The article offers a descriptive and statistical analysis of the differentiation of registered unemployment across local areas in Poland. The author looks at the determinants of registered unemployment and changes in the trend from 2002 to 2011. The article analyzes the impact of geographical, historical and administrative factors on differences in unemployment at the county level. Moreover, on the basis of a theoretical model, the author analyzes changes in unemployment in the studied period. The research makes use of several basic methods used in spatial econometrics, including a fixed-effects regression model. The study shows that some of the differences in unemployment rates by county are due to administrative factors, such as the distance from a county to the largest city in a specific province and the type of county: whether it is an urban or rural area. Counties located closer to the administrative centers of their respective provinces and counties located in urban areas tended to have lower average unemployment rates during the studied period, the authors say. Historical factors also play a role, according to the authors: unemployment rates in areas of Poland formerly under Austrian and German rule tended to be lower than in other regions. The analysis of changes in unemployment rates in the studied period shows that labor market adjustments at the county level are dynamic and asymmetrical, the authors say. The analysis also shows that past unemployment rates have a significant impact on the growth of unemployment in future periods and that the scope of changes in unemployment is asymmetrical. Generally, labor market adjustments in urban counties were more asymmetrical than in rural counties, the authors say.