The Fairchild Deal : Trade War: When Chips Were Down

WASHINGTON â€”  To some it was like selling Mount Vernon to the redcoats: Fairchild Semiconductor, a pioneer of America’s high-technology industries and the mother company of Silicon Valley, was being purchased by foreign rival Fujitsu of Japan.

It was, in business terms, a friendly takeover. Executives of the ailing California computer chip maker had gone to Japan in search of a financial savior. Fujitsu had been invited, even urged, to bid.

But to then-U.S. Commerce Secretary Malcolm Baldrige the deal was bigger than a $200-million transaction between two private companies. At stake was America’s future--its military security, its economic health and its technological leadership.

Attacked Sale

In a move that insulted Japan, angered some in the Reagan Administration and generated shock waves that continue to ripple through the international trade and investment communities, Baldrige attacked the sale as a threat to national security.

Over breakfast with reporters at the Sheraton Carlton here last March, Baldrige “drew a line in the sand� and told Japan it could not cross without a fight. His message: America’s high-technology industries, already bloodied by what he regarded as predatory trade practices, had to be protected from Japanese acquisition and the battle would start here, with the Fairchild-Fujitsu deal.

“I’ve come out against it,� Baldrige said simply. “That’s something that gives me great problems.�

From the moment Fujitsu announced on Oct. 23, 1986, that it was buying an American corporate icon, the Fairchild deal had been a business transaction entangled in politics. Administration “techno-hawks� saw it as “part of a master plan of the Japanese to take over the U.S. information industry.� But White House economists worried about who would underwrite the national debt if Japanese investors and their money felt unwelcome here.

Ideological Brawl

The debate within the Administration dissolved into what some called an ideological brawl, “an unholy war� waged in meetings of the Committee on Foreign Investment in the United States (CFIUS)--a little-known interagency panel chaired by the Treasury Department. Even the Justice Department’s routine antitrust review became embroiled in the political conflict.

The issue was going to the Cabinet the morning that Baldrige had breakfast with reporters; the morning that a $200-million business deal between two private companies exploded under the pressures of international politics. Five days later, Fujitsu abandoned its bid.

The U.S. semiconductor industry, worried about a “domino effect� and who would be next if Fairchild ownership went to Japan, cheered the news. A sentimental symbol of the industry had been kept out of Japanese hands. The first domino hadn’t fallen. Not yet.

The extraordinary intervention also slashed Fairchild’s market value and cleared the way for U.S. rival National Semiconductor to acquire Fairchild at a “fire sale price,� about half the original Japanese offer. That raised questions about who benefited most: the country, the computer chip industry or National Semiconductor.

How Fairchild, a past-its-prime computer chip company, became a pawn in the U.S.-Japan trade war is a study in fear--corporate survival fears, national security fears, economic fears and an underlying distrust of the Japanese.

Chapter II

Is Nothing Sacred?

Fairchild. In the annals of high-technology, its very name is mythic. Its roots crowded out the plum orchards of California’s Santa Clara Valley and replaced them with industrial plants; prunes gave way to the new cash crop--silicon.

Fairchild begat the Silicon Valley. Founded by eight engineers from a Palo Alto company who left to start their own shop with a lot of brash ideas and a little venture capital, Fairchild Semiconductor became a magnet for brilliant electrical engineers and savvy young businessmen; it nurtured them in their apprenticeships and then was the model when they struck off on their own ventures. Fairchild was a breeding ground for ideas and people who would change the world.

It was at Fairchild that a young engineer named Robert Noyce developed the first practical integrated circuit, a miracle of design that could store, channel and manipulate millions of electrical signals on a fingernail-sized sliver of silicon, the basic ingredient of sand.

The integrated circuit--or semiconductor, or computer chip--was the foundation for a new world of miniaturization. Now room-size computers fit on desktops, calculators into shirt pockets and televisions onto wristwatches.

Varied Uses

Chips control pacemakers, video games and guided missiles. They put voices in automobile dashboards, intelligence in home appliances and men into space. Without it, any hopes for a Strategic Defense Initiative would be nothing more than another “Star Wars� fantasy.

In recent years, Fairchild had fallen behind--some called it “the dog of the industry.� The technology race was being led by its offspring, companies founded by bold, independent Fairchild alumni who came to be viewed as the “Eli Whitneys, the Henry Fords and Du Ponts of the day . . . the new captains of industry.�

Noyce and Gordon E. Moore, another Fairchild co-founder, had left to establish Intel. Charles E. Sporck, the pragmatic production expert, had been lured away to save a near-bankrupt National Semiconductor. And W. J. (Jerry) Sanders III, the flamboyant marketing genius, had left to found Advanced Micro Devices. Intel, National Semi and AMD had surpassed Fairchild and had become the “big three� of the valley’s chip business.

