November 23, 2023, 11:00–12:30
Room Auditorium 4
Behavior, Institutions, and Development Seminar
Abstract
We study the local and aggregate welfare effects of foreign multinational enterprises (MNEs) in an economy with wedge-like distortions. We start with a guiding general equilibrium (GE) framework that highlights the ex-ante ambiguous role distortions can play in amplifying or attenuating the welfare effects of foreign MNEs, and the primitives needed to take the theory to the data. For our empirical application in Mexico, we track the size of four sectors of the economy -- domestically-owned establishments (split into formal or informal) and foreign MNEs (split into maquila and non-maquila) -- across time (1994 to 2019) and commuting zones (CZs). We also construct direct measures of initial wedges and distortions such as crime, labor taxes, subsidies, and credit constraints (varying across establishment types and CZs). We show that Mexico faces substantial distortions that can generate resource misallocation within and across CZs. Then, using an instrumental variable strategy, we find that increases in foreign MNE employment lead to expansions of the local domestic economy. This effect is mainly driven by formal domestic sector growth. Moreover, foreign MNEs in the maquila program have weaker positive effects on the domestic economy than non-maquila MNEs. Finally, we use the reduced-form estimates and the microdata to calibrate the model and quantify the GE effects. We simulate a 10% MNE productivity shock and find that the expansion of MNEs in Mexico between 1994 and 2019 led to an aggregate 6% welfare gain. A quarter of this gain comes from reallocation effects in the presence of distortions, while a fifth comes from productivity spillovers from foreign MNEs to the local domestic sector. (With Jose P. Vasquez (LSE) and Roman David Zarate (World Bank))