A Breath of Fresh Air? Firm types, scale, scope and selection effects in drug development

This paper measures differences in the innovation performance of different types of firms in the pharmaceutical industry. We compare the innovation performance of incumbent firms with entrants, controlling for differences in the scale and scope of research, both at the firm level and at the project level. To do so, we develop a simple analytical framework of drug development, which we use to estimate a structural model, using data on 3,000 drug R&D projects in preclinical and clinical trials in the US during the 1980s-early 1990s. Key to our approach is a careful attention to the issue of selection – firms choose which compounds to advance into clinical trials. This choice depends upon the likelihood of success, but also upon economies of scale and scope, and strategic considerations about product cannibalization. It also depends upon how the costs of development and the rewards of success are shared within organizations and between alliance partners. After controlling for selection, we find that: a) incumbent pharmaceutical firms draw their compounds from better statistical distributions; b) over time, learning or environmental selection make entrants firms more similar to the established firms both in terms of selection behavior and research productivity; c) compounds licensed by pharmaceutical firms are at least as likely to succeed as internal developed projects, inconsistent with the “lemons” hypothesis; d) firm scale improves innovation performance but not scale at the project level.