The lab and the plant: Offshore R&D and co-location with production activities

The literature has highlighted that the propensity of MNEs to co-locate offshore R&D labs with their production plants can vary substantially according to firm and industry characteristics. In this paper, we apply a novel two-stage estimation procedure that allows us to tease out this heterogenous behavior and investigate the factors that are associated with a higher propensity to co-locate production and R&D activities abroad. Using data on 1483 greenfield international investments in R&D activities made by 855 firms in 587 cities worldwide, we uncover that the strength of the co-location effect is indeed highly heterogenous across firms. In particular, it is higher among firms with less international experience and geographical dispersion of international activities, as well as with a lower share of intangible assets. These results are consistent with the idea that co-location is a substitute for firms' ability to coordinate complex and dispersed organizational structures, and that firms relying relatively less on codified knowledge can use co-location of offshore R&D and production to facilitate knowledge transfer across activities.