The Birth of a Multinational: Innovation and Foreign Acquisitions

We model the joint decision to innovate and entry into foreign market in a continuum-time stochastic environment. We test the assumptions and predictions of the model using panel firm-level data from Spanish firms with detailed information on innovation and international activities. We find that innovation is a lumpy and disruptive process: it occurs sporadically and it is associated with a drop in technology immediately after. We incorporate this technological feature into the model and arrive to novel empirical predictions: (1) Firms that eventually become multinationals innovate more often than those that remain domestic. (2) The headquarter unit is more likely to innovate immediately after of becoming multinational, relative to firms that remain domestic and relative to established multinational firms. (3) Although in the cross-section multinational firms are more productive than domestic firms, productivity in the headquarter drops immediately after becoming multinationals. The three predictions are confirmed empirically.