We present a family of mechanisms which implement Lindahl allocations in Nash equilibrium. With quasