Compaq and HP: What's an Investor to Do?

BusinessWeek crunches the numbers to help sort through the claims and counterclaims

The low-profile investor advisory Institutional Shareholder Services may have done what millions of dollars in ads and related proxy-fight expenses couldn't do: tip the scales in favor of Hewlett-Packard Co.'s (HWP ) contested purchase of Compaq Computer Corp. (CPQ ) On Mar. 5, the Rockville (Md.) firm advised institutional shareholders controlling 23% of HP's shares to vote for the merger. There's no assurance that all of them will, but ISS has strongly influenced votes in the past. That puts pressure on dissident board member Walter Hewlett, who has persuaded investors owning 21% of the stock to oppose the merger. "If it had gone the other way, the deal would have been dead," says Merrill Lynch & Co. analyst Steven Milunovich. "Now, it's a horse race."

Shareholders must soon place their bets. From now to when the votes are cast on Mar. 19, both sides will turn up the volume on what they see as the pros and cons of the $22 billion deal. For HP's 900,000 shareholders, sorting through the blizzard of claims and counterclaims could prove as dizzying as a climb up the Tetons. Even stock analysts have been loath to release numbers on how the merged company might perform. And so the question remains: Will the largest tech merger in history create a computer colossus or a another merger mess?

A financial dissection by BusinessWeek, with the help of Banc of America Securities, shows that a go-it-alone HP will grow faster and be more profitable in the near-term. Further out, however, an HP-Compaq combo could deliver juicier profits--providing the companies can smoothly integrate the two giants without losing momentum or creating snafus.

For the risk-averse investor, a solo HP is the safer bet. With the economy providing a small lift, HP could hit net income of $2.3 billion in calendar 2002, up 54% over a year ago. A combined HP-Compaq would be considerably bigger in size, yet its earnings growth would not be as high: Profits would be up 50% in 2002, to $2.7 billion. And because the merger would balloon the number of HP shares outstanding, earnings per share of a stand-alone HP would be $1.17, vs. $0.89 for the merged entity.

For the bold investor who believes management can integrate the companies, the merger makes sense. A combined HP-Compaq could deliver a bigger earnings punch by 2004. By then, the merged company could boost profits 30%, to $5.6 billion. That compares with 10% earnings growth for an independent HP, which would hit $3.4 billion in net income that year. "If HP pulls it off, the stock could go to 30," says an investor who has purchased more than 10 million shares since Jan. 1. Today, HP's stock is hovering at 20.

Still, there's a lot that can go wrong in executing such a complex merger: Executives must manage such diverse businesses as $25 ink cartridges and $3 million supercomputers while dealing with morale issues as the payroll is slashed by at least 15,000 workers. Indeed, ISS Vice-President Patrick McGurn says: "My initial reaction [to the merger announcement last September] was in line with Walter's. My gut instinct was, they've got to be kidding." But after 20 meetings and incalculable hours poring over management's plan for weaving the two companies together, the ISS team was won over. Says McGurn: "We had a high enough confidence level to say the risk was manageable."

On paper, it's easy to see why ISS likes the merger. HP would not only enjoy top market share in printers, PCs, and storage, it would have the second-largest server business and the third-largest tech-services organization. The marriage would double the sales staff, to 15,000, so HP could woo corporations of all sizes, not just the top few hundred accounts HP handles today. "Without this merger, HP wouldn't have the sales horsepower to compete with Sun or IBM," says analyst John B. Jones of JB Ventures.

Best of all, the new computer goliath would have not one but two cash cows. HP's $9 billion ink-cartridge business cranks out more than $2 billion a year in profits. Plus, the merger would create a computer-repair business double the size of HP's, pumping out $1 billion in annual profits. Indeed, HP says the new company could fund more than $4 billion in research and development--up from the $2.6 billion HP now spends.

And investors could benefit big time from huge cost savings. By eliminating redundant administrative functions, HP says savings would reach $2.5 billion a year by 2004. The company would likely write off most of the more than $1 billion cost of the merger in 2002. It figures it would suffer a 5% dip in sales in 2002 and 2003 as redundant product lines are killed and rivals swoop in to grab customers while management focuses on the merger. Even assuming revenue slippage of 7.5% for 2002 and 2003--which is more in line with most analysts' estimates--HP's net margins would still rise to 6.5% in 2004 from 3.4% today, according to BusinessWeek's analysis.

That's the optimistic view. Lots could go wrong. For starters, there's no assurance that the merger will fix what ails HP's $9.2 billion corporate-computing business, which includes servers, storage, and software. Operating margins for this business were -8% in the quarter that ended Jan. 31, vs. the -6% margins logged the quarter before. And there's more trouble on the horizon. Today, a third of HP's server business--its powerful Unix machines--has healthy profit margins and is tied with Sun Microsystems (SUNW ) for top market share. But most analysts expect Sun and IBM (IBM ) to outpace HP in this sector when tech spending recovers because rivals are offering newer Unix models. At the same time, HP's $1.6 billion Intel-based server business continues to leak red ink.

Buying Compaq will help, but how much is unclear. For starters, Compaq's enterprise business has woes of its own. Sales slumped 25% in 2001 as corporate spending on tech gear dried up. And aggressive price cutting on servers by Dell Computer and IBM sliced a point off Compaq's 14.8% market share last year, according to Gartner Inc. Low-end Intel-based servers, of which Compaq is the No. 1 seller, are fast becoming commodity items with thin profit margins.

