What exactly is MSFT's dividend strategy?

Dividends are a significant part of the historical return attributed to stocks and the market generally. Indeed, of the oft-quoted 10% average market return, some 4.5% of that represents dividends. Which is why a company's dividend and its strategy wrt to said, are important to investors.

MSFT declared its intention to pay a dividend in January of 2003. This was coincident with the last ill-advised split, and amounted to $0.08 ($0.16 pre-split). Neither announcement was well received by investors, with many "growth" players seeing the dividend as an admission that MSFT's best growth was behind it and abandoning the stock (which, unfortunately, proved to be correct - at least to date). For those who remained, the initial dividend was far below market and therefore near-meaningless. This was addressed, to some extent, over the following 12-18 months, as MSFT followed up with some fairly aggressive and consistent increases - which provided a basis for future optimism. Then came the brain-dead and fiscally-irresponsible $3 one-time, which proceeded to tank the stock some $5 or more before the dust settled. Since then, it's been difficult to discern what MSFT's strategy is for the dividend. The early (directionally positive) momentum, has been replaced with long periods of no increases or, more recently, modest $.01/Q per year (or so) ones. Currently, MSFT pays just 1.38%, while the S&P average for dividend-payers is 2% (1.52% for all 500 issues combined). Now, bear in mind that MSFT hasn't been able to even keep pace with that index for the past 5 years (although it has a shot at doing so this year). Also, unlike almost all those companies, MSFT is still sitting on ~$32B of cash and short-term investments, as well as another $8.7B of "Equity and other investments" (all figures as at last Q) - the necessity for the latter being a subject unto itself.

The bottom line? If, despite recently paying the management team $1B in bonuses and pouring $B's more of shareholder cash into buybacks (which, thankfully, aren't being diluted even faster for a change), Ballmer is still having trouble figuring out ways to spend MSFT's enormous cash horde - which appears to be the case (see here, here, and here) - how about this suggestion: increase the dividend? Quite frankly, how any leadership team sporting that abysmal track record of performance versus the S&P can still be in power, is difficult enough to understand. But how they can do so while simultaneously sitting on a mountain of cash, and nevertheless paying a grossly under-market dividend, is unfathomable. You can't have weak growth, poor relative performance to the market, AND a far below market dividend, and still expect anyone to find your stock particularly attractive. So if you're truly "shareholder-focused", start demonstrating that fact by raising the dividend to at least the S&P dividend-payer average of 2%. You know... like say, INTC (2.32%)?

OT but related: watching CSCO's surge today following very strong numbers last night, I see that they've now joined the growing list of companies, including SAP, ORCL, AAPL and others, who's stocks have outperformed MSFT over the past 10 years. Ballmer says that shareholders need to take a "long-term" focus. Isn't a decade a reasonable timeframe in which to assess management's performance in their primary mandate of increasing shareholder value? If MSFT can no longer outgrow many of its competitors for investing $, stubbornly refuses to ditch money-losing new ventures or make deep cuts in existing core businesses to better drive earnings, and continues to lag many peers and all major indexes wrt the stock, then hadn't it better start providing at least a market dividend in exchange for anyone bothering to hold it as anything other than a trade?