The three chapters of this dissertation examine evidence on relationships between
patterns of agglomeration within the United States and workers’ industry and occupation
decisions. The first chapter investigates agglomeration economies in the form of job
search and matching. Using data from the U.S. Census and the Current Population
Survey, I show that, on average, workers change occupation and industry less in more
densely populated areas. These results provide evidence in favor of increasing-returns-toscale
matching in labor markets. The second chapter examines the distribution of new
work across U.S. cities. Using Census microdata, I find that new work—that is, new
types of activities that closely follow innovation—concentrates in cities that initially have
more college graduates and a more diverse industrial base. This evidence is consistent
with concentrations of human capital facilitating regional adaptation to new technologies.
Finally, in the third chapter, I explore why industrial diversity is important for regional
adaptation to innovation. Using employment estimates for detailed new occupations from
the Occupation Employment Statistics, I find that diversified cities use new activities
more intensively, even as these activities mature. These results suggest that a “nursery
cities” model of industrial structure and regional innovation may not fully describe the
agglomeration pattern of new activities.