3Com Bear's Den
by Jim Surowiecki ([email protected])

Minuscule profits, declining margins, a mishandled merger, troubles in Asia, questionable accounting, and fierce competition from two of the best-managed companies in America --that's a pretty good description of what investors in 3Com have faced for the past year and what they're going to be facing for the foreseeable future. At this point, the answer to the question, "Are you bearish on 3Com?" should be a foregone conclusion. The real debate is about whether or not you should be shorting the stock.

The mighty have indeed fallen a long, long way. At the beginning of last year, the entire computer networking industry could do no wrong in the eyes of the stock market, and although Cisco was the class of the field, 3Com seemed to be as strong a second-place contender in terms of its management and its growth strategy as it was in terms of market share. But what the company has shown us over the last fifteen months is precisely how easy it is to stumble in a business that places a premium on innovation and reliability. It has also shown us how vulnerable even relatively large companies are to the technological and pricing power of Intel <% if gsSubBrand = "aolsnapshot" then Response.Write("(Nasdaq: INTC)") else Response.Write("(Nasdaq: INTC)") end if %>, which has recently moved strongly into the low-end network interface card (NIC) business, where 3Com had previously been the dominant player.

The results are as instructive as they are ugly. In the second quarter of FY 1998 (third-quarter earnings will probably be released around the time this is published), 3Com earned just $15.1 million on sales of $1.22 billion, a net margin of 1.2% and a year-over-year decline of almost 90%, even though 3Com had become a much bigger company thanks to its overpriced and error-ridden acquisition of U.S. Robotics. That acquisition makes annualized sales comparisons difficult, but on a sequential basis sales *fell* by nearly $400 million in the second quarter, even as SG&A expenses rose by 10%. More strikingly, inventories skyrocketed, while the company's ratio of receivables to accounts payable remained at about 3-to-1, even though ideally you want to owe more than you are owed.

To be sure, much of the decline in sales came from the "client-access" side of the business, which means modems and NICs, and much of that decline was the result of U.S. Robotics' clogged inventory channels, which 3Com decided finally to clear out at the expense of current revenue rather than continue pushing new sales on its vendors. But even 3Com's systems business -- where it was the No. 2 player in the industry -- declined on a sequential basis, and the company warned that it had inventory problems in systems as well as in modems.

Gross margins for 3Com as a whole have fallen from 55% in late 1996 to 47% today, a decline that could not bode worse for the company in the next year or so. In the simplest terms, 3Com is paying more to make its goods while selling them for less, hardly a recipe for earnings growth. Almost across the board, this has become a company that has serious problems doing the things a company needs to do: forecasting demand, controlling supply, and managing its production facilities.

The merger with U.S. Robotics now looks like one of the great business gaffes of 1997. 3Com appears to have hurried its way through the due diligence, failing to scrutinize U.S. Robotics' inventory management and its accounts receivable, and ended up paying potentially billions more for the company than its bottom line actually justified. By the time the merger was completed, modem inventories were up to 18 weeks, triple what they should have been, while accounts receivable had risen to 3 1/2 months. 3Com then compounded its problems by being less than frank in its accounting practices, loading up the quarter of the merger with write-downs and setting up a situation where the problems with the merger would be hidden and the results going forward would look better. 3Com has now restated those numbers after SEC "queries."

3Com does have the PalmPilot, one of the truly most excellent products of recent memory and one around which a fervent cult has already sprung up. On the other hand, even 3Com execs describe the PalmPilot's branding campaign as among the worst-handled in history, and Microsoft's entry onto the scene -- with a product called the PalmPC, no less -- means that all bets are, unfortunately, off. More to the point, the handheld-computer market is certainly a growing one, but it's still small enough that it won't be able to carry 3Com all by itself.

The networking business will, of course, continue to grow rapidly in the future, and 3Com remains in a good position to take advantage of the networking of small businesses and of telecommuters. But the company's recent performance offers investors no reason to believe that its execution of its strategic plans will be particularly efficient. The really remarkable thing about 3Com, in fact, is that it continues to trade at a P/E ratio of more than 30 at a time when its earnings expectations are anyone's guess and may not even reach double figures. The company's current valuation is a legacy of the days when networkers were golden boys and when U.S. Robotics seemed to be the gem of the modem-makers. But that valuation is a massive overvaluation, and the future is nowhere near bright enough to make buying 3Com a sensible investment. Not long-term, not short-term. No term.

Next: The Bull Responds