Why Siemens is exiting the chip business -- Enough already
Richard Richtmyer
Siemens has had enough.
After investing heavily in its DRAM fabs and then losing more than $700 million in the chip business last year, the German electronics giant said last week that it will spin off that unit, as well as its other electronic-component segments.
The plan to spin off the businesses, outlined in Munich last week by Siemens AG president and chief executive Heinrich von Pierer, was motivated by the enormous capital requirements and huge profit fluctuations that are common for commodity-chip makers.
The move is also part of a broader 10-point reorganization scheme the company unveiled in July, designed to better focus the $70 billion conglomerate.
"It makes sense to try to rationalize the business," said Anita Farrell, an analyst with Merrill Lynch & Co. Inc. in London. "It's too big and too diverse."
Although it identifies itself as an electronics company, Siemens is a global in-dustrial conglomerate whose business lines include power plants, information and communications, health care, and transportation, as well as consultation, engineering, and integration services.
"It's a very large company. And it's been, in some ways, rather unfocused," Farrell said. "And certainly the losses this year have spiraled because of the DRAM business. So they made a decision not to remain there."
And those losses came shortly after Siemens spent a lot of money beefing up its DRAM capacity around the world.
In 1996, as the memory market began softening and other chip makers started to retreat from DRAMs, Siemens embarked on a capacity-expansion tear. As its competitors slowed or halted production at their fabs, Siemens ramped up production at its Dresden facility, constructed a new fab in the United Kingdom, and launched DRAM joint ventures with Motorola Inc. in Virginia and Mosel-Vitelic Inc. in Taiwan.
"The surprising part about it is that over the past two years, they've invested quite a bit of capital in upgrading their facilities and trying to make a big push into getting more market share and becoming a leading force in the DRAM industry," said analyst Jim Feldhan of Semico Research Corp., Phoenix.
But von Pierer conceded that the industry's latest cycle-perhaps the most punishing in history-has taken its toll.
"The semiconductor business is marked by high yet unsteady growth, with huge fluctuations in earnings," he said. "This has a major impact on our company's overall results."
Indeed, the semiconductor unit posted a pretax loss of $723.1 million, compared with a $65.7 million profit last year, the company reported.
The businesses in Siemens' components division, consisting of semiconductors, passive components and electron tubes, and electromechanical components, amassed combined sales of roughly $6.6 billion and employed about 47,000 during the recent fiscal year, which ended Sept. 30.
Those business groups, von Pierer said, will first be converted into separate legal entities. Then the semiconductor group, and possibly the passive-components and electron-tubes group, will be listed on a public stock exchange.
For the electromechanical group, the company is seeking a strategic partner, according to von Pierer.
Of the business groups that will be affected by the divestment plan, all but the semiconductor group were profitable during fiscal 1998.
Also, von Pierer said, several non-memory segments of the group have performed well.
"In a number of areas-such as high-frequency ICs and power semiconductors-[the group] has a leading market position and generates value for the company," he said. "The business with chip-card ICs is also strong. In the sector of memory chips, Semiconductors is among the world leaders in costs and product quality."
When pressed for more details about how the spun-off semiconductor unit will be structured, or if Siemens is considering strategic partnerships, company executives deferred comment until Siemens' analysts' meeting next month.
Siemens' decision comes as the latest in a line of semiconductor divestitures over the past several years, and given the current direction in which the semiconductor market seems to be headed, Semico's Feldhan said, it makes sense to spin the unit off now.
"If you look at it from a broader perspective, a number of semiconductor companies have repositioned themselves," he said. "Texas Instruments, for example, divested themselves out of the DRAM business. They've divested 12 different business units and are refocusing that company into DSP and analog."
National Semiconductor Corp. last year sold off Fairchild Semiconductor, a company it created in 1996 to concentrate on its older, commodity-product lines.
The Siemens divesture "is surprising because we hadn't heard any rumors about it," Feldhan said. "But when you do hear about it, it's not all that surprising that they're being repositioned. After a couple of bad years, companies start to look at how they can spin things off to set the stage for the next growth."
Many research firms, Semico included, expect to see a turnaround in the semiconductor market and some DRAM shortages starting sometime during the second half of 1999. And since Siemens said it expects the initial public offering to take at least 12 months to complete, "from the longer perspective, it may be a good time to do it and reap the appreciation over the next 24 to 36 months," Feldhan added.
Meanwhile, executives at Siemens' passive-components and electron-tubes group-which is also expected to go public in about a year's time-contend the divestiture will put the company in a better position to gain share in the commodity passive-components market.
For about three years, the passives group, which has a joint venture with Japan's Matsushita, has been building up capacity in an effort to become a larger player in the commodity passive-components market outside Europe.
"We see this as a great opportunity for us," said Denny Salmang, chairman and chief executive of Siemens Passive Electronic Devices Inc., South Iselin, N.J.
Being freed from its corporate parent, the company will be able to reinvest more of its profits, which Salmang said is the key to its expansion plan.
"We'll also have access to money which comes from the general market, which is normally cheaper than money you can get outside, and that will make it easier for us to make acquisitions," he added.
---
Siemens' DRAM Journey: A Time Line
- May 1996: Announces $1.5 billion 64-Mbit DRAM fab plan for Richmond, Va., in joint venture with Motorola Inc.
- October 1996: Forges strategic alliance with Mosel-Vitelic Inc. and enters into joint venture to build $1.7 billion fab in Taiwan.
- July 1997: Accelerates development of its embedded DRAM technology, moves toward 0.25-micron CMOS process from 0.35-micron, 64-Mbit DRAM process.
- September 1997: Christens new 0.35-micron fab in North Tyneside, England; extends partnership with Mosel-Vitelic.
- August 1998: Begins phasing out production at
North Tyneside fab, plans to shutter plant by
year's end.
- November 1998: Announces spin-off of semiconductor group and other businesses in components division.
techweb.com |
