Don't Count On Cash Flow
The one measure you thought you could trust can be twisted too.
FORTUNE
Monday, May 13, 2002
By Herb Greenberg
With the quality of its earnings under the microscope, technology-services company EDS did something unusual when it announced fourth-quarter results several months ago: It went out of its way to tout the strength of its cash flow. Not earnings, but the stuff that earnings spring from. The stuff that's allegedly unmalleable, that tells the true story. If operating cash flow is strong, the theory goes, the quality of earnings should be strong too. "Investors are looking for some measure of truth to hang their hat on," says Charles Mulford, an accounting professor at the Georgia Institute of Technology and author of The Financial Numbers Game. "There's a feeling that if they can deposit it in the bank and you can see it and spend it, then it's real--and unlike earnings, not subject to the whims of the accountants or the vagaries of GAAP."
Truth seekers, we have bad news for you. Like everything else in financial statements, even cash flow is corruptible--or at least subject to quality issues of its own. And while cash flow may be harder to manipulate than earnings, Mulford says, there's a "surprising amount of flexibility" in how cash flow can be reported. That's critical if you're considering this measure as a clue to how solid earnings really are.
Cash flow--referred to as operating cash flow in annual reports--is generally defined as net income before the effects of such noncash expenses as depreciation and amortization. (Despite what many companies claim, Ebitda is not cash flow.) Investors in capital-intensive companies, such as EDS, want a further level of detail, calculating "free" cash flow, which is usually defined as operating cash flow minus capital spending.
Yet some items can trick investors into thinking cash flow is stronger than it really it is. Take the tax benefit companies get when employees exercise stock options. Accounting rules consider that benefit part of operating cash flow. Almost all number sleuths, however, believe it's an unpredictable nonoperating gain that investors should subtract from operating cash flow. (As stock prices collapse, so does the benefit, because fewer options are exercised!) After falling for a year, for example, operating cash flow at Lucent turned positive in 2000. But analyst Albert Meyer of David Tice & Associates, who had long been warning of trouble at Lucent, wasn't fooled. "A watchful eye could see that it was entirely attributable to tax benefits," Meyer says. So much so that without the tax benefit, Lucent's 2000 operating cash flow would've been a negative $760 million rather than a positive $304 million. Instead of getting better, it was getting worse!
Then there's American Software, which could serve as a poster child for the way that capitalizing expenses can artificially inflate cash flow. The company, which capitalizes the cost of software development--amortizing it over months or years rather than expensing it immediately--is not doing anything wrong. Under GAAP a host of expenses can be capitalized--from direct-response advertising to landfill development to debt issuance. That makes actual expenses lower than they would otherwise be, pushing earnings higher. And by those rules, costs that are capitalized count as an "investment" item on the cash flow statement. The trouble is that when a company reduces the amount it capitalizes, cash flow can suddenly drop too. That's what happened at American Software, which sliced its capitalized costs by 62% in 2001. As a result, cash flow, which had been positive, turned negative.
Even if nothing on the cash flow statement is out of the ordinary, there's no guarantee that earnings are as solid as they appear. Which gets us back to EDS. While its free cash flow may have been strong, the question should be, Strong relative to what? EDS's operating and free cash flow have been falling in recent years while earnings have been rising. That's rarely a good combo. EDS claims cash flow is being held down by costs associated with new contracts, but UBS Warburg analyst Adam Frisch says it could also signal aggressive revenue recognition, lax credit terms, or the increased use of off-balance-sheet transactions. "We view this as a material and ongoing issue," he recently wrote in a report explaining why he rates EDS a hold. "We prefer other business models that can grow revenues, margins, and cash flow simultaneously." Which, of course, is the way it should be. |
