Direct estimation of synergy: A new approach to the diversity-performance debate

This study examines the linkages between relatedness and synergy in the context of diversification among U.S. pharmaceutical firms for the period 1960-1980. Rather than assume (as in the entropy, Herfindahl and concentric indices of diversification) that the levels of synergy generated by different related combinations of business units are identical, the authors estimate synergy directly using a modified version of the concentric index. In addition to estimating synergy using capital market performance of the firm as a whole, they examine the effects of nondrug diversification on the innovative productivity of firms’ pharmaceutical divisions along. The present two main findings are that production relatedness, such as that between drugs and chemicals, in fact did not imply synergy over the period of the study; and that the patterns of synergy for different types of relatedness shifted over time with the industry life cycle.