Working Paper: NBER ID: w28059
Authors: Treb Allen; Dave Donaldson
Abstract: How much of the spatial distribution of economic activity today is determined by history rather than by geographic fundamentals? And if history matters for spatial allocations, does it also matter for overall efficiency? This paper develops a forward-looking dynamic framework for the theoretical and empirical study of such questions. We derive conditions on the strength of agglomeration externalities under which equilibria are unique and yet temporary historical shocks can have particularly persistent, or even permanent (i.e. path-dependent) consequences. When estimated using U.S. data from 1800-2000, this model displays multi-century persistence from small and temporary shocks as well as path dependence (with large aggregate welfare effects) throughout much of our estimated parameter range.
Keywords: spatial economy; path dependence; agglomeration externalities; historical shocks; economic geography
JEL Codes: C33; C62; F1; R11; R13; R23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| small historical shocks (N13) | long-term economic outcomes (F69) |
| temporary shocks (E32) | permanently different economic outcomes (P19) |
| historical shocks (N13) | trajectory of economic activity (E32) |
| agglomeration externalities (R11) | unique equilibria influenced by historical events (D59) |
| historical spillovers (N13) | current economic productivity (O49) |