This paper develops the theoretical reasons why user charge finance leads to improved government efficiency and reduced expenditures. It is argued that user charges constrain bureaucratic behavior, thus forcing the bureau to operate closer to the true demand curve for the service. Further, it is argued that user charges reduce consumer demand by connecting use and payment and by breaking fiscal illusion. This analysis leads one to expect that greater reliance on user charge finance would lead to lower government expenditures. An empirical test is performed which supports this hypothesis. The theoretical results combined with this empirical test suggest that user charges do generate lower government expenditures and a more efficient government.