible free cash flows. But like overgrown teenagers—as physically mature as they were likely to get—their intellectual maturity was somewhat lacking.
Hey, tech: Maturity is not a bad thing! If we roll the clock all the way back to the end of 1991—predating the bubble phase of tech sector investing by three years or more—the total return from the S&P 500 Information Technology sector has been beaten by the Dow Jones US Select Dividend Index. Tech won the battle for growth, with a compound rate of price appreciation of 9.7% a year to dividends’ 7.1%. Yet dividends still won the war: Their average yield of 4.4%, nearly 6 times the tech sector’s 0.7%, allowed the Select Dividend index to return an average of 11.5% a year, topping tech by a full percentage point.
With this and many other experiences in mind, it will probably come as no surprise that I’ve long held a grudge against the tech sector. Until very recently, the only tech stock involved in my dividend-focused portfolio at Morningstar was 
Microsoft (MSFT). I owned the stock between April 2005 and July 2007, and the experience remains instructive. I bought Microsoft on the premise that its dividend could double or even triple without hindering the company’s internal growth potential. But while the means may have been obvious, the motive was sorely lacking. Microsoft instead stuck with a series of mostly low-double-digit hikes that, thanks to an ultra-low dividend yield at the start, failed to drive much capital appreciation. Microsoft trades no higher today than it did when I sold it; all that’s changed is that shares that once yielded only 1.4% now pay 3.2%.
Yet it rarely pays to leave investment biases unchallenged forever, and even a cynic like me has to admit that things are getting a bit better. Just this year, Apple (AAPL) and Dell (DELL) started paying dividends, Cisco (CSCO) raised its dole 75% in only its second year as a payer, and even that foot-dragger Microsoft delivered another 15% hike. The sector as a whole still yields only 1.6%, but I estimate that the total dividend payout of the group is up 59% year-to-date (with Apple accounting for three fifths of the gain). Better yet, a handful of the biggest tech stocks in the US now yield more than the S&P 500 average.
Could it be that tech stocks have become investments that long-term income seekers can’t ignore? No, at least not yet. But one tech stock--Intel (INTC)--has demonstrated the combination of business fundamentals, capital allocation policies, and valuation to offer genuine, enduring appeal.
Where the Money Goes
To me, the ideal capital allocation and dividend policy for a company is rather simple. First, out of a company’s operating cash flow, it should reinvest enough internally to maintain its current earning power and competitive position, while taking advantages of any high-return opportunities to deploy additional capital within existing operations. Second, cyclical businesses—and almost all have at least some kind of cycle—should maintain a buffer so that dividends don’t become burdensome during temporary downturns. Third, the rest of the company’s ongoing cash generation should be devoted to regular cash dividends that grow with earnings. Debt, naturally, should be kept under control. As for acquisitions, anything larger than a bolt-on should be funded with equity—or not be made at all.
Despite one questionable mark on the last count—last year’s purchase of security software firm McAfee—I think Intel passes this test. A 42% payout ratio is generous but not too generous in the context of the industry’s cyclicality. Lately Intel has been spending north of $10 billion a year on expanding and improving its manufacturing capacity, but plenty of free cash flow is left, and—thanks to Intel’s wide economic moat (aka its competitive advantage)—it’s reasonable to assume it will lead to good dividend growth in the future.
So why do so many of Intel’s peers flunk this test? Software and services, being less cyclical and capital intensive than Intel, could actually provide even more generous payouts, but they choose not to.
Hurdles to Better Dividends
I think shareholders bear a certain part of the blame. Investors in other areas of the market have started to realise both the importance and the practicality of
dividends, but in tech, too many investors (individuals and professionals) are still trying to party like it’s 1999. Dividends carry that stigma of maturity, as if maturity really was something to be ashamed of.
That said, the companies themselves present a bigger challenge. Executives and directors certainly realise that growth rates peaked long ago, but their salaries
are still linked to the size of their companies (making acquisitions attractive), and stock-based incentive pay encourages buybacks rather than dividends.
There also persists this myth that growing companies need their earnings to keep expanding, but for most tech companies, this simply isn’t true. They invest for growth primarily through research and development, which comes out of current profits.
I don’t expect fast-growing Salesforce.com (CRM) to pay a dividend; it’s barely turning a profit. But better-established firms within Big Tech are so profitable, in terms of returns on tangible capital, that they can continue growing at above-average rates while generating far more cash than they consume. Here we find Cisco, which now has a 29% payout ratio and plans to return 50% of free cash flows to investors through dividends and buybacks. This is a lot better than nothing, but leaves an unanswered question: Why retain the other 50%?
Another major problem—made explicit when Apple announced its dividend in March—is the US corporate tax policy. Big Tech, even more than other large-cap American businesses, tends to book a large share of sales and profits overseas where corporate tax rates are lower—Ireland, for example, where the corporate income tax rate is just 12.5%. As long as those profits stay in offshore bank accounts, the income won’t be taxed again in the United States. But if the cash is repatriated to the US parent, whether to pay dividends, repurchase shares, or whatever, the company owes the difference between the tax it has already paid abroad and the US rate of 35%.
As I pick apart the possible explanations for low dividends in tech, they still smell more like excuses
This goes a long way toward explaining why Apple paid an effective tax rate of just 24% in 2011, and as of June was sitting on a cash hoard of $117 billion—some 69% of which was offshore. It also explains, at least in part, why Apple declines to pay dividends or even make share repurchases with the cash generated from its overseas operations. I won’t debate corporate tax policy here, but Apple is not supposed to be a money market fund. If management won’t repatriate the cash to provide returns to shareholders, how much is that cash—or the ongoing cash flows of overseas subsidiaries—really worth?
This is not just a dividend policy problem; it also speaks to very real concerns regarding earnings quality and corporate governance. (By contrast, the possible expiration of favourable dividend tax rates on shareholders hasn’t prevented big increases in tech payouts this year.)
Not So Unique Anymore
Still, these reasons are not wholly unique to tech. Multinationals like General Electric (GE) and the big drug companies manage to post low effective tax rates, but nevertheless provide far larger dividends. As I pick apart the possible explanations for low dividends in tech, they still smell more like excuses.
In the end, I believe Big Tech has matured to a point where we shouldn’t treat the stocks much differently than we’d treat large and successful (albeit cyclical) firms in other industries. The only additional threat is that Big Tech remains vulnerable to technological evolution that doesn’t threaten the maker of Cheerios. But if that threat is more important than anything else, how can hoarding cash be a better solution for shareholders than paying the largest possible dividends?
Aside from Intel, it’s high time for these firms to match their maturity and economic importance with respectable dividend policies. If other Big Tech stocks get with the programme, I would consider buying more of them—but for now, Intel is the only one I want.
The original version of this article appeared in the November 2012 edition of the Morningstar DividendInvestor newsletter, which is published out of the US office.
