Production Networks and Skill-biased Technology

DEPARTEMENT ECONOMIE ET SCIENCES DE LA DECISION

Intervenant : Banu Demir PAKEL (Bilkent University)

HEC Campus - Bâtiment T - salle T004

Abstract:

Technological progress is at the root of economic development, and in the last few decades, technology has mostly been skill biased. While there is a vast literature on the development of skill-biased technologies and on the incentives for individual firms to adopt them, less is known about the role of the production network in the process of adoption and diffusion. Yet, in most examples of technological advances, a firm’s decision to adopt is intertwined with the same decision by its suppliers and customers. We study theoretically and empirically this interconnection between firms’ decisions to adopt higher, skill-intensive technologies and their propagation through the network. First, we combine domestic firm-to-firm trade transactions data from Turkey with matched employer-employee data to document a novel fact: skill intensive, high-wage firms disproportionately buy and sell inputs to other skill-intensive, high-wage firms. Next, we develop a quantitative general-equilibrium model of endogenous network formation and technology adoption to account for these empirical regularities. In the model, firms that adopt the higher technology use relatively more skilled labor and skill-intensive inputs.  Firms’ decisions to adopt higher technologies are intertwined because they change the relative cost and demand for skill-intensive inputs in the network. Finally, we structurally estimate the model to data and use a counterfactual to study how trade shocks influence, directly and indirectly through the network, the adoption of firms to international markets.