APL (Ajiken Power Lunch)

Thursday, 12th of September, 12:00 to 1:15 p.m.

IDE-JETRO C23 Meeting Room 　

Over the past few decades, industrial policies — policies that are purposefully targeted at particular industries — have been an active area of policy debate. This paper studies the econometric evaluation of such policies when industries are linked through production networks and firms in each industry engage in strategic interactions. I develop a general equilibrium model with these two features to define a causal policy effect as a ceteris paribus difference in outcome variables across different policy regimes. The key mechanism of my model is that when firm-level production functions exhibit constant returns to scale, policy effects are mediated by changes in firms’ marginal profits not only through adjustments of their own actions but also via those of competitors’ actions (i.e., strategic complementarities), and that both of these changes are compounded by the production network. To identify such policy effects, I develop a new procedure that first characterizes them in terms of sector-level variables and firm-level variables — firm-level sufficient statistics, and then recovers these building blocks with the aid of the control function approach of the industrial organization literature. Using my framework, I examine the causal impact of one part of the U.S. CHIPS and Science Act of 2022 on GDP. My estimation predicts that accounting for firms’ strategic interactions even flips the sign of the policy effect with the magnitude roughly the same, highlighting the policy relevance of strategic interactions in the presence of a production network.

Dr Ko Sugiura (Ghent University)

Yuta Watabe