This paper explores the strategic value of location decisions by Multinational Enterprises (MNEs). Using a game-theoretic model, it finds that when learning is contained to a single market (local learning), strong firms will chase weak competitors into new markets in order to disrupt their rivals’ learning processes and maintain a competitive advantage. The effectiveness of this strategy weakens, however, when learning is transferable to other markets (global learning) because firms cannot so easily affect what, how and when their competitors learn. These outcomes suggest that, in some cases, strategic interaction alone could be sufficient explanation for geographic clustering.

JEL Classification Numbers: F23, L13, D43, M10
Keywords: FDI, location choice, firm heterogeneity

I acknowledge the helpful comments of Daniel Byrd, Gordon Hanson, Kai-Uwe Kühn, Jeff Furman, Francine Lafontaine, Erin Martin, Joanne Oxley and Bernard Yeung on prior drafts as well as seminar participants at Harvard, INSEAD, Michigan, NYU, USC and INFORMS. Errors remain my own.