Cityscapes of Inequality: Segregation and the Politics of Public Goods in Urban Brazil
Many thanks to my field assistants and survey enumerators (names here) for all their efforts with data collection for this project.
Why do cities of comparable size and fiscal capacity exhibit such stark disparities in the provision of basic urban services? Why are even country capital cities, such as Bras\'{i}lia and Cape Town, laggard in provision when they should, in theory, have better access to resources from the federal government? My book manuscript sheds light on these questions using a mixed-methods approach and original data collected over 19 months of in-person field research. I demonstrate that patterns of urban class- and racial segregation, the geographic layouts of cities, is at the heart of the answer.
I argue that politically polarized cities under-provide public goods. In particular, the geographic configuration of cities (i.e., their degree of class-based segregation) determines when cities are politically polarized and when they, instead, develop cross-class political coalitions to address the spatial externalities between groups. The geographic configuration of cities shapes how the urban middle class forms policy preferences and mobilizes to realize them. Segregation along class lines in cities reduces the spatial externalities of inequality (e.g., sewage run-off, waterborne diseases, organized crime) that spill over from impoverished areas (e.g., ``slums'' or favelas) to the middle class. In these segregated settings, reduced cross-class externalities increases the efficacy of privately-provided services (e.g., private guards, personal firearms, household water wells) to the extent that these options substitute demand for public provision. Conversely, in integrated (de-segregated) cities, the scale of spatial externalities undermines the efficacy of small-scale private services, thereby inducing middle-class preferences for public goods that address externalities.
Thus, while segregation encourages exit into private service provision, class-based integration –through the externalities mechanism– aligns the urban middle class with the poor to form preferences and coalitions for public goods in place of private alternatives. The geography of cities produces enduring patterns of spatial externalities that, in turn, shape the policy preferences and political behavior of the urban middle class. Drawing on semi-structured interviews with city councilors, retired mayors, and civil servants, focus groups with neighborhood associations from different city cases, a novel quasi-experimental strategy for causally estimating the effects of segregation, ethnographic data collection, a set of geocoded conjoint survey experiments administered across neighborhoods, and a large-scale door-to-door survey of over 4,000 households across 420 neighborhoods in the megacity of São Paulo, Brazil, I document how the argument applies to the location of streetlights, the extension of sewer lines, the stationing of policing units, and investments in public schooling. The nine-chapter draft of the manuscript is available upon request.
Chapter 1: Introduction
Chapter 2: Theory: How segregation shapes public and private goods preferences
Chapter 3: Political causes of segregation
Scope conditions: traces historical periods when geography is politically engineered
Introduces identification strategy to separate segregation as cause
Chapter 4: Data Collection, Survey Designs, and Main Empirical Results I
Chapter 5: How voters realize their preferences
What do voters vote for in city-level elections?
Main Empirics II: Effects of segregation on voting, political polarization, and urban coalitions
Chapter 6: Model of Politics: How politicians aggregate preferences
Chapter 7: Illustrates the theory using 5 city case studies
Chapter 8: Racial identities and geographies encode class externalities
Conjoint survey experiments across varying geographies
Chapter 9: Public Goods in Place of Redistribution as a Solution to Class Externalities