Rather than run from this trend, though, HP CEO Carlton S. Fiorina believes she must turn it to her advantage. While higher-end Unix systems now bring in most of the industry's profits, over time, less expensive machines based on Intel Corp.'s (INTC ) new Itanium chip and either Windows or the Linux software will take over many jobs. While the Unix market is expected to grow around 6%, Fiorina says the market for Windows and Linux models will grow 20% a year as these systems prove their mettle at tougher computing jobs. Indeed, market researcher International Data Corp. estimates that worldwide revenue from sales of Windows and Linux servers will outpace Unix machines by 2005.

Even if Fiorina's plan works, the road there won't be smooth. Why? Intel's Itanium chip isn't taking off as fast as HP and Compaq had hoped. The chip is fast, yet few software developers are writing programs to run on it, and widespread adoption is probably at least three years away, says Compaq Executive Vice-President Michael Winkler. Since both companies have announced that they will dump their own high-end server chips by mid-decade, customers that need high-end screamers might be more inclined to go with new models from IBM or Sun. And since many corporate customers plan their IT projects around their big servers, this could also deny HP sales of storage, software, and service deals. "Itanium's lack of success would not only hurt revenue but would severely impact HP's ability to be a total IT provider," says Banc of America Securities analyst Joel Wagonfeld.

Taking on Compaq's $15.2 billion PC business is an even bigger risk. On first blush, it looks like a nice fit: The company would merge HP's profitable home-PC line, which enjoys 50% market share in U.S. retail stores, says researcher NPD Intelect, with Compaq's well-heeled corporate-PC business, which is No. 2 to Dell. Indeed, Compaq is moving furiously to outsource production, streamline distribution, and sell more on the Web to become superefficient. Its PC inventory turns--the number of times a year a company moves all of its products through its factories--doubled last year, to 62, still behind Dell's 97.

There's far more work to be done. Despite the progress, Compaq's PC unit still lost $587 million in 2001. If Fiorina is to eke out the 3% operating profit she has projected for the PC business, Compaq will have to quickly extend its efficiency gains and teach HP to do the same. "We and the industry need to be more cost-efficient, and we don't think we're done yet," says compaq CEO Michael D. Capellas. "It's time to put it in place." But to the degree that HP relies on computer dealers to sell its wares, Dell's leaner model will have a low-cost edge and allow it to wage price wars without breaking a sweat. "Dell's advantage is large and increasing," says Dell CEO Michael S. Dell.

In computer services, buying Compaq is only a partial answer. The merger will double the size of HP's maintenance-and-support business. But it does little to help HP in other service sectors, such as consulting and outsourcing. These businesses would make up less than 20% of the company's service revenue. Since they are key to landing big contracts with corporations and governments, analysts say HP may still need to do an acquisition, which is unlikely in the wake of the Compaq deal.

What can HP accomplish alone? For starters, many analysts say the company could limit PC and server losses by not chasing market share at all costs. Instead, the company would do better to hold firm on price to ensure profits, says Gartner analyst Martin Reynolds. That might chase off thrifty customers who just want the lowest price but not customers who have more complex--and potentially more profitable--needs. Say an airline wants to let travelers at the airport print out their own customized magazine to read on their flight, such as all the week's articles in national newspapers on HP's proxy fight. That setup would require a full suite of servers, storage gear, software, printers, and consulting services, so customers interested in a total solution would likely pay HP's computer prices. HP is testing such a system.

A more focused HP might also make more of its franchise printer operation. HP has built a thriving business in photo printers and all-in-one printer-copier-fax gizmos. These categories brought in 32% of HP's $5 billion in printer sales in its most recent quarter. Since photos require more ink than plain documents, the photo printers drive sales of highly profitable ink cartridges.

Yet there are far bigger printing horizons for HP--if the company doesn't get distracted by merger issues. In commercial printing, management has pointed out that just 4% of all documents are printed on HP-style printers. The rest--training manuals, annual reports, and junk mailings--are printed by offset printers that charge hefty fees. For years, HP has talked about a better way: giving corporations lower-cost printers that can match the image quality of offset-printers but can be tied into a corporate information system.

Can HP turn this vision into reality? Whichever way the merger vote goes, Fiorina is going to try. Just days after unveiling the Compaq deal, HP announced a $1 billion purchase of Indigo Technologies Ltd., which makes a high-speed, high-quality printer. Bill McGlynn, who runs a one-year-old division focused on this commercial-printing opportunity, has been helping big companies deploy the printers along with everything else required to use them: servers, customer-tracking software, and consulting services to set it all up. If McGlynn reaches his goal of building a $5 billion business over the next five years, HP could add $200 million more to annual profits by 2004, based on the operating margins of the printers, servers, and other gear involved. "What makes HP unique is that they can tie it all together," says professor Frank Romano at Rochester Institute of Technology's School of Printing Management & Sciences.

Will buying Compaq make that more or less likely? Predictably, supporters of the deal say it will help--if only by giving HP twice as many chief information officers to sell to. Still, many analysts worry that management will have its hands more than full with the merger to focus on such opportunities. Creating the uber-computer company designed to match Dell's hyper-efficiency plus IBM's technical might may take all the brainpower HP execs can muster. Like a restauranteur intent on besting both McDonald's Corp. and Le Cirque, that's a tough balancing act. And it creates a difficult choice for shareholders.

By Peter Burrows in San Mateo, Calif., and Andrew Park in Dallas

    Before it's here, it's on the Bloomberg Terminal.