Essays in Urban and Spatial Economics

My thesis consists of three chapters on how infrastructure and regulation shape urban and spatial outcomes. While rather different in terms of the specific topic, they share at least three common themes. First, all chapters are policy-oriented and speak to currently urgent issues. Second, they study how access—whether to goods, services, jobs, or essential needs—is shaped, by either physical infrastructure or policy. Third, ultimately, they are about how we can enhance the sustainability of the world. In terms of methodology, all chapters employ quasi-natural experimental designs, while two additionally adopt structural models, including dynamic game of entry and exit from empirical industrial organization and quantitative spatial model from urban and spatial economics. Below I briefly introduce each chapter. In Chapter 1, I study the cost impacts of minimum parking requirements (MPRs) for retail and local services firms. As a common North American land use regulation, MPRs require firms to provide parking proportional to size upon opening and operating. I derive exogenous sources of variation from a 2012 reform in Seattle that reduced MPRs in parts of the city delineated by arbitrary boundaries. I find extensive margin effects of MPR reductions, which positively impacted firm entry, survival, and related local outcomes, implying MPRs increase firm entry costs and fixed costs. Estimates from a dynamic discrete choice structural game theoretic model of entry and exit indicate that MPRs raise entry costs by 24 percent. A census tract would have at least one percent more firms if its MPRs were removed. Chapter 2 is about a quasi-natural experiment in August 2015 when Houston rolled out a complete redesign of its bus route network. This came in response to falling ridership, a new light rail system, and new employment centers. This paper evaluates the consequences for commuting patterns and welfare. Using old bus routes as the instrumental variable, at the census tract pair level, I find a one-minute travel time reduction due to the redesign led to a 0.94% increase in the number of commuters from 2014 to 2018. I also find large extensive margin effects. Evaluated through a quantitative spatial equilibrium model with mode choice, the redesign was welfare-improving. My counterfactual exercises bring new insights on the importance of connecting buses to intermodal transfer points, population, and employment centers. My contributions include that I provide a first study of the bus to the transit policy evaluation literature, and that I study a cheap re-optimization, instead of a costly expansion, of transit. Chapter 3, coauthored with Stephan Heblich, examines the impact of global warming on the presence of ice roads in far north Canada, which are crucial for transporting essential goods to remote northern communities, particularly First Nations. In recent decades, the operational duration of ice roads has diminished, resulting in a higher reliance on costly supply flights. To assess the consequences on food security, we exploit data from Nutrition North Canada, which offers subsidiesfor shipping to eligible communities. We find winter temperatures are negatively correlated with the availability of ice roads. When they are less available, more aircrafts fly in as a substitution. Using temperature as instrument, we show shorter ice road durations cause more shipping post winter, when only aircraft is available. So, food prices may rise and consumption may decline. To summarize, urban and spatial policies deeply affect access, growth, and equity. Applying empirical tools, I evaluate unique, live policy issues while preserving external validity, and enhance our understanding of mitigation of climate change. I believe the three essays are meaningful contributions to urban and spatial economics.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.