IBM Sure Is One Undervalued Net Stock
At least compared to Internet pure plays. So how do you put a proper value on Big Blue? Here's one attempt

In 1993, just before Louis V. Gerstner took over as CEO of IBM (IBM), a plan circulated through the corporate headquarters to to maximize shareholder value by breaking up the company into seven or eight smaller companies, probably to be called Baby Blues. Gerstner promptly killed the idea and pronounced that Big Blue was worth more as a whole than in parts. From that point on, IBM has enjoyed a resurgence that shocked many of its competitors, surprised Wall Street, and is still going strong.

No one is talking about the old breakup plan today. Its stock closed at $105.25 on Dec. 2, which gives it a price-to-2000 earnings ratio of 33. That's small-time, however, compared with the valuations of pure-play Internet stocks like Sun Microsystems (SUNW), whose 2000 p-e is 77, and e-commerce king (AMZN), which doesn't have a p-e because it doesn't have any profits, yet whose stock has appreciated 15-fold in the past two years.

Which raises the question: Should IBM be valued as an Internet stock? Don't expect that to happen anytime soon. But if you look at each of IBM's divisions as if it were a stand-alone company, a picture of systemic undervaluation comes into focus.

GROSS MARGINS. IBM consists of three main businesses. The biggest revenue producer is hardware manufacturing, which brought in $26.8 billion in revenue in the first nine months of 1999. This division is pretty evenly split between personal computer production and mainframe and server production. After that comes global services, which can be roughly defined as IBM's consulting arm, and which brought in $23.4 billion in the same period. Finally, the company's software division had $9 billion in sales the first three quarters of this year.

Software is probably the highest-margin of the three businesses. Of that $9 billion, only $1.6 billion went into production, which means the company makes a gross profit margin of almost 80%, roughly on a par with Microsoft's (MSFT) gross margins. While IBM's software operations aren't expected to grow as rapidly as Microsoft's, there's no reason to believe that they'll retreat, either. In fact, software revenue grew 15% in the first nine months. And that happened during a time when many corporations held off on software purchases because of spending on Y2K projects.

The global services business is by far the fastest-growing division in IBM: In the first nine months it grew 17%. Much of that is attributable to IBM's e-business division, which is expected to produce $3 billion in revenue this year. In fact, the majority of IBM's services business has something to do with the Internet.

OUTSOURCE OR SELL. Hardware is the only IBM division that is stuck in a growth rut. Indeed, the PC division lost money last year. This is the part of IBM's business that analysts like the least. "They have a $12 billion PC business that's not doing well and is distracting them from their Internet focus," says John Jones, an analyst with Salomon Smith Barney who rates the stock a buy and expects it to hit $155 before the end of next year. "If I were in charge, I would either outsource their manufacturing to someone who can do it more efficiently, which is what Hewlett-Packard does, or I would create an independent company within IBM with its own tracking stock and incentivized managers."

Few analysts are calling for IBM to take its name off of computers. "It's like McDonald's advertising hamburgers but making its money off of
fries and shakes," says Steven Milunovich of Merrill Lynch, who has a neutral rating on IBM right now. "But it's wrong to look at [hardware] as a loss leader. They need to figure out a way to generate profits in that division, or else they shouldn't be in that business."

Gerstner claims that the unit is on the cusp of profitability and should get over the top by the first or second quarter of next year -- as soon as the Y2K buying freezes that many companies imposed are removed. Meanwhile, the company has already exited the consumer retail market, a move in the direction of making money.

IMPRECISE EXERCISE. So IBM has a hardware business that's in the doldrums but should experience a moderate turnaround soon, a software division that is pleasingly profitable, and a services division that is turning in gonzo growth. As an imprecise exercise, let's compare each division to a similar public company and give it that company's market valuation.

Because IBM doesn't break out operating profits by division, we'll have to do this using price-to-sales valuations. Let's start with software. Since Microsoft isn't really a direct competitor, let's compare IBM to Oracle (ORCL), which makes database software, also has an Internet-centric business model, and whose software gross margins are about he same as IBM's. Oracle's stock has a price-to-sales ratio of 12.1. Apply that to IBM's software sales over the last four quarters, and you have a market valuation of about $145 billion.

IBM's services division is a much harder comparison, because it's involved in so many businesses. But as an exercise, let's compare it to Sun's price-to-sales ratio of 9.1. True, Sun is primarily a server and software company. But it is getting into the services business and, more important, is one of the few non-dot.coms that has been given an Internet valuation because of its huge Net-related business. Multiply the last four quarters of IBM's services revenue by Sun's multiple, and you get a market valuation of $282 billion.

Then there's the hardware division. Few would argue with comparing this business to Compaq (CPQ), which has gone through similar hard times and is in the process of mounting a comeback to profitability. Compaq's price-to-sales ratio is 1.1, which would give IBM's hardware division a market valuation of about $35 billion.

LESS RESPECT. Total all three of these valuations, and you get a market capitalization of $462 billion. IBM's current market capitalization is $190 billion. That's a 150% undervaluation, which means that if this exercise made sense in the real world, IBM's stock should be worth about $250 a share, rather than $105.

Of course, this scenario doesn't translate well in the real world. That's because the three divisions aren't separate, and IBM isn't about to split them up. Even if they did get their freedom, no one can say the units will be as successful as an Oracle or a Sun. Still, it's abundantly clear that IBM's stock doesn't get the same respect from Wall Street that its Silicon Valley competitors do.

How can it get that respect, short of a tumultuous breakup? One option would be to give each unit greater independence and a tracking stock. Salomon's Jones wants to see that happen with the hardware division. And Ulric Weil, who follows the stock for Friedman Billings, Ramsey, would love to see it happen in the e-business division. "It would put more value on their books, and it would stop people from looking at them as a hardware company," Weil says.

Another option is for Lou Gerstner to continue his path of turning the overall company into an Internet giant and let Wall Street unlock the value whenever it's ready to. Maybe then you could apply's price-to-sales ratio of 24.4 to IBM's entire revenue stream -- producing a market valuation of more than $2 trillion. That would never happen, laughs Jones, who adds: "IBM makes too much money to get an Amazon valuation."

Jaffe writes about the markets for Business Week Online

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _


Inside IBM: Internet Business Machines

TABLE: IBM's e-Business Strategy

TABLE: How IBM Uses the Net

Gerstner on IBM and the Internet

ONLINE ORIGINAL: IBM Sure Is One Undervalued Net Stock

E-Mail to Business Week Online

Copyright 1999, Bloomberg L.P.
Terms of Use   Privacy Policy