Working Paper: CEPR ID: DP13700
Authors: Joaquin Blaum; Claire Lelarge; Michael Peters
Abstract: Commonly used firm-based models of importing imply that firm productivity should have no effect on the allocation of expenditure across a common set of sourcing countries. Using French data, we show that this homotheticity property is soundly rejected: larger firms concentrate their import spending on their top varieties, holding the sourcing strategy fixed. To rationalize this finding, we propose a novel model of importing that features (i) a complementarity between firm productivity and input quality and (ii) heterogeneity across countries in their ability to produce high quality inputs. This model implies that large firms bias their spending towards countries with a comparative advantage in producing high quality inputs and hence generates a non-homothetic import demand system. We provide empirical support for this and other predictions of this theory.
Keywords: Trade in Intermediate Inputs; Firm Heterogeneity; Firm Size; Nonhomothetic Import Demand; Quality-Productivity Complementarity
JEL Codes: F11; F12; F14; D21; D22; D24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Larger firms (L25) | Concentrate their import spending on top sourcing countries (F10) |
| Larger firms (L25) | Nonhomothetic import demand system (D11) |
| Firm size (L25) | Import demand patterns (R22) |
| Larger firms (L25) | Pay higher prices for inputs (L11) |
| Larger firms (L25) | Bias spending toward countries producing high-quality inputs (F14) |
| Input quality choice model (C25) | Nonhomothetic import demand system (D11) |