The Impact of the Eisenhower System of Interstate Highways on Agglomeration.

摘要

The Eisenhower System of Interstate Highways resulted in the construction of more than 40,000 miles of new or enhanced highways. The object of this study is to determine how this improvement in infrastructure impacted agglomeration or the concentration of population and economic activity. In the first chapter, I explore demographic trends in the United States from 1960 to 2000 including urbanization, suburbanization and income growth. Some clear indications that emerge are depopulation of central cities, an increasing outward movement towards suburbs and exurbs, disproportionate growth in automobile transportation and an increasing predominance of the service sector. The remainder of the chapter provides a synopsis of recent literature in the fields of agglomeration, urbanization and economic growth.;In the second chapter, I develop a theoretical model, based on the core-periphery framework, and examine the response of utility optimizing agents to reductions in transportation costs. Using several utility functions to describe relative preferences for modern and traditional goods and introducing finite commuting costs in the modern sector, I find that improvements in speed of commuting lead to increasing levels of agglomeration and income. This result is fairly robust and is not impacted by differences in production technologies, urban population densities or a wide range of utility specifications.;In the third chapter, I measure growth rates of population, employment and net migration from 1950 to 2000. I find strong evidence for agglomeration both within interstate counties and from counties without interstate highways to counties with interstate highways. By increasing the speed of travel, the interstates have a unique and unconventional impact on agglomeration. They do not increase metropolitan population densities but allow the boundaries of the metropolitan regions to expand and accommodate greater populations. I test the robustness of these results by accounting for a host of alternate specifications and using instrumental variable techniques.