Patents as Indicators of R&D Internationalization: A Note on Method

A significant strand of research on the internationalization of R&D in multinational enterprises (MNEs) uses patent databanks as sources of information, especially those provided by the U.S. Patent and Trademark Office (USPTO). The present paper examines the reliability and validity of the common practice of (1) identifying the geographical location of inventions by the address of the first-listed inventor in patent applications, and (2) including in international comparisons both U.S. and non-U.S. MNEs. The first inventor criterion is problematic principally because most patents list more than one inventor, and these sometimes reside in different countries. Comparing the patenting activities of U.S. and foreign firms is also problematic, because of differences between the propensities to patent domestically and abroad.

The analysis draws on two sets of empirical data, (1) USPTO pharmaceutical patents (patent classes 424 and 514) for the years 2000-2015, and (2) results from a small questionnaire survey directed to U.S. and UK patent attorneys. The survey provided no support for the assumption that the address of the first inventor is a reliable indicator of where the R&D underlying the invention was performed. The analyses of patent data confirm that the chosen criterion matters considerably and demonstrate dramatic differences in patenting in the U.S. between domestic and foreign firms. The results raise questions regarding the reliability and validity of the methods employed in much of the inherited literature and provide conclusions as to how best to employ patent data in research on international R&D.