(Archived document, may contain errors)

5/30/96 277

LET THE U.S.JAPAN SEMICONDUCTOR AGREEMENT EXPIRE

(Updating Backgrounder No. 805, "The U.S.-Japan Semiconductor Agreement: Keeping Up the Managed Trade Agenda," January 24, 199 1.) The Clinton Administration and some Members of Congress argue that the decades old U.S.-Japan Semiconductor Agreement has helped save America's ailing semiconductor industry and increase pre- viously slumping sales in Japan. They are wrong. The semiconductor agreement, signed in 1986 by the Rea- gan Administration, nearly destroyed America's computer industry, forced higher prices on American con- sumers, and has had little to do with the U.S. industry's rebound in this crucial technology. The President and Congress should let this agreement expire as scheduled on July 31, 1996.

The 1986 Agreement The United States and Japan agreed in 1986 to monitor foreign semiconductor market share in Japan. There were two reason for this. First, the U.S. industry complained that Japan was protecting its semicon- ductor market from foreign competition. Thus, they argued, the U.S. was not selling as many semiconduc- tors in Japan as it could. Second, the U.S. industry argued that Japanese semiconductor manufacturers were selling these products in the U.S. at prices below their cost of production. I The result was the 1986 trade agreement with Japan, which sought to monitor both the level of foreign semiconductor market share in Ja- pan and the prices of Japanese semiconductors sold in the U.S. Some supporters of the semiconductor agreement argue that it established a "guarantee" by which for- eign semiconductor manufacturers are granted a 20 percent market share in Japan. If foreign market share falls below the 20 percent figure, the U.S. and others can impose trade sanctions on Japan. Yet the 1986 agreement makes no such guarantee. In fact, the market share number is based on a secret side letter that the Japanese government supposedly sent the U.S., stating that it would work to help American companies achieve higher market share in Japan. The U.S. treated the Japanese pledge as binding, but the Japanese

I While the U.S. International Trade Commission found that many Japanese companies were selling their semiconductors in the U.S. at prices below their production costs, there is little evidence that this was done to undercut the prices of U.S. producers. Instead, it was more the result of intensive competition within Japan and a drop in demand for personal computers in the mid- I 980s, which caused the prices for semiconductors to fall rapidly. In order to cut their losses, many Japanese semiconductor firms which had overproduced and had bulging inventories sought to cut their losses by quickly selling off their stockpiles of semiconductors. For more information, see Bryan T. Johnson, "The U.S.-Japan Semiconductor Agreement: Keeping Up the Managed Trade Agenda," Heritage Foundation Backgrounder No. 805, January 24, 199 1.

were more ambiguous, preferring to make it in private. The Japanese government refuses to confirm the ex- istence of this side letter.

Unsatisfied with America's slow progress on increasing market share in Japan, the Reagan Administra- tion had imposed some $4 billion in penalties on the Japanese industry in the mid- 1980s. Most of these pen- alties were suspended in 1986. The Reagan Administration, however, applied some $165 million in retali- atory duties to Japanese imports in 1987. The Japanese were compelled to raise prices for their semiconduc- tors sold in the U.S. in order to avoid the imposition of special tariffs and duties resulting from U.S. anti- dumping investigations. As a result, Japanese computer chips cost American computer manufacturers 30 percent to 40 percent more in the late 1980s than before the agreement. Personal computer prices shot up drastically in the late 1980s, almost destroying America's competitiveness in this crucial high-tech product area.

The 1991 Agreement When the semiconductor agreement expired in 199 1, it was clear that foreign market share had grown only marginally in Japan, from about 9 percent in 1986 to only 14 percent. Nevertheless, the Bush Admini- stration sought to extend the agreement for five more years. This time, the strategy was different. Rather than focus on numerical targets and guaranteed market shares, the Administration backed off the claim that the 20 percent market share number was a guarantee which, if not met, could invoke sanctions. The Admini- stration sought to include a reference to the 20 percent foreign market share figure, but both it and the Japa- nese government specified that this figure was not a guarantee. The 1991 agreement states: The Government of Japan recognizes that the U.S. sen-dconductor industry expects that the foreign market share will grow to more than 20 percent of the Japanese market by the end of 1992 and considers that this can be realized. The two Governments agree that the above statements constitute neither a guarantee, a ceiling, nor a floor on the foreign market share.2 Thus, the 1991 agreement should have removed any doubt that the 20 percent foreign market share fig- ure was to be used as a benchmark in measuring the openness of the Japanese market to foreign semicon- ductors. Yet, despite the text in the negotiated agreement, both the U.S. government and U.S. semiconduc- tor industry continued to focus almost exclusively on this number, leading many economists to call this a managed trade agreement. Even so, the agreement paved the way for U.S. and Japanese firms to enter joint ventures. Both the U.S. and Japanese agreed not to enforce strict antitrust laws which could stifle such rela- tionships, and to sponsor meetings between U.S. and Japanese companies to arrange such relationships. These joint ventures were beginning to occur, even without the agreement, in the late 1980s and early 1990s. They would continue to occur without the agreement. After the 1991 agreement, instead of promot- ing the use of numerical targets, the Bush Administration focused on private sector solutions. The foreign share of Japan's semiconductor market has increased from 14 percent in 1991 to more than 29 percent to- day.

The Cause of U.S. Success The U.S. semiconductor industry is more competitive and prosperous today than it was before the U.S.- Japan agreement. It now is the world's largest producer of semiconductors. The question is: How much does this success have to do with the agreement, and how much is the result of market forces and the sound business decisions of private companies? The answer is simple: America's renewed success in Japan has oc-

2 "Arrangement Between the Government of Japan and the Government of the United States of America Concerning Trade in Semiconductor Products," United States Trade Representative's Office, Washington, D.C., 1986, as amended. curred because of private sector initiatives by U.S. and Japanese companies to work together, and despite the U.S. government's managed trade policies. The 1991 agreement emphasized the use of "long-term relationships between Japanese semiconductor purchasers and foreign capital-affiliated producers, including joint product development with Japanese cus- tomers." This statement gave rise to an avalanche of trans-Pacific strategic alliances or joint ventures be- tween U.S. and Japanese semiconductor companies. Some of these alliances could have been challenged under America's antiquated antitrust laws. But the emphasis on joint ventures in the agreement eased the concerns of many U.S. companies seeking partnerships with Japanese firms. In 1991, for example, there were fewer than a dozen such partnerships between U.S. and Japanese companies. Three years later, in 1994, there were more than three dozen. Located in Japan, many of these alliances gave U.S. firms direct access to Japan's complex distribution system, advanced manufacturing techniques, and new and innovative marketing techniques. Moreover, they allowed U.S. and Japanese firms to pool resources, thereby reducing the research and development costs of new and advanced semiconductor products, and helped U.S. firms gain access to the Japanese semi- conductor market. U.S.-Japanese joint ventures and strategic alliances therefore are the principal reason not only for America's regaining its role as the largest producer of semiconductors in the world, but also for America's increased market share in Japan. Misguided Focus on Market Share The semiconductor agreement is both a failure and an unwise policy. It is based on a fundamental mis- conception about trade: that because Country A's market share in Country B is significantly lower than Country A's global market share, Country B must have a closed market. This assumption is fallacious. Japan has 47 percent of the world market in semiconductors, excluding North America, but only 23 percent market share in semiconductors in North America. Japan also has 33 percent of the world market share in microprocessors, excluding North America, but only 6 percent share of the microprocessor market in North America. Few would argue that the North American (including the U.S.) semiconductor or microprocessor market is closed to Japan. American computer companies dominate the U.S. market because they are more competitive than the Japanese, not because the U.S. government pro- tects them from competition. There is a danger that extending the sen-dconductor agreement could cause the U.S. to be held to the same standard of market openness as it expects of Japan. If other countries were to decide that America is closed, they could retaliate by imposing tariffs and duties on America's exports, further harming U.S. busi- nesses and threatening American jobs. Under these circumstances, the U.S. would have no ground on which to stand against foreign retaliation.

Objectives Fulfilled Even if the semiconductor agreement's goal was to give foreigners a 20 percent share of the Japanese semiconductor market, as its supporters contend, this goal has been exceeded, thanks largely to U.S.-Japa- nese joint ventures and strategic alliances. Today the foreign market share for semiconductors in Japan is over 29 percent. Moreover, while total foreign sales of semiconductors in Japan in 1986 amounted to only $900 million, foreign sales today are worth over $6 billion-a sevenfold increase. The Clinton Administration believes that the recent success of America's semiconductor industry is the result of the U.S.-Japan Semiconductor Agreement. It is not. The U.S. semiconductor industry is successful today because it sought joint ventures with Japanese firms which gave American companies access to the Japanese market, advanced manufacturing techniques, and reduced research and development costs. The semiconductor agreement sets a bad precedent by focusing on market share numbers. If other countries applied the same policy to the U.S., many of America's exports could faced increased tariffs abroad. The President should let the agreement expire. Bryan T. Johnson Policy Analyst