Fairchild had been the world’s top computer chip maker in 1966. But it was a legend in trouble in 1983 when Donald W. Brooks of Texas Instruments was called to the rescue. Schlumberger, the French-controlled oil field services firm that bought Fairchild in 1979 and tried to run it like a heavy equipment company, had finally recognized it needed veteran semiconductor management to reverse its decline.

Brooks found a company lagging in leading-edge technologies. Fairchild was losing money even as the rest of the semiconductor industry was booming.

But the boom of 1983-1984 masked some ominous trends for the entire industry. The Japanese had already entered the market for memory chips--common kinds of semiconductors known as commodity chips--and had done so with the support of the Japanese government. On both sides of the Pacific a massive buildup of manufacturing capacity was setting the stage for an international chip glut.

Lulled by long market dominance and technological leadership, U.S. companies were neglecting quality problems and inefficiencies in manufacturing.

The Japanese companies--some of the same conglomerates that already had taken over home appliance and consumer electronics markets--used U.S. technology and their own superior production methods to make better and cheaper memory chips. The conglomerates offered built-in customers for Japanese chip-makers while trade and cultural barriers kept U.S. firms out of Japan’s lucrative chip markets.

Lower Interest, Taxes

Japanese companies also enjoyed lower borrowing rates and tax advantages. They could take early losses to capture markets and thus gain sales volume for eventual profits.

As a pricing war escalated, U.S. chip makers cut into their profits to compete, but the Japanese reached into the deep pockets of their conglomerate parents and lowered prices even more. To the American companies, this was “dumping�--selling chips below the cost to make them.

Noyce, whose Intel saw the price of a $17 chip fall to $4.50 in eight months, warned that the Japanese intent was “quite clear--they are out to slit our throats.�

American companies were staggered. In the crucial market for memory chips called D-RAMs, for example, Japanese companies went from a 5% share to control 90% in only four years. By the mid-1980s, American companies had virtually abandoned the commodity chip market, many moving to more sophisticated devices with higher prices but smaller markets. Fairchild was among them.

Then the boom turned to bust. Orders from computer makers slowed to a standstill, but the plants kept churning them out. Suddenly, the market was flooded with cheap chips.

In 1985, the U.S. semiconductor industry lost more than 40,000 jobs and its five largest manufacturers recorded $343 million in losses only a year after profits of $1.3 billion. By the end of that year, U.S. companies were clinging to a narrow 49% to 42% lead in markets for all types of chips.

As competitive hostilities were heating up, National Semi’s Sporck said: “We’re at war with Japan--not with guns and ammunition, but an economic war with technology, productivity and quality.�

There were renewed calls for U.S. government intervention from some of the most independent-minded business executives on the globe. Many, like AMD’s Sanders, said they wanted Washington to declare that the U.S. semiconductor industry was “sacred . . . and we will not permit it to be destroyed.�

An early partisan of the semiconductor industry cause was Mac Baldrige, the commerce secretary. In an unprecedented move, Baldrige filed an anti-dumping claim against the Japanese. And with U.S. Trade Representative Clayton K. Yeutter in 1986, he pushed through a “hellishly difficult� trade agreement in which the Japanese promised to end dumping and open their markets to the United States.

An uneasy truce in the trade war prevailed in the summer of 1986. U.S. resentment still lingered toward companies like Fujitsu, earlier found guilty of dumping chips at 60% below fair market value.

But to Brooks at Fairchild, Fujitsu was no enemy--Fujitsu was the “white Samurai.�

Schlumberger, beset by troubles in the oil industry, was becoming more impatient with its loss-ridden high-tech subsidiary. Concerned that Schlumberger might dismember or completely shut down Fairchild, Brooks initiated talks with Fujitsu.

This time it was Brooks looking for a rescue. After weeks of secret negotiations, Fujitsu offered a $200-million lifeline. It would buy Fairchild.

Warren Davis, a lobbyist for the Semiconductor Industry Assn. (SIA) and a Fairchild alumnus, summed up valley reaction: “Fairchild is the mother company of Silicon Valley . . . is nothing sacred?�

Chapter III

Of Chips and Politics

Brooks, the Texas transplant, had never been one of the good ol’ boys of Silicon Valley. He hadn’t started his own company like Noyce or Sanders or LSI Logic’s irascible Wilfred Corrigan, another former Fairchild executive. Brooks kept Fairchild outside the SIA. And he avoided any role in the industry’s political activities.

