The paper offers an empirical analysis of the effectiveness of spending on junior high schools in Poland on the basis of students’ performance during external exams in 2002-2006 and Central Statistical Office (GUS) data on local government budgets. The study makes use of multilevel models, including an “educational value-added model.” Even though the methods used by the author do not make it possible to determine the exact cause-and-effect relationships involved, they are more reliable than methods used in most previous studies of this kind in Poland. The results obtained by Jakubowski suggest that, under the existing institutional arrangements, the total expenditure of Poland’s central and local governments per student has no influence on the average increase in knowledge among junior high school students. Greater spending neither upgrades the quality of education nor helps equalize educational opportunities, Jakubowski concludes. These findings carry a clear message for decision makers and local government officials responsible for educational policy showing that their policies do not necessarily contribute to an increase in the country’s human capital stock. Further investment in the national education system requires a rethinking of the effectiveness of individual measures, Jakubowski says. At the same time, he adds that further research is needed to look for more effective educational programs to improve the quality of education in Poland.