Done Deals, and Those Not Taken

Will the planned
mega-merger between NEC Electronics and Renesas Technology work
out? The deal creates a $13B semiconductor company a year from now,
one that is about the same size as current No. 3 Texas Instruments
and No. 4 Toshiba Corp. That $13B is an impressive number. But as
Tom
Starnes, an analyst who follows the MCU market at Objective
Analysis, points out, bigger is not necessarily better. “It
could be that Renesas is too big already,” Starnes said.

“Is this a solution?”
Starnes asks, pointing to the struggles at Renesas, formed from a
merger between Hitachi, Ltd. and Mitsubishi Electric Corp., to
do more than just field parallel product lines, which he calls
“cohabitation” rather than a true merger. Starnes, a
former Motorola engineer and ex-Gartner MCU/DSP analyst, said he
has been impressed with what NEC Electronics has done recently to
create new offerings for automotive applications. But in both
automotive and general-use MCUs, NEC and Renesas will have
duplicate product lines. In 32 bitters, for example, NEC offers its
V series, while Renesas has its Super H and M32 product lines.

With Japan’s tradition of
consensus decision-making and the post-war culture of lifetime
employment, it may be difficult for NEC and Renesas to fully merge
their product lines, fabs, and personnel in a reasonably quick
manner. Japanese companies believe in doing things well, even if
that requires more time to get the job done. Some call that
over-engineering, some admire the devotion to quality, but
that’s the way it is.

In a sense, NEC and Renesas are
doing some of the hard work prior to the merger. The companies have
announced
separate plans to close older fabs and sharply reduce staff.
The plan is to cut as much fat as possible before the merger
concludes next April, the start of the fiscal year in Japan.

The merger will mean the end of NEC
as a stand-alone semiconductor producer, by itself no great world
tragedy, but worth noting. Starting in the mid-1980s,
NEC reigned as the world’s largest chip producer (see
Wikipedia’s good year-by-year history of the largest chip
vendors.)

NEC was the largest semiconductor
company until 1992, when Intel vaulted three spots to take over the
lead as personal computers took off. And then NEC remained the No.
2 chip vendor, behind Intel, for seven more years, until 1999. NEC
got to the top, and stayed there for a long time. By 2006 it had
dropped out of the Top Ten chip vendors, and racked up huge losses
for four years running.

Hajime Sasaki,
who ran NEC’s semiconductor operations from the top of
the NEC Super Tower, once was asked why NEC’s star fell so
fast. NEC, Sasaki said, was like “a dog that tried to chase two
rabbits,” the memory and logic/MCU businesses.  

A merged NEC-Renesas really needs a
tough-minded, smart CEO from outside the two companies. Yukio
Sakamoto, currently the CEO of Elpida Memory, would be a good
choice. When Sakamoto worked at Texas Instruments Japan, he became
known as the person to turn to to get tough restructuring jobs
accomplished, according to former TI executive Wally Rhines, now
the CEO of Mentor Graphics Corp. When a customer needed some
impossible task accomplished quickly, Sakamoto would tell his TI
bosses, “Don’t worry, I’ll take care of
it.” And he did, over and over again, Rhines recalled.

The NEC-Renesas merger announcement
brings to mind all the deals which haven’t worked out, and
another set of deals which the principals probably regret not
making. AMHS vendor Asyst Technologies, now in voluntary Chapter
11, rebuffed two takeover attempts early last year. Axcelis spurned
its suitor (Sumitomo Heavy Industries), as did ASM International
(Applied Materials). Given the current downturn, do those
companies, and their shareholders, wish they had been more willing?
Probably.

Starnes, however, makes the point
that most semiconductor mergers and acquisitions just don’t
work out. Two merged companies that have separate fabs don’t
find it cheap or easy to come up with a common process technology,
for example. Even STMicroelectronics, which eventually achieved
success as a merged entity, still has its Italian and French
factions.

The engineers at NEC Electronics are
no less likely to give up their company’s still-proud banner
next April. Nor should they. Who knows? If the world wakes up and
moves quickly to electric vehicles (which depends more on rapid
progress in battery technology), NEC-Renesas may contend for
greatness again.