But even as an outsider, Brooks sensed that the Fujitsu purchase would be poorly received in his neighborhood. What he underestimated, however, were the obstructions it would encounter in Washington.

In the weeks before announcing the sale, Brooks and his senior executives conferred privately with veteran Washington lobbyists who recommended “a very strong public relations campaign.�

They noted growing resentment of Japan in the Administration and Congress over the dumping issue, the intractable trade deficit and frustrating trade barriers. What some called “the flow of history� was working against Brooks--whether he knew it yet, or not.

“I could smell things beginning to happen,� recalled one of those advisers, who characterized Brooks as politically naive.

The education of Donald W. Brooks began immediately.

Brooks summoned Andrew Procassini, president of the SIA, to a private briefing at Fairchild’s Cupertino headquarters just hours before announcing the deal with Fujitsu.

Sought Assurances

“I want to be assured that the SIA won’t take a negative position on this,� Brooks told him. Procassini shrugged and agreed that it was a private matter between the two companies. The SIA would not interfere, he promised.

Despite Procassini’s assurances, there was enough concern in the valley that within a week several industry executives--including Procassini, Sporck, Noyce and Corrigan--met at National Semiconductor to discuss how the SIA should respond.

Corrigan, who said the Japanese business strategy is to “obliterate competition,� warned that the bid for Fairchild was “like a probe . . . being observed by other major Japanese companies.� If Fujitsu succeeded, according to Corrigan’s “domino theory,� by the end of 1987 “all the major Japanese companies would be coming in� to acquire U.S. semiconductor manufacturers.

Outside that Oct. 31, 1986, meeting, Sporck told reporters he objected to the Fairchild sale because it was too favorable to Fujitsu and was “not a good deal for the U.S. semiconductor industry.� But privately Sporck and his colleagues decided the SIA would not engage in any official campaign to block the deal.

And Donald W. Brooks left for Washington in the closing days of October, 1986, confident that he had saved his ailing company. He had only to silence the distant rumblings of bureaucratic objections. When he reached the Potomac, however, Brooks discovered that he and Fairchild had stepped from the financial frying pan into the political fire.

He was summoned first to the Defense Department, where he was introduced to a skeptical Stephen D. Bryen, the deputy undersecretary known around the Pentagon as “the prince of doom and gloom.� Bryen quizzed Brooks about national security concerns and about the goals of the Japanese.

It was Bryen’s view that the Fairchild purchase could be “the opening gun� of a Japanese barrage aimed at a vital U.S. high-technology industry. “If one of the flagship companies of our semiconductor industry could fall into the hands of the Japanese, we could end up with no U.S. semiconductor industry. We could lose the technology race by default,� he said.

Another stop was at 600 17th St., a small, unpretentious building a block from the White House--the office of Ambassador Michael B. Smith, the deputy U.S. trade representative. Smith and Bruce Smart, one of Baldrige’s top aides, met Brooks together. The two officials had been primary negotiators on the 1986 chip agreement with Japan, veterans of the simmering trade war. They told Brooks the Fairchild sale was a bad idea.

“They gave me some indication that it was an unhappy situation with Japan and that this was another blow,� Brooks would recall later.

The Fairchild executive was learning what one American lobbyist for Japanese clients already knew: “The Japanese have to deal with a lot of suspicion in this town.�

Brooks’ hostile reception was building before he got to Washington. The very day newspapers reported the Fujitsu offer, Charles Ferguson of the Massachusetts Institute of Technology’s Center for Technology, Policy and Industrial Development picked up the phone and began ringing alarm bells throughout the capital.

He called the Defense Department, the White House, contacts in the intelligence community--and Smith. Like a Paul Revere warning that “the Japanese are coming,� Ferguson said Fairchild was too important to be lost to Japan.

“It was a profoundly disturbing situation,� said Ferguson, who has written extensively about the competitive decline of the U.S. semiconductor industry and its strategic implications. “It wasn’t in the national interest for Brooks to sell out to Fujitsu.�

Ferguson said it was an illustration of “lifeboat diplomacy,� an attempt to save Fairchild first while betraying the American semiconductor industry--and perhaps the national interest.

The message fell on receptive ears. At both the Commerce Department and U.S. trade representative’s office, aides began investigating the issue immediately. And at the Defense Department, Bryen told his aides bluntly: Stop the sale.

