Working Papers

Breaking Invisible Barriers: Does Fast Internet Improve Access to Input Markets?

(with Banu Demir and Beata Javorcik)

How do improvements in internet infrastructure affect supply chain organization and distribution of economic activity across space? We exploit an episode of massive investment in optical fibre networks across Turkish provinces to shed light on this question. Using rich micro-data on firm-to-firm transactions and information on expansion of optical fibre networks, we show that fast internet affects firms' supplier networks by facilitating doing business. Firms not only reallocate their purchases towards suppliers with better internet connectivity but also diversify their input sourcing patterns. We develop and estimate a tractable spatial equilibrium model with endogenous production networks under rational inattention. We find that the elasticity of manufacturing firm-to-firm trade with respect to internet connectivity is sizeable. Through the lens of the estimated model, we find that improvement in high-speed internet infrastructure leads to a 2% gain in the real income in the median Turkish province.

Endogenous Spatial Production Networks: Quantitative Implications for Trade and Productivity

Best Conference Paper Award, 13th FIW Research Conference

Best Paper Prize, 19th GEP/CEPR Annual Postgraduate Conference

Distinguished CESifo Affiliate Award, 2021 CESifo Area Conference on Global Economy

I develop a model of endogenous production network formation between spatially distant firms. Unlike other such models, it is tractable even for very large numbers of firms, that is, it delivers closed-form predictions for firm-to-firm trade, it can be estimated via maximum likelihood, and it can be used for firm-level counterfactual analysis. I exploit novel micro-data on Indian firm-to-firm production networks for estimation. The estimated model implies that upon market integration across Indian states, over half of the variation in changes in firms’ sales to other firms can be explained by endogenous changes in network structure.

Triad Trade & Small Worlds of Large Spatial Production Networks

Do firms’ production network contacts (suppliers, customers) help form connections with newer suppliers or customers? If they did, one would observe transitive triads — a group of three firms all trading with each other — in production networks. Using rich administrative data on firm-to-firm linkages from India, I provide first evidence that triads are excessively prevalent in firm-to-firm production networks. Further, I find that proximity through the production network is an important determinant of trade frictions. I develop a quantitative general equilibrium model of network formation between spatially distant firms. The model lends to elegant aggregation and features endogenous trade frictions unlike standard trade models. The estimated model implies that network proximity explains a dominant majority of trade frictions.