Working Paper: NBER ID: w15997
Authors: Howard Bodenhorn; David Cuberes
Abstract: In this paper we argue that in 19th century U.S, households and firms that were located in cities with banks enjoyed a higher level of both consumption and production amenities than those who were located in cities without banks. We use data on banks location and city population growth in the Northeastern United States in 1830-1870 and document a positive and strong correlation between financial development and subsequent population growth. The correlation is robust to controls for geographical characteristics of the city, the percentage of population working in different sectors, and its initial population. Propensity score matching estimators that compare similar cities in terms of observables also yield a positive association between finance and urban growth. Our estimates suggest that the presence of a bank at a given location in the late 1830s is associated with increases its population growth in the 1840s by one to one and a half percentage points per year. Because urban growth was correlated with economic development in the nineteenth-century U.S, we believe our results provide further support for the finance-growth nexus.
Keywords: financial development; urban growth; nineteenth century; banks; population growth
JEL Codes: N11; N90; R11
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Presence of banks in cities during the late 1830s (N93) | Subsequent population growth rates in the 1840s (J11) |
| Towns with banks (G21) | Compound annual rates of population growth greater than towns without banks (G21) |
| One standard deviation increase in bank assets (G21) | Population growth over the following three decades (J11) |