Government attorneys were asked to find any legal grounds for blocking the sale. The Pentagon ordered its general counsel to respond within 24 hours.

“The DOD general counsel ended up working on it for three or four days. It didn’t matter. The fact was, except in extreme cases, there was no such authority,� said a former Administration official.

It also was apparent early that any antitrust concerns would be insufficient to stop the deal. A Fairchild-Fujitsu combination would not have controlled enough of any segment of the chip market to warrant legal action.

Nor could serious objections be raised over its defense contract work. Although 35%-45% of Fairchild’s sales were to defense contractors, the company was not the sole technology supplier to any of them. Also, it was not directly involved in any classified projects. And defense concerns over foreign ownership already had been resolved in 1979 when French-controlled Schlumberger bought Fairchild and established a New York-based subsidiary to operate it.

‘Red Herring’

“The national security issue was a red herring,� said one Administration official.

The last resort for opponents of the Fairchild sale seemed to be “the old guy with no teeth�--the Committee on Foreign Investment in the United States (CFIUS).

It may be one of the oddest creatures in the federal bureaucracy, a committee designated to monitor and investigate foreign investments but not empowered to intervene. CFIUS was born in the aftermath of the Arab oil embargo amid fears the oil cartels would use their burgeoning cash reserves to buy America.

The interagency panel is composed of high-level representatives from the departments of treasury, state, defense and commerce, as well as the U.S. trade representative and the Council of Economic Advisers.

Inevitably, CFIUS gets only the most politically sensitive transactions. Fairchild clearly qualified on that score. In fact, Fairchild was the only case singled out for investigation in 1986 from among $209.3 billion in foreign investments. The $200-million sale represented less than 0.1% of that total.

In the decade since its creation, however, CFIUS had never opposed a single foreign investment. On the rare occasions when substantial policy questions could not be resolved, the panel sent its recommendations to the Cabinet. But neither had the Cabinet ever discouraged a foreign takeover.

No laws bar foreign investment and, in fact, Reagan Administration policy formally welcomes such investment. In 1983, President Reagan called a “free and open international investment climate� the key to U.S. and world economic recovery.

“Foreigners had a huge role in building this country. The one thing we’ve never done in 200 years is prevent foreign investment,� said R. A. Cornell, deputy assistant treasury secretary, a top CFIUS official.

But in late October, 1986, Baldrige and top U.S. trade officials--some of whom had helped draft the Reagan open investment policy--believed the world had changed and that it was time to reevaluate whether some critical American companies should be off-limits to foreign control.

That question would be at the heart of the CFIUS debate.

But as its Fairchild review began, CFIUS appeared headed in the same direction it had always gone: away from intervention.

The sides were entrenched from the beginning. Treasury, the State Department, a significant faction of Pentagon officials and virtually all Administration economists lined up behind the open investment policy and disputed the techno-hawk view that Fujitsu should be an exception.

If the Baldrige camp was going to stop the Fairchild sale, it would have to win some converts fast. The Justice Department antitrust review was headed for a mid-December deadline and its anticipated ruling in Fujitsu’s favor would clear the way for the sale to be consummated. Debate would be moot.

“We all wanted to beat Justice,� said a trade official.

Chapter IV

The Winter Impasse

The winter’s night was closing in on Washington and still Mike Smith couldn’t go home. At his desk in the warmly lit office he drafted one more memo--a letter to Frank C. Carlucci, then the newly appointed national security adviser (he was to become secretary of defense late this year). It was a call for help.

Smith, the deputy U.S. trade representative, was deeply troubled by the Fairchild sale. He thought Washington should be troubled by it, as well. There was more to national security, he was arguing in CFIUS debates, than inventories of bombers, bullets and missiles. Economic security was equally vital.

“We can’t be a world power if we don’t make our own semiconductors,� Smith had been telling CFIUS.

It was a chilly walk past the White House from Smith’s office to the Treasury Building where CFIUS met in a paneled fourth-floor conference room. The meetings were not going well.

From the outset, Smith and his Commerce Department counterparts found themselves butting up against the Administration’s open investment policy, a formidable obstacle to winning the hearts and minds of aides to Secretary of State George P. Shultz and Treasury Secretary James A. Baker III, both strong proponents of that policy.

It was Smith’s position that the open investment policy should be linked to open trade policies, that Japanese violations of trade agreements and unkept promises to open Japan’s markets to the United States should be considered sufficient grounds for economic retaliation--including investment restrictions.

