Essays on infrastructure and urban development in developing countries
OA Version
Citation
Abstract
This dissertation presents three chapters that focus on questions studying the infrastructure and urban development in developing countries. More than 1 billion people worldwide inhabit slums or informal settlements, characterized by inadequate access to public infrastructure. Although slums offer affordable housing options near employment opportunities, their residents often face the risk of displacement due to urban development initiatives and land use policy changes. The first chapter examines the impacts of a zoning reform in São Paulo, Brazil, on formal and informal housing. To evaluate these impacts, we compile an extensive dataset covering the years 1996-2016, tracking slum evolution, zoning changes, public housing constructions, detailed formal housing records from property tax, and socio-economic variables at the neighborhood level. Our identification strategy combines propensity score matching and difference-in-differences design. Wed find that the reform led to an increase in formal housing supply, a decrease in slum prevalence, and an uptick in public housing development. Neighborhoods affected by the reform show some evidence of higher income and education attainment, indicating gentrification. While growing literature has documented distinct features of aid projects from China and traditional donors, disparities in aid effectiveness remain poorly understood. Chapter 2 fills this gap by comparing the Chinese and World Bank project impacts on African local economies. Using detailed, geocoded project data and a stacked difference-in-differences approach, we find that Chinese infrastructure projects significantly increase nighttime light in recipient regions, with effects persisting over time, while World Bank projects show no significant impact. Location and project characteristics only partially explain the differences. We further rule out three potential mechanisms: follow-up projects, political favoritism, and implementation by Chinese companies. Access to basic infrastructure is a critical component of quality of life and an important measure of economic development. However, on-the-ground data about infrastructure access, especially in low-income countries, is often sparse and costly to collect. We train and calibrate a machine learning model to extract data on infrastructure access for each 6.72x6.72km area of Africa from satellite images. The model achieves accuracy levels of 77.1% to 84.7%. We show the value of this novel dataset with two applications. First, we use a spatial regression discontinuity design to study how much of the heterogeneity in infrastructure access across countries comes from differences in institutional quality. Second, we study the role of political favoritism in explaining within country heterogeneity.
Description
2024