Niu, YiCities started the unprecedented growth about one hundred years ago. Their importance and significance are reflected by their high productivities and spatial concentrations. The understanding on urban development would help improve urban management and policies and increase wellbeing of urban residents. The three related essays in this dissertation try to improve the understanding from the perspectives of employment centers and agglomeration economies, interactions between labor and housing markets, and the behavior of local governments. The first essay examines the role of employment centers on economic development. The theoretical literature suggests that agglomeration economies are the main force behind the formation and evolution of employment centers, as well as behind economic growth in general. Applying the birth model to employment centers in Maryland, I find agglomeration effects are increased by the centers, particularly those with high employment size or industrial diversity. Ignoring employment centers may overestimate the agglomeration effects when using the fixed distance measurement. Policy implications are local officials may use employment centers as a vehicle to promote economic growth. In the second essay I test the impact of job loss on housing foreclosures. A great challenge in this study, as well as in interactions between labor and housing markets in general, is the geographic mismatch between employment and residential locations. This partially explains the mixed effects of job loss on foreclosures found in the literature. In order to gauge this effect, I develop a job loss vulnerability index using home-work commuting pairs. After fixing the attenuation bias from measurement errors, I find that job loss plays an important role in foreclosure decisions. This essay provides evidence for impact from labor market bust to housing market depression. The third essay estimates the spending pattern of off-budget revenues. The literature assumes different spending preferences of budgetary and off-budget revenues, but empirical evidence are scarce due to the lack of off-budget data. I use land revenues to proxy off-budget revenues in Chinese cities. I find that off-budget revenues do not crowd out budgetary expenditures, and they tend to support visible and tangible projects, rather than some other traditional public spending items that are not quite obvious.DissertationUrban planning