Over the past 25 years social scientists attempting to explain the dramatic
changes in the relative distribution of urban and rural population growth
have gravitated toward two competing explanations. The regional restructuri
ng hypothesis holds that changes in the spatial distribution of employment
opportunities have been dominant whereas the deconcentration hypothesis att
ributes these changes to changes in residential preferences of workers and
consumers. We develop an empirical test of these two explanations based on
whether commuting and migration are positively or negatively related after
controlling for other economic factors. Our econometric results support the
deconcentration hypothesis.