Mannerisms

Then, the big UK companies, GEC, Plessey, Ferranti, Thorn, EMI and STC were indulging in mergers and mayhem.

Today the Japanese majors are struggling with the same forces which turned the UK majors into industrial dinosaurs.

In the UK, the result was the demise of the majors, and the emergence of a plethora of start-ups, as top-flight engineers went off to do their own thing, in their own way.

The pressures on the Japanese electronics industry are threefold: increased Asian competition cutting into, or eliminating, profitability; the urgings of the government to restructure by merger; and the outside pressure of financial institutions hoping to promote M&A for reasons of their own (i.e. fees from M&As and profits from buy-outs).

As in the UK, the initial brunt of the pain in Japan is being visited on the capital-hungry, erratic, semiconductor sector. The formation of Renesas and Elpida, the spinning off of NEC Electronics as an independent company, the sale of Sanyo Semiconductor, the likely absorption of much of Sony Semiconductor by Toshiba, are the tangible results of these pressures.

No one doubts there will be more restructurings to come. The government was urging the merger of Toshiba and Fujitsu three years ago.

NEC Electronics is effectively in the position that it will have to go fabless, or share a fab, at the 32nm node.

Fujitsu’s semiconductor business survived as an IDM at the 90/65/45nm nodes by getting customers to pay up-front for foundry wafers. At that time, under a CEO who was a former process engineer, Fujitsu had an exceptionally low power process which was sufficiently attractive to make the foundry proposition attractive.

Now, under a CEO with a design background, the word is that the company is late in defining and choosing the materials for its 32nm node, and may not be able to repeat that model.

That means Fujitsu will also have to go either fabless, or for a shared fab.

The only Japanese semiconductor company which can afford a fab at 32nm is Toshiba and, at the moment, it doesn’t appear to be terribly keen on sharing it with anyone.

Meanwhile venture capitalists are waiting with ready money to snap up disenchanted, or laid-off, engineers who are looking for start-up opportunities. And many engineers inside the big companies are said to be disenchanted as, no matter how hard they work, their companies don't do any better. So they question the judgment of management.

So far, however, the start-up situation is in its infancy in Japan. There have been pitifully few semiconductor start-ups which have got to IPO: THine, Megachips, RealVision, Axell. That’s about it.

But the Japanese government-backed VC fund ENOVA has invested in six semiconductor start-ups, two of which are close to IPO, and JASVA (Japan Semiconductor Venture Association) has over 30 privately-held device manufacturers as well as software companies, EDA companies and equipment companies numbering 230 in all, 100 of which are start-ups.

So, as the majors stutter, big changes could be on the way for the Japanese semiconductor industry.

As in the UK, it could be the beginning of something much more dynamic and exciting than what went before.