One Step Forward for Japan's Chipmakers

The country's Big Five companies have survived the tech drought by slimming down. Now, they need to start growing again

By John Yang

For the past two years, the top five Japanese chipmakers -- Hitachi, Toshiba, Mitsubishi Electric, Fujitsu, and NEC -- have struggled. After the technology bubble burst in 2001, the price of dynamic random access memory (DRAM) chips hit rock bottom, and inventory piled up as customers cut back on spending. To make things worse, a hopeful bet on cell-phone sales overseas -- namely in the U.S. -- fell short. In the fiscal year ending March, 2003, the Big Five chipmakers posted a total net loss of $934 million (based on an exchange rate of 120 yen per U.S. dollar). That's still quite an improvement vs. a loss of $13 billion a year ago.

While Japan's chipmakers were mired in the downturn, their counterparts in Taiwan and Korea posted record profits. Chipmakers in those countries fared better because they concentrated on their core competencies, while the Japanese continued to make a wide portfolio of semiconductors for products ranging from power generators to home appliances. Once the Big Five realized they could face everlasting sleep if they didn't adapt to the changing marketplace, they restructured their business models.

First, they spun off low-return assets, decreasing them by 12% over the last two years. Fujitsu (FJTSY ) was the most aggressive, trimming its total by 19%. Semiconductor unit spin-offs aided the group's balance sheet the most, helping to lighten capital-spending costs. On average, upgrading a semiconductor division's fixed assets account for 40% of total capital-spending costs.


  For example, Hitachi (HIT ) and Mitsubishi Electric (Melco) (MIELY ) created a joint venture in 2002 called Renesas, which focuses on producing large-scale integrated circuits. Hitachi went one step further, creating the Elpida venture with NEC (NIPNY ), which is considered the last pure Japanese DRAM maker since the others quit the business.

NEC also plans to sell 34 million shares of its semiconductor company, NEC Electronics, on the Tokyo Stock Exchange on July 24. NEC Electronics focuses on research, development, manufacturing, and sales of chips that aren't commodity DRAM, and this could be one of Japan's largest IPOs this year.

Toshiba (TOSBF ) and Fujitsu announced an alliance last year but haven't set concrete plans. The latter realized that memory chips were becoming a financial burden and transferred its flash-memory business to a joint-venture company controlled by Advanced Micro Devices (AMD ). As a result, Fujitsu's semiconductor portion of total capital spending dropped to 25%, from 40%.


  The Big Five also reduced fixed costs by 15% to 20% over the last two years, implementing measures such as early retirement for senior employees and closing overseas factories. Fabrication plants were moved from Japan to China and other countries where labor is cheaper. For example, profits at NEC's cell-phone business improved as NEC used Chinese fabs more aggressively. These efforts have boosted the Big Five's total operating income to $5 billion in the fiscal year ending March, 2003, from a loss of $4 billion a year earlier.

Furthermore, the group has made notable progress in reducing total debt. Since March, 2002, combined total debt has declined 14% from $83 billion, which is quite impressive. NEC, the leader in improving its balance sheet, shaved 34% from its previous debt of $19 billion. Likewise, Melco trimmed its debt by 24%. All five outfits also reported operating income (as opposed to losses), and their interest-coverage ratio benefited dramatically -- another sign that the group's credit risk is improving.

Management has done a good job defining and pinpointing core competencies, too. In previous years, a herd mentality led the Big Five to the same markets that included DRAM chips, cell phones, and system integration. After the restructurings, it's clear where each company is making its bet. For example, having purchased IBM's (IBM ) hard-disk-drive operations, Hitachi will focus on storage and microstorage devices such as 2.5-inch and 1.5-inch hard drives.


  Toshiba will continue to muscle ahead with NAND-type flash memories (NAND refers to a Boolean logic function that describes the chip's circuit design) and discreet devices. Melco will zero in on old-economy products, such as power generators and industrial-automation equipment. And NEC and Fujitsu will continue to focus on telecom and networks.

Another clear trend is that the Big Five are shifting away from commodity businesses or any product that's prone to become a commodity in the near future. Except for Elpida, Japanese chipmakers basically withdrew from the DRAM business. They also shrunk the number of chips they make for home-appliance products.

Unfortunately, even after these changes, we at Standard & Poor's still see rough times ahead for the Big Five. Although total debt has come down, credit risks remain. The Big Five's average debt-to-equity ratio of 2 is alarming to us. For NEC alone, debt is 3 times equity. In comparison, many American chipmakers carry zero debt, and very few have debt-to-equity ratios more than 1.

Second, restructuring measures taken by the five companies were aimed at survival rather than stirring growth. In the process of stripping out certain assets, the chipmakers may have lessened their competitive edge. Lastly, the Japanese still lack true earnings drivers. Given these factors, we're not ready to recommend that investors buy these companies as bargains yet, regardless of their low share prices.


  We have a 3 STARS (hold) recommendation on shares of Mitsubishi Electric, a 2 STARS (avoid) ranking on Hitachi, and a 1 STAR (sell) ranking on Toshiba, Fujitsu, and NEC.

The Big Five's collective share prices have dropped more than 60% since peaking in the summer of 2000. NEC and Fujitsu have been hardest hit, dropping 79% and 88%, respectively. Prices have gained momentum since mid-June, reflecting the pull from U.S. markets and a sense of relief that Tokyo won't let the country's banks go belly-up again, like in 1997-98. Still, we don't think the current price-to-book of 2 times can be justified, given the credit risk and weak earnings outlook for the Big Five. Therefore, we foresee a major correction in share prices once the summer rally is over.

Analyst Yang follows Japanese semiconductor companies in Tokyo for Standard & Poor's

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