Working Paper: NBER ID: w22946
Authors: Arnaud Costinot; Dave Donaldson
Abstract: In this paper we develop a new approach to measuring the gains from economic integration based on a generalization of the Ricardian model in which heterogeneous factors of production are allocated to multiple sectors in multiple local markets based on comparative advantage. We implement this approach using data on crop markets in approximately 2,600 U.S. counties from 1880 to 1997. Central to our empirical analysis is the use of a novel agronomic data source on predicted output by crop for small spatial units. Crucially, this dataset contains information about the productivity of all units for all crops, not just those that are actually being grown—an essential input for measuring the gains from trade. Using this new approach we find substantial long-run gains from economic integration among US agricultural markets, benefits that are similar in magnitude to those due to productivity improvements over that same period.
Keywords: Economic Integration; Agriculture; Trade Costs; Productivity
JEL Codes: F0; F1; F11; F15; F17
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Economic integration (F15) | Gains in U.S. agriculture (N52) |
| Economic integration (1880-1920) (N93) | Annual growth in real output (O40) |
| Economic integration (1954-1997) (F15) | Annual growth in real output (O40) |
| Productivity improvements (O49) | Gains in U.S. agriculture (N52) |
| Observed outputs and local prices (P22) | Unobservable productivity shocks (O49) |
| Trade costs (F19) | Unobservable productivity shocks (O49) |