This chart shows Dell's free cash flow and amount spent on share repurchases on the left axis, and the year-over-year change in diluted shares outstanding on the right axis:

Dell's Balancing Act Is Dell's stock buyback program shareholder-friendly or not?

(Note: All numbers are in millions of dollars. Free cash flow is cash flow from operations, minus depreciation and amortization, which I use as a proxy for maintenance capital expenditures. I did not subtract cap ex, because this would penalize Dell unfairly for its investments in growth. Normally, I also subtract "tax benefits of employee stock plans," which in my opinion have nothing to do with operating activities, but I have not done so here because free cash flow is being compared with amounts spent repurchasing shares resulting from those same employee stock plans.)

Dell generated healthy free cash flow, totaling $10.8 billion, in the 13 quarters covering fiscal years 1999, 2000 and 2001, plus the recently reported first quarter of fiscal year 2002. (This figure significantly exceeded net income of $6 billion during this period, demonstrating how the company's direct model combined with effective working capital management -- not to mention $2.6 billion of tax benefits of employee stock plans -- enhances its cash flow.) Dell used 56% of its free cash flow over these 13 quarters to buy back its own shares, which can be a shareholder-friendly action.

This sounds great, but there are two problems with what is happening. First, it is unclear whether Dell's share repurchase program in the past few years has been creating or destroying shareholder value. The stated purpose of the program is to offset the dilution due to Dell's stock option program, but that's not necessarily a good thing. As Warren Buffett wrote in his 1999 annual letter to Berkshire Hathaway shareholders, "There is only one combination of facts that makes it advisable for a company to repurchase its shares: First, the company has available funds ... and, second, finds its stock selling in the market below its intrinsic value, conservatively-calculated."

Dell has had plenty of cash in recent years, but was the stock really trading below conservatively calculated intrinsic value above $40 and even $50 within the past year? If not, then Dell was destroying shareholder value when, for example, it spent $2.7 billion in FY 2001 buying back 65 million shares at an average price of $41.54. Dell continues to spend nearly all of its free cash flow to buy back shares in the belief that, according to Reid, "We don't think we're a $24 to $26 stock long term." Current and prospective Dell investors will have to decide for themselves whether they think this is a good use of Dell's cash.

The second problem is that Dell is getting less and less bang for its share repurchase buck. Let's look at the data for each year:

• In fiscal year 1999, Dell spent $1.5 billion on share repurchases, equal to 65% of free cash flow. Consequently, diluted shares outstanding fell by 2.6% during the year, which is exactly what a shareholder would like to see (not to mention that the stock nearly quadrupled).

• In fiscal year 2000, Dell scaled back its share repurchases to only $1.1 billion, or 28% of free cash flow. Wise move, given the stock's lofty valuation -- it traded at a price-to-earnings ratio above 70 for much of the year -- but consequently, the share count only fell 0.7%.

• In fiscal year 2001, as the stock fell dramatically, especially in the second half of the year, Dell ramped up its share repurchases to $2.7 billion, equal to 68% of free cash flow. Yet the diluted shares outstanding rose 1.9%.

• The trend continued in the first quarter of fiscal year 2002, as Dell spent 99% of its free cash flow buying back shares, yet the share count rose 0.7%.

Stock Options

Wait a second! How could the share count be rising when Dell is spending so much on share repurchases? Can you say stock options? Like many companies, especially those in the technology industry, Dell issues a large and increasing number of stock options to its management and employees.

Last year, as Dell's stock price fell, many options became worthless because they were struck at much higher prices. This is demoralizing to employees and no doubt made the company nervous about retaining its talent. So what did Dell do? It more than tripled the number of options granted in fiscal year 2001 vs. fiscal year 2000 from 50 million to 154 million. (This information is in the fiscal year 2001 10-K under Note 6 of the Notes to Consolidated Financial Statements.) It's not surprising, therefore, that Dell's share count rose despite the massive buybacks.

Reid at Dell notes the jump in stock options granted last year was due partly to many new employees and also to a special double option grant. However, unlike regular options, which begin vesting immediately, the extra options don't begin vesting for two years (all options vest over five years). Reid said the recently granted options are neither vested nor in the money, so the stock price will have to rise over time for them to be worth anything, thereby aligning Dell employees' incentives with those of shareholders.

By my calculations, had Dell wished to keep its share count steady in fiscal year 2001, it would have had to buy back 52 million more shares (on top of the 65 million shares it purchased for $2.7 billion). Using this average price of $41.54 per share, Dell would have had to spend an additional $2.2 billion, for a total of $4.9 billion. Obviously, this was not feasible given that free cash flow was only $4 billion.

Implications

Even though the cost doesn't appear on the income statement, stock options are clearly a form of compensation, and there is a real cost associated with them. Let's say a company spent 100% of its free cash flow buying back shares, yet the share count remained constant. Isn't this just a compensation expense, and therefore in this scenario the company generated no free cash flow for shareholders?

At Dell, the situation was even more extreme last fiscal year: Had it spent 100% of its free cash flow on buybacks, the share count still would have risen 0.8%. Thus, one could argue that Dell had negative free cash flow for shareholders last year.

To some extent, this analysis penalizes Dell for having a stock that's done well over time (though obviously not in the past year or so). The dilution that took place in fiscal 2001 was due mostly to stock options granted many years ago, when the stock was at a much lower price. Had the stock done badly, then many of the options would have expired worthless and the company also would have been able to buy back more of its shares for the same amount, but shareholders would, of course, be worse off today.