The paper offers a statistical analysis of the determinants of wage differences in Poland in 2002-2006. These analyses are made with the use of a theoretical model that combines the Solow-Summers efficiency wage model (a neo-Keynesian economic model) and neoclassical growth models. This combined model shows that wage differences are determined by macroeconomic variables such as labor productivity and the unemployment rate. The authors use statistical data collected by the Central Statistical Office (GUS) on wages, labor productivity and unemployment in individual counties in Poland in 2002-2006. The analysis shows that the highest wages are reported in large urban centers and “counties that are centers of local economic development,” the authors say. On the other hand, the lowest wages are in typically agricultural counties and those that were home to many former state-run farms. In counties with a high rate of unemployment, relative wages in 2002-2006 were usually lower than in counties with low unemployment rates, while a high level of relative labor productivity was usually accompanied by a high level of relative wages. Regression equation estimates for each of the country’s 16 provinces show that relative wages had a different impact on unemployment rates and relative labor productivity in individual provinces.