The US-Japan semiconductor trade arrangement: Political economy, game theory, and welfare analyses

While advocates of managed trade have often singled out the semiconductor industry as the best case for government intervention in trade, this analysis of the 1986 US-Japan Semiconductor Trade Arrangement reveals the failure of good intentions. Discussion focuses on the effects of the Arrangement on the Dynamic Random Access Memory (DRAM) market.First, a political economy approach is employed to show that there were two key conditions to the US semiconductor industry's success in gaining trade protection. The first condition was an alliance between US semiconductor producers and users protection was granted only when government did not face antagonistic consumer pressure. The second condition was pressure from Congress for protection due to a variety of reasons this pressure motivated the executive branch to grant trade protection to the semiconductor industry in order to prevent broader congressional action.Second, a three-country oligopoly model is used to analyze cyclical dumping in the DRAM industry and the effects of various methods of protection proposed during the negotiation of the 1986 Arrangement. Threatened US antidumping duties were rejected by negotiating parties in favor of price floors because this shifted rents from the US Government and third countries to Japan. However, the Japanese Government's rational response to protection was to make use of Japan's position as the dominant DRAM supplier it mandated cuts in exports, thereby raising prices. Evidence of such behavior is presented. Higher prices destroyed the US producer-user alliance which consequently sabotaged the 1986 Arrangement.Finally, welfare analysis is employed to estimate the short-run welfare gains and losses resulting from the effect of the 1986 Arrangement on DRAM prices. In the short run, producers gained and consumers lost in all countries. However, Japan experienced a net gain while the US, Europe and ROW each experienced a net loss. This analysis is the first comprehensive estimation of the welfare effects of the 1986 Agreement and represents one of the first efforts to calculate the welfare effects of bilateral agreements in third countries. Data from wide-ranging sources include previously unpublished series.