‘Unjust Enrichment’

Baldrige and his Commerce Department objected to what the secretary called “the unjust enrichment� of Japanese companies that had engaged in predatory trade practices. Now those same companies, their pockets bulging with cheap dollars, were coming in to buy a weakened member of the very industry they had damaged. The entire industry seemed vulnerable.

“Mac (Baldrige) thought if we let them get away with this, we were sending an awful signal to Japan. It could be the beginning of the end for our industry,� an Administration official said.

The Pentagon’s Bryen, routinely accompanied to the CFIUS meetings by a uniformed Navy captain, hammered at the national security issue. He didn’t want to lose a U.S. chip maker to the Japanese. Besides, he felt Japan’s failure to take security seriously had allowed some sensitive U.S. technology to leak to the Soviet Union.

Also, it was Bryen who raised “the Cray argument,� an issue that would become central to the antitrust investigation.

Eliminate Competition

Cray Research Corp. of Minneapolis, the world’s leading manufacturer of supercomputers, relied on Fujitsu and Fairchild for its essential emitter-coupled logic (ECL) chips. The merger would have made the Japanese firm the world’s top supplier of ECL chips and eliminated a competing source for Cray.

The Pentagon’s super-secret National Security Agency, which relied on Cray supercomputers, worried that because Fujitsu also made its own supercomputers it might withhold the vital ECL chips and keep future advances to themselves.

“The NSA was very disturbed,� Bryen said. So was Bryen.

Other members of CFIUS, however, came to be disturbed about what they regarded as Bryen’s strident advocacy of the national security threat. One Administration official referred pointedly to Bryen when he complained about the “disgusting level of debate that sometimes lacked an intellectual base.�

And another Administration official said: “Some of us got real tired of having our patriotism impugned by that man� for not taking a hard line against Fujitsu.

‘Japan Bashing’

“It was sheer Japan bashing,� complained another critic who opposed U.S. intervention.

Bryen did not speak for the entire Pentagon. Richard L. Armitage, assistant secretary of defense for international security affairs, represented a sizable faction that supported the Fujitsu-Fairchild merger.

Armitage, involved in negotiations to get Japanese technological support for the Strategic Defense Initiative, had argued before CFIUS that Fujitsu could contribute new technologies to Fairchild that would, in turn, be available to the U.S. military.

The argument underscored one of the ironies of the Fairchild controversy: that its purchase by French-controlled Schlumberger in 1979 had not been opposed.

One Defense Department consultant said: “This Japanese company could infuse a lot of technology into a failing Fairchild. Why was Fujitsu ownership bad when it was OK for a French company, that couldn’t give it technology, to buy it?�

Opponents scoffed at Armitage’s theory of “reverse technology flow.� MIT’s Ferguson called it as likely as “converting a panther to vegetarianism.�

The hard positions of both sides showed no signs of changing when Smith stayed late to write the letter to Carlucci. Smith, seeking to expand the debate, wanted the National Security Council to recognize that economic issues were significant enough to warrant a national security review of the Fairchild deal.

If Smith and his allies could establish that Japanese ownership of a major U.S. semiconductor company might affect the nation’s economic health through a loss of technological leadership, it could be grounds for high-level intervention despite the Administration’s open investment policy.

The NSC letter immediately succeeded in expanding the CFIUS panel as NSC aides joined the table. So did aides from the Office of Management and Budget, the Labor Department and the White House Office of Science and Technology Policy. As the panel expanded, so did debate.

“Fairchild was like a clay pigeon,� Brooks would later recall. “All these agencies were shooting at it.�

CIA Officials Present

Also present for various CFIUS exchanges were people from the CIA, National Security Agency and the National Science Foundation.

And the Justice Department, which was conducting the independent antitrust review, sent an observer to monitor the CFIUS review. Opponents of the deal raised a wide range of objections designed to expand and, inevitably, slow the antitrust inquiry.

“We did everything we could within the law to delay that process,� said a leading opponent of the sale, who conceded, “I don’t think there ever was an antitrust problem� with the deal.

Additional Data

When the transaction became “a political animal, there was no way in hell the Justice Department would come out� with its antitrust finding before the Administration debate was resolved, a Pentagon official said.

In mid-December, when the antitrust review deadline expired, Justice demanded a massive amount of additional data that would take Fairchild and Fujitsu months to assemble. The delay gave foes of the deal time to press for Cabinet intervention.

“Justice would have had to find in favor of Fujitsu,� conceded one opponent. “But they were smart enough to recognize a gnashing of souls and they were willing to accommodate us for a time.�

Experienced Washington attorneys representing the companies knew exactly what was happening but could do little except counsel patience. “The (antitrust review), process was being grossly abused--they were jamming the door with it, keeping the deal from closing,� one of the lawyers said.

A Justice Department official denied that the second request was intended to delay the transaction or that outside agencies improperly influenced the antitrust process, but he acknowledged other agencies recommended areas for possible investigation.

‘An Ax to Grind’

“I’d be hard-pressed to find any antitrust case where someone--everyone--doesn’t have an ax to grind. That’s something we deal with all the time,� said the official, who asked not to be identified.

At the same time, several members of Congress were fretting over the implications of the Fairchild-Fujitsu deal. Sen. J. James Exon (D-Neb.) wrote letters to several Administration officials warning that the U.S. technological lead over the Communist Bloc was being eroded and that the semiconductor industry was crucial to national defense.

Exon urged Trade Representative Yeutter and other CFIUS participants to “take action to preserve at least a minimum and prudent level of (America’s) vital defense industry within our borders and jurisdiction.�

And Sen. Howard M. Metzenbaum (D-Ohio) voiced fears in a letter to the Justice Department that the Fairchild acquisition “could permit Fujitsu to dominate the American supercomputer market.� It was a worry shared by several in the Administration, especially Baldrige.

In fact, supercomputers became a focal point of the debate over whether to block the Fairchild sale. In addition to national defense concerns, Baldrige worried that Fujitsu would use its new American company to sell its own supercomputers in the United States, undercut other American competitors without fear of trade law violations and muscle in on the young, lucrative market. Bryen said it was a way for Fujitsu “to launder� its nationality, a way “to look American.�

It was a scenario for the destruction of the U.S. supercomputer industry--a repeat, perhaps, of the demise of America’s once-healthy television-set industry.

While Baldrige and some in Congress gave the matter great urgency, industry experts knew it was unlikely that the Fairchild deal posed such a threat. Fairchild, indeed, had a valuable distribution network--but for $15 chips, not $15-million supercomputers.

“It’s like saying I’m buying a Toyota dealership in Orange County to sell supercomputers,� said an attorney for the firms.

Complex Backdrop

Into early 1987, the Fairchild drama was being played out against a complex backdrop of international politics.

--The hard-won semiconductor trade agreement, which took effect a mere two months before the Fujitsu bid was announced, already was faltering. The Administration was considering trade sanctions against the Japanese for continuing to dump chips and failing to open their markets.

--U.S. trade representatives angrily broke off negotiations with Japan on a range of subjects, including supercomputers and telecommunications, after an official of the Ministry of International Trade and Industry, Makoto Kuroda, said Japan was unwilling to buy supercomputers from American firms.

--Congress, watching the trade deficit balloon, was threatening to enact retaliatory legislation aimed at the Japanese or any other trading partner that failed to open markets and enforce fair trade practices.

--Sporck and other prominent semiconductor executives were lobbying for $2 billion in federal funding for Sematech, a chip-manufacturing consortium. It was proposed as a joint industry-government response to Japan’s giant high-technology conglomerates that also receive substantial government backing.

On Feb. 19, the top officials of CFIUS met once more on Fairchild. The long, T-shaped table in Conference Room 4121 of the Treasury building was the staging ground for the most contentious session yet. Everyone was there--and so were the wide divisions that had separated them from the outset.

The Office of Management and Budget and the National Security Council had lined up with Treasury, State and the Council of Economic Advisers against intervention. The President’s science policy office was “waffling,� but seemed troubled by the Fairchild sale.

The Defense Department remained split. Armitage and Richard N. Perle, Bryen’s boss, sat side-by-side as the discussion moved around the table. But the two assistant secretaries took opposite positions. No one was certain how Defense Secretary Caspar W. Weinberger would resolve the division.

It was clear to Assistant Treasury Secretary David C. Mulford, who chaired the meeting, that consensus was impossible.

In silence, Smith and Mulford glared across the table at one another. After nearly four months of often “red-faced debate� and political maneuvering, the impasse was complete.

Mulford would send the CFIUS report to the Cabinet’s Economic Policy Committee without a recommendation. The Fairchild question was headed for a showdown at the highest level of the U.S. government.

Also contributing to this story were staff writers Melissa Healy in Washington and Sam Jameson in Tokyo, and library researchers Barclay Walsh in Washington and Susanna Shuster in Los Angeles.