Qiao, YuAgglomeration economy has long been proposed to account for an individual firm’s favor for denser environments. Previous strides have linked firm creation and productivity growth to the magnitude of agglomeration. This dissertation addresses three aspects of agglomerative impact on firms’ dynamic that have not been adequately emphasized in the literature. Specifically, the research provides an understanding of how agglomeration affects firms’ decisions on R&D investment, closure and relocation. In Chapter 2, I develop a simple Cournot type, two-stage competition model that reveals firms tend to reduce their R&D investment more in denser locations than in less dense ones with the presence of knowledge spillover. This implies that local agglomeration strengthens the negative relationship between knowledge spillover and R&D efforts. I then use firm-level data from China to test this theoretical prediction. The Tobit model yields estimated results that are consistent with the theoretical prediction. That is, the R&D effort is negatively correlated with knowledge spillover and the magnitude of the negative relationship increases along with localization agglomeration. The impact of geographic concentration on firm survival is studied in Chapter 3. Agglomeration economy encourages firm birth and growth, while agglomeration diseconomy accelerates firm death. The net impact of agglomeration on firm survival depends on the relative strength of agglomeration economy and diseconomy. Drawn upon an establishment-level data from Maryland, the essay finds empirical evidence supporting the claim that urbanization negatively affects survival, while specialization, diversity and employment centers reduce hazards for some industries. The finding indirectly evidences that the firm selection effect contributes to the productivity advantage of big cities. Firms frequently make spatial adjustments to accommodate their change in operation over time. Agglomeration economy could be one essential influence on a firm’s relocation decision-making. Chapter 4 delves into the relocations of service firms within the Baltimore Metropolitan Region. The nested logit model shows a higher probability for firms choosing a location with a high level of agglomeration. The estimates suggest diversity might be more important than specialization at the margin for intra-metropolitan relocation. Also identified is a more prominent localization effect than urbanization effect on firm intra-metropolitan relocation.enDissertationEconomicsUrban planningRegional studiesAgglomerationKnowledge SpilloverR&DRelocationSurvival