UC Berkeley

This paper examines the changing process of strategic alliance formation in the Japanese electronics industry between 1985 and 1998. With data on 123–135 Japanese electronics/electrical machinery makers, we use a dyad panel regression methodology to address hypotheses drawn largely from embeddedness theory on how the firms’ horizontal and vertical keiretsu business group affiliations and prior alliance networks supported and constrained partner choice in new R&D (innovation) and nonR&D (implementation) domestic economy alliances. We find that in the first half of our series (1985–1991; the “preburst” period), keiretsu served as infrastructure for new strategic alliances that had both innovation (R&D) and implementation (nonR&D) goals. In the second half of our series (1992–1998, the “postbubble” period), the keiretsu effects on R&D alliance formation were gone, but the groups’ role in nonR&D alliances had magnified. Moreover, as the keiretsu effect on new alliances fell or rose, those of prior direct and indirect alliance ties moved in the opposite direction. We conclude with some discussion of whether these period shifts in the embeddedness of the electronics industry alliance formation process were or were not effective adaptations to the turbulence and uncertainty of the postbubble Japanese economy.