Letter to Financial Journalist

"I enjoyed your column on stock buy backs. If I may, I have a comment or two. You wrote: "It's true that buybacks can offset new shares emanating from employee stock options."

It is actually the other way round. First, companies issue stock to employees and then they go and buy them back. Here is the issue: Management claims that stock-based compensation is not an expense. Now that they have to expense stock-based compensation, they claim that the expense is a non-cash event and as such should be added back, or ignored. Cisco and Broadcom, among others, take this view.

Seeing you made reference to Dell, I'll use it for illustrative purposes. During the past 12 years, Dell took a stock-based compensation expense of $15.6 billion on its tax returns. This is also what it would have cost the company to buy back the stock it issued to employees at the time of the option exercise, net of employee strike price contributions. If stock-based compensation is not really an expense, how come it turns up as one on the tax return?

As for the non-cash nature of stock-based compensation, this is where stock buybacks come in. At the beginning of 1995, Dell had 2.432 billion (split-adjusted) shares outstanding. At the end of 2005, it had 2.485 billion shares outstanding, an increase of 53 million; or net decline of 247 million, not counting the 300 million shares issued with the conversion of preferred shares. During this period the company repurchased 1.078 billion shares. One could argue that the bulk of the repurchases related to stock previously issued to employees. From 1995 to 2005, the company granted employees approximately 770 million options and a similar amount of options were exercised by employees. During the period 1999 to 2005, the company spent $16.8 billion on stock buy backs, but net of employee strike price contributions of $7.2 billion, the cash outflow was $9.6 billion. As the share count has hardly moved, despite all this buyback activity, one has to question the assumption that these expenditures were truly an allocation of capital.

To labor the point, consider a shareholder who owned, say, 10% of Dell back in 1995. Today, this shareholder owns 9%. However, if Dell's stock repurchases truly accomplished what we are all lead believe, the same shareholder would have had a 15% ownership. The share count would have been trimmed from 2.3 billion to 1.5 billion. Under the stock-based compensation regime, the billions of dollars spent on stock buybacks kept the share count more or less stable; ergo, these buybacks are not an allocation of capital. We should not allow FASB's defective reporting standards to fool us so easily.

FASB, ever eager to please management, has written accounting rules that kept stock-based compensation out of earnings, even though a hefty stock-based expense lands up in the tax return. In addition, FASB ruled that if management spends real hard cash to buy back all or some of stock it issued to employees in lieu of cash wages, the cash outflow should be reported in the financing section of the cash flow statement. Forget about accounting rules and think like a fifth grader. If management gives employees a "coupon" instead of cash wages, which it later redeems, why is that redemption not identified as compensation expense? The cash outflows that go merely to mop up the dilution caused by stock option exercises should by rights be reported as a reduction of operating cash flow. However, because it lies hidden in the financing section of the cash flow statement, we write about these buy backs as if they represent capital allocation decisions. I would argue that the billions of dollars Dell spent on stock repurchases were for the most part nothing more than a compensation expense in disguise - call it the settlement of a deferred compensation liability. There are analysts on Wall Street that make this argument, e.g., David Zion at CSFB. If financial journalists were take this stance and adjust earnings and cash flows of the Dells of this world for the cost of buying back the stock that they issued in order to avoid reporting a realistic compensation expense, the capital markets would discover that these darlings of the tech world are only half as profitable as the accountants would want us to believe. Such journalism would add serious value and help investors allocate capital on a more rational basis. It would also show up FASB's deceit.