This paper presents an experimental examination of the Falkinger (1996) mechanism for overcoming the free-rider problem. The basic idea of the mechanism is that deviations from the mean contribution to the public good are taxed and subsidized. The mechanism has attractive properties because (i) it induces higher contributions to the public good and can implement an efficient level of contributions as a Nash equilibrium, (ii) the government budget is always balanced irrespective of the level of individual contributions, (iii) it is simple and policy makers need only little information to implement the mechanism. To examine the empirical properties of the mechanism we conducted a large series of experiments. It turns out that the introduction of the mechanism generates immediate and large efficiency gains. This result is robust throughout many different experimental settings. Moreover, in the presence of the mechanism the Nash equilibrium is a rather good predictor of behavior.