This paper investigates the stylized empirical facts regarding U.S. urban growth
in the 1880s. The author estimates the covariation of empirical proxies for various
theorized sources of growth with the growth rates in output, capital, and labor.
Results support Barro [Quarterly Journal of Economics, 106, 407-444 (1991)] and
others who have found an important role for convergence and other neoclassical mechanisms.
Importantly, it is found that externality-based factors impacted growth in inputs
but had no direct relationship with productivity growth.