Essays on Inequality in Cities

Abstract

This dissertation investigates the equity consequences of one the largest and most ambitious public works programs in the world—the Interstate highway system—on inequality across groups and the intergenerational outcomes of children.

In my primary dissertation project described in chapters one and two, I quantify how interstate road infrastructure construction during the 1960s increased racial inequality in American cities. As a mechanism for why these disparities arise, I explore the central role of exclusionary institutions that limited access to residential neighborhoods for minority families and how they shape the distributional effects of interstate highway policy.

I develop a quantitative model with rich spatial heterogeneity to capture how in equilibrium, institutions interact with social preferences and the housing market to influence inequality in highway impacts. The model structure crucially enables distinguishing the extent to which segregation is de jure, versus due to economic differences or social homophily, to measure which barriers are most meaningful for the residential choices of Black families. I estimate the model using spatially-granular historical records constructed from restricted Decennial Censuses for 25 cities and archival road network maps that I digitized for 71 cities. The empirical variation for estimation derives from quasi-random placement of interstate routes relative to historical comparison roads and from spatial discontinuities in where racial exclusion prevailed across locations.

With the empirical estimates and the model framework, I find that highways produced commuting benefits that accrued largely to suburban neighborhoods and environmental harms concentrated in central areas, where the Black population resided, leading to their losses from interstate highways. Exclusionary institutions are the primary force behind segregation in the 1960s and account for the majority of the racial gap in interstate impacts. When I eliminate constraints on where Black households are permitted to live, rather than being harmed by highways, the Black population achieves gains. This shift in welfare occurs because they are no longer confined to the urban core, where highway costs outweigh benefits, and the findings highlight the key role of institutions in shaping the disparate incidence of place-specific shocks.

In my secondary dissertation project described in chapter three, I measure the consequences of place-based interventions on children’s long-run outcomes, continuing with the setting of the interstate highway system.

The first step to making this research possible is building an infrastructure of intergenerational linkages encompassing 100 years of economic events. At the U.S. Census Bureau, I have played a leading part in creating novel parent-child linkages that cover the universe of the U.S. population born between 1964 and 1979—on the order of tens of millions of individuals. Construction uses machine learning models, string-comparison methods from natural language processing, and immense administrative tax datasets.

Using these linkages, I document that upward mobility for Black children was depressed during this period, where children from families in the top-quintile of parental income were more likely to enter the bottom-quintile in adulthood rather than remain in the top-quintile. I then investigate how the interstate highway network affected the economic mobility of these households. Place-based policies, such as transportation infrastructure, aim to benefit areas they directly target, and highways do so by improving a neighborhood’s access to employment. However, when a policy is sufficiently large, migration in equilibrium alters the peer composition of both the locations households seek out and the locations they depart from. In the case of the interstate network, as this migration was highly differential by economic status (higher educated, White families left central areas for suburbs with increases in connectivity), inequality was amplified across space—suburban neighborhoods experienced job access increases and enhanced peer externalities while central neighborhoods faced solely declines in peer externalities. These direct and indirect impacts then subsequently influence intergenerational mobility by race.