Why do firms decide to offshore certain parts of their production process? What qualities certain countries as particularly attractive locations to offshore? In this paper we address these questions with a theory of international production hierarchies in which organizations arise endogenously to make efficient use of agents' knowledge. Our theory highlights the role of host-country management skills (middle management) in bringing about the emergence of international offshoring. By shielding top management in the source country from routine problems faced by host country workers, the presence of middle managers improves the efficiency of the transmission of knowledge across countries. The model further delivers the prediction that the positive effect of middle skills on offshoring is weaker, the more advanced are communication technologies in the host country. We provide evidence consistent with this prediction.

We generalize the Antràs and Helpman (2004) model of the international organization of production in order to accommodate varying degrees of contractual frictions. In particular, we allow the degree of contractibility to vary across inputs and countries. A continuum of firms with heterogeneous productivities decide whether to integrate or outsource the production of intermediate inputs, and from which country to source them. Final-good producers and their suppliers make relationship-specific investments which are only partially contractible, both in an integrated firm and in an arm's-length relationship. We describe equilibria in which firms with different productivity levels choose different ownership structures and supplier locations, and then study the effects of changes in the quality of contractual institutions on the relative prevalence of these organizational forms. Better contracting institutions in the South raise the prevalence of offshoring, but may reduce the relative prevalence of FDI or foreign outsourcing. The impact on the composition of offshoring depends on whether the institutional improvement affects disproportionately the contractibility of a particular input. A key message of the paper is that improvements in the contractibility of inputs controlled by final-good producers have different effects than improvements in the contractibility of inputs controlled by suppliers.