Abstract: This dissertation consists of two essays exploring the often noted dispersion of economic activity within cities. Focusing in particular on the phenomenon of polycentricity, these essays explore the relationship between employment centers and spatial and economic outcomes of cities. The first essay explores the implications of two common proposed criteria for identifying an employment center. Does the area represent a local concentration of employment? Does the area affect the local population density of the city? Using data on both place of employment and place of residence, I propose a new method for testing the relationship between concentrations of employment and population density within a metropolitan area. First a recently developed statistical method is used to identify concentrations of employment using data on place of employment. Second, I propose two methods for estimating the extent of the radius of influence for an employment center, using the relationship between tract of employment and tract of residence. Third, I propose a new specification for the entrance of distance into the polycentric regression. This new specification allows the impacts of the concentrations of employments on density to be positive, following the theoretical hypothesis. I use this new specification to jointly estimate the local gradients of 21 identified concentrations of employment in the Houston metropolitan area on their local population density. I find that not all identified employment concentrations have the expected significant positive gradients, and thus do not qualify as employment centers. I also find that the estimated gradients are sensitive to estimates for the radius of influence for each employment concentration, and that the level of employment in an employment concentration, alone, is not a strong predictor of significant local impact on population density or on the size of the estimated gradient. The second essay tests for the theoretically predicted relationships between the number of employment centers in a city, and the city's transport costs and wages. Urban area vehicle miles travelled rise with an increase in the number of employment centers in an urban area, while commute times are unaffected. These findings contradict the common hypothesis that additional employment centers lower transport costs by allowing workers to live closer to work. Instead, it appears that if transport costs are falling they do so through a fall in per unit distance price. I find that urban area average wages fall with an increase in the number of employment centers. I also find that average wages increase as a larger share of employment locates within employment centers. These two findings support the belief in the presence of agglomeration economies within employment centers that increases in concentration. In a competitive equilibrium the formation of additional employment centers have externalities in both the costs and benefits, thus it is not clear if the efficient number of employment centers will be formed within an urban area. This is explored through an investigation of the determinants of the share of urban area employment that locates in employment centers. I find that the predicted employment share maximizing number of employment centers increases with urban area size.

"This dissertation contributes to two literatures: Urban Economics and Labor Economics. In the first chapter I estimate the effect of home ownership on individual workers' unemployment and wage growth, as well as other labor market outcomes. Because of higher moving costs, home owners will be less willing than renters to relocate for work and could therefore face longer unemployment spells. To elaborate on this hypothesis, credited to Oswald (1996), I build a simple search model and obtain a set of labor market predictions to test. The current microeconomic literature has reached mixed results regarding home ownership's impact, with most studies concluding that home ownership reduces unemployment. I show that the instruments used are likely to be invalid because of, among other reasons, Tiebout (1956) type sorting into housing markets. I use an instrumental variable free of the endogeneity present in other work: the county level home ownership rate when and where the worker grew up. This IV affects workers' preferences for housing but not, conditional on my covariates, their labor market ability. My results indicate that home ownership is a significant hindrance to mobility, and homeowners suffer longer unemployment spells and slower wage growth because of it. In the second chapter I use a dynamic model of neighborhood choice to estimate household preferences over the demographic characteristics of a neighborhood. I focus on the racial mix, average income and housing price level of a neighborhood, and whether households prefer neighbors that are similar to themselves. Identification of these preferences is complicated by the social aspect of neighborhood amenities. A household's valuation of a particular choice (neighborhood) is a function of the choices other households in the market have made and will make in the future. I show that demographic characteristics of a neighborhood are therefore endogenous to neighborhood quality. Standard estimates of preferences over neighbors may be biased by the presence of such unobservable local amenities. I develop a framework to correct this problem based on a careful delineation of the information households could have access to before and after they make their decisions. The model I build has the advantage over the literature of being able to produce self-consistent predictions about demographic changes. I deal with the low frequency of observations in my data set, the decennial census, by simulating local housing markets between data collection periods. After controlling for type-specific preferences for the physical amenities of neighborhoods, I find a universal preference for higher income neighbors. In contrast to much of the literature, my results suggest white households have no aversion to minority neighbors. In the third chapter I estimate the effect of parental credit scores on the child's probability of attending and completing college. Parents in the US are increasingly supplementing the student loans available to their children with unsecured debt in their own name. This is the first paper on this topic to make use of direct observations of credit scores, rather than rely on proxies such as wealth shocks. I find that good parental credit significantly improves the child's probability of attending college, with a smaller (although still significant) effect on the probability of completing a four-year degree. I provide evidence that the estimated relationship is causal and not biased by, for example, unobserved ability. Additionally, I show that credit scores may affect attendance through channels other than access to the student loan market. I hypothesize households substitute the potential to borrow for precautionary savings"--Pages iii-iv.

This thesis consists of three unrelated essays in the fields of labor and urban economics. The first essay exploits the creation of a formal college system in Quebec in the late 1960's as a quasi-experiment to estimate the value of community college. Focusing on the effect of the policy on English-speaking Quebecois, the creation of the CEGEPs (Colleges of General and Vocational Education) is shown to increase schooling by about a third of a year for both men and women, without diverting students from university. Despite increasing educational attainment, estimates of the impact of CEGEP on wages are negative. Analysis suggests the negative estimates can be understood as a combination of lost labor market experience, a decrease in the return to university, and an insignificant return to CEGEP. The results are robust to the inclusion of controls and across years of data. Possible interpretations of the results are discussed. The second essay, co-authored with William Wheaton, examines the relationship between labor market agglomeration and wages. Using the 5% public use micro sample of the 1990 U.S. census, we find that observationally equivalent workers in the manufacturing sector earn higher wages when they are in urban labor markets that have a larger share of national or metropolitan employment in their same occupation and industry groups. Quantitatively, the effect is large, with an elasticity (measured at the means) of between 1.2 and 3.6 for these effects. We interpret the willingness of firms to pay more for equivalent workers in dense markets as evidence of an agglomeration economy in urban labor. The third chapter estimates the effect of employment dispersion on average commute times in American cities. Using a sample of over two hundred cities, I find that residents of cities where employment is more geographically disperse have lower average commute times than residents of cities where employment is more centralized. The results are robust to the inclusion of city fixed effects. An instrumental variables strategy is employed to try to account for potential simultaneity between changes in employment dispersion and changes in commute times.

American cities, after decades of decline, are regaining affluent and highly educated residents. This dissertation examines these trends at the neighborhood level and documents that resurgence, in the form of gentrification, is prevalent in cities, specifically in downtown neighborhoods near employment centers. My results indicate a fundamental shift from city center decline to growth around 1990, which motivates my focus on exploring two novel causes of gentrification. In Chapter 1, I demonstrate that declining gender wage gaps since 1980, and the associated influence on female labor force participation and marriage decisions, are one cause of gentrification. As women increasingly invest in human capital and delay marriage they are more likely to move to, and remain in, urban neighborhoods close to employments areas, allowing these neighborhoods to develop high-quality amenities which facilitate further redevelopment. I document that as the gender wage gap declines so too does the probability of marriage and that, in turn, marital status factors heavily into family residential location decisions, with singles systematically opting to live closer to employment centers. Overall, I find that falling gender wage gaps had a significant but heterogeneous effect on neighborhood prosperity that benefited those neighborhoods nearest the city center. Specifically, I find that the drop in the gender wage gap from 1970 to 2010 can explain 40\% of the average national change in city center income over the same period. One potential factor that influenced the decline in gender wage gaps was a shift in the labor market returns to social skills, a shift that disproportionately aligned with female skill endowments relative to men. I return to this topic in Chapter 3. In Chapter 2, I document the role that condominium development played in gentrification. The advent of condominiums offered high income individuals a legal form to own, rather than rent, high-density real estate close to employment centers. I use condominium conversion ordinances at the municipal level as a source of exogenous variation in condominium development. Using a differences-in-differences set-up, I find that the passage of ordinances limited the development of condominiums in cities subject to regulations that made it more costly to convert housing stock to condominiums. With this approach, I am able to establish a causal effect of condominium development on certain gentrification outcomes, including income and education. In Chapter 3, I introduce a framework for estimating the labor market returns to social capital and explore related mechanisms. I find that the wage return to increasing one's high school network by approximately five friends is equivalent to the return to one additional year of schooling. To better understand the mechanisms that underlie this return, I introduce a game theory model wherein students optimize their time allocation between studying and socializing. Empirical results are consistent with model predictions, specifically in that students make social investments in activities, such as drinking alcohol, that generate friendships at the expense of academic achievement. As an application, I demonstrate baseline estimates that suggest there are positive returns to attending a so-called "party" school for college.

Simple models of urban search matching -- Extensions of urban search-matching models -- Non-monocentric cities and search-matching -- Simple models of urban efficiency wages -- Extensions of urban efficiency wage models -- Non-monocentric cities and efficiency wages -- The spatial mismatch hypothesis : a search-matching approach -- The spatial mismatch hypothesis : an efficiency-wage approach -- Peer effects, social networks, and labor market outcomes in cities -- General conclusion -- Appendix A: basic urban economics -- Appendix B: Poisson process and derivation of Bellman equations -- Appendix C: The Harris-Todaro model.

Over the past thirty years, urban economic theory has been one of the most active areas of urban and regional economic research. Just as static general equilibrium theory is at the core of modern microeconomics, so is the topic of this book - the static allocation of resources within a city and between cities - at the core of urban economic theory. An Essay on Urban Economic Theory well reflects the state of the field. Part I provides an elegant, coherent, and rigorous presentation of several variants of the monocentric (city) model - as the centerpiece of urban economic theory - treating equilibrium, optimum, and comparative statistics. Part II explores less familiar and even some uncharted territory. The monocentric model looks at a single city in isolation, taking as given a central business district surrounded by residences. Part II, in contrast, makes the intra-urban location of residential and non-residential activity the outcome of the fundamental tradeoff between the propensity to interact and the aversion to crowding; the resulting pattern of agglomeration may be polycentric. Part II also develops models of an urbanized economy with trade between specialized cities and examines how the market-determined size distribution of cities differs from the optimum. This book launches a new series, Advances in Urban and Regional Economics. The series aims to provide an outlet for longer scholarly works dealing with topics in urban and regional economics.