"Any way you slice it, operationally, we beat estimates by a wide margin." -- Hewlett-Packard CEO Carly Fiorina, speaking at the company's third quarter 2000 conference call.
Let's come right out and say what Wall Street really thinks: Carly, you're full of it.
Shares of Hewlett-Packard (NYSE: HWP) dove more than 9 percent today after the company reported third quarter results. Earnings were better than expected, but not by as much as HP executives asserted.
All it took to smash HP's third quarter smile -- a rosiness reflected in the initial news articles, including one written by me -- was a 20-minute span during the Q&A portion of yesterday afternoon's conference call.
Bear Stearns analyst Andrew Neff fired the first damaging bullet on the call:
"I'm having difficulty getting to the 97 cents (reported by HP as operating earnings per share)... It seems like your upside came mainly from (the) Interest and Other (line of the income statement). Maybe you could clarify that."
Neither Fiorina nor CFO Robert Wayman could bring clarity, at least not in the eyes of analysts. That failure became evident over the next several questions, including the following wound from Prudential Securities' Kimberly Alexy:
"When we exited last quarter, with the (Unix server) growth rates suggesting possibly we had seen acceleration in Unix growth ... Can you give some more color about what you saw during the quarter that may have changed to lead to some of the lower revenue growth in Unix, besides the compares?"
The "compares" cited refer to HP's insistence that the third quarter had a tough comparison with the year-ago period, when HP saw an influx of revenue from orders that were previously delayed.
Analysts weren't buying that claim. A few minutes after Alexy took aim, Banc of America Securities analyst Kurt King unloaded a double barrel of buckshot:
"Carly, I just wanted to come back to the Unix topic that others have asked questions about. If my notes from a year ago are correct, your revenue growth in Unix (for the third quarter of 1999) was in the mid-teens...
"The N-class server (which HP execs pointed to as the reason for last year's robust third quarter) began order-taking just a month before (last year's) quarter began, so I think I'm struggling with the same issue others are, in determining how the compare really qualifies as all that difficult. And I'm wondering if you could address that topic a little bit further in conjunction with your guidance at the analyst meeting in May, about your expectation of acceleration, and how the number you just reported actually showed any sign of acceleration."
Don't you love analyst syntax? Long-windedness aside, King's point is valid: less than three months ago, HP said it would accelerate its Unix revenue growth rate. Instead it fell on a year-over-year basis, and you shouldn't blame a difficult precedent set by last year's third quarter, because that was already a known factor when HP made its lofty prediction less than three months ago.
Fiorina insisted the year-over-year comparison holds up:
"I'm not sure that I can give you additional data beyond what we've already said, so I apologize in advance if I'm being repetitive here, but we were throwing a lot of data at you, so lets try and reiterate what we've said..."
You can bet analysts don't appreciate the implication that they're incapable of understanding "a lot of data." That didn't stop Fiorina from continuing:
"First, we believe that having normalized our growth for last Q3 from the anomaly of about five months worth of ship into July -- and when I say 5 months I not only mean the entire third quarter, but I also mean about $100 million dollars from Q2 that was delayed into Q3 -- if we normalize for that anomaly, we think our growth rate for this quarter would have been in excess of 25 percent."
King and others find it hard to swallow that claim of 25 percent growth, which explains today's Banc of America research note:
"The market will likely take a more skeptical view of HP's 100 percent bullish approach to dealing with Wall Street under the new CEO. Management came up short of some key guidance metrics in 3Q and in our view wasn't entirely credible in dealing with some of the concerns that were raised on last night's conference call."
Perhaps the biggest hurt came from Sanford Bernstein's Tony Sacconaghi, lobbing a question at CFO Robert Wayman:
"Bob, I'm still struggling a little bit with what expectations were based on with EPS and what you delivered..."
Unix growth notwithstanding, Sacconaghi went on to point out that if all non-operating items are excluded from HP's third quarter results, earnings per share barely exceeded analyst forecasts. Sacconaghi particularly noted that the quarter did not include certain expenses -- about 5 cents per share worth -- that were in last year's third quarter.
When all the non-operating events are added up, HP earned roughly 87 cents per share by one analyst estimate, or only 2 cents higher than First Call's 85 cents consensus.
"I'm just questioning Carly's statement about, by no matter how we measure this, this was a clear outperformance versus expectations," Sacconaghi said. "And I'm just struggling to reconcile that."
Replied HP's director of investor relations:
"I think we're going to move on to the next question."
Lost in all this was the fact that HP really did post a good quarter. But the company's labored financial explanations -- King and HP executives during the call actually tussled over the meaning of "acceleration" -- obscured the actual results.
Printing and imaging remained strong. HP's Unix server growth, though slower than predicted, continued to roar in the low- and mid-range markets. PCs did extremely well. Service revenues saw a solid uptick.
More important, HP's fourth quarter looks better than it did before. Every analyst that issued a research note today -- even the ones who downgraded -- raised earnings estimates for the fiscal year.
But at the moment, fairly or not, few investors trust the company to report accurate results. 22GO>