This paper develops and estimates a spatial general equilibrium job search model to study the effects of local and universal (federal) minimum wage policies. In the model, firms post vacancies in multiple locations. Workers, who are heterogeneous in terms of location and education types, engage in random search and can migrate or commute in response to job offers. I estimate the model by combining multiple databases including the American Community Survey (ACS) and Quarterly Workforce Indicators (QWI). The estimated model is used to analyze how minimum wage policies affect employment, wages, job postings, vacancies, migration/ commuting, and welfare. Empirical results show that minimum wage increases in local county lead to an exit of low type (education<12 years) workers and an influx of high type workers (education_12 years), which generates negative externalities for workers in neighboring areas. I use the model to simulate the effects of a range of minimum wages. Minimum wage increases up to $14/hour increase the welfare of high type workers but lower welfare of low type workers, expanding inequality. Increases in excess of $14/hour decrease welfare for all workers. I further evaluate two counterfactual policies: restricting labor mobility and preempting local minimum wage laws. For a certain range of minimum wages, both policies have negative impacts on the welfare of high type workers, but beneficial effects for low type workers.