Essays in Urban and Environmental Economics

Abstract

This dissertation comprises three chapters in urban and environmental economics. The first two chapters evaluate the impacts of urban greening in the context of rail trails. The third examines a rental housing regulation intended to help vulnerable populations adapt to climate change. In Chapter 1, I investigate how decades of rail trail openings have changed neighborhoods in the Boston metropolitan area. Using a staggered event study, I find that housing values begin to rise within a decade of trail openings, accompanied by modest increases in the share of college-educated residents and little change in other socioeconomic measures. Over time, these differences grow substantially: tracts with rail trails became increasingly more expensive, with more educated and wealthier residents in the decades following the trail opening. These patterns suggest that rail trails catalyze significant neighborhood change, and early compositional shifts amplify over time. Chapter 2 examines the long-run effects of rail trails in the United States, where over 26,000 miles of abandoned rail lines have been converted into trails, with another 9,000 miles underway. I conduct a long-differences analysis across U.S. metropolitan areas, where rail trails are observed but opening dates are unavailable. To address endogeneity in trail placement, I instrument for trail creation using inefficiently connected historical rail segments, identified by edge betweenness centrality. Estimates show that between 1970 and 2020, tracts with rail trails experienced substantially higher housing value growth and an expansion of housing supply. These changes are accompanied by demographic shifts, including increases in median household income, the college-educated population, and the White share of residents, consistent with residential sorting. The effects are heterogeneous across space, particularly by proximity to the urban core. Chapter 3 examines how a policy in Montgomery County, Maryland requiring landlords to provide air conditioning in all rental properties affected rents. Using a difference-in-differences approach, I find that rents increased by $56 to $97 for units that experienced tenant turnover and by $41 to $53 overall across the market. This suggests that landlords more than fully passed the costs of A/C installation onto tenants. These effects are more pronounced in rental properties located in median income census tracts. To explain this substantial effect, I present descriptive evidence that landlords may be undertaking additional renovations concurrently with the required A/C installations.