Are Short-Lived Localized Productivity Shocks Persistent? Historical Evidence
Does history matter for urbanization? This paper examines whether short-run shocks to urban population arising from idiosyncratic productivity shocks can have a persistent impact on long-run urban growth. The analysis addresses a fundamental question of urban economics: whether agglomeration economies, and thus growth, derive from an independent role for city size as well as specialization. The study takes advantage of the history of urbanization in the U.S. American state of Oklahoma during the period 1907 to 2010. A unique dataset documents the series of highly localized, short-term exogenous positive shocks to urban populations from oil discoveries. These shocks created boomtowns almost overnight during the “Black Gold” oil boom of 1900 through 1927. The dataset also geocodes the results of a severe episode of land erosion during the late 1920s and early 1930s, known as the “Dust Bowl.” It resulted in depopulation and out-migration. These episodes provide clean natural experiments to identify plausible causal impacts. Our new dataset also allows us to control for a large set of geographical variables and initial conditions in the context of very early urbanization. Our study of urban growth over the period 1940 through 2010 finds that positive shocks to urban population induced by the oil boom had a persistent impact up to 70 years after they occurred. Negative shocks from the erosion crisis persisted for about 40 years. These results suggest that historical shocks can drive path-dependent urban growth. We also investigate the mechanisms through which the oil shocks generated such persistence.
David Cuberes is Professor of Economics at Maynooth University. His research fields are urban economics, economic history and economic growth.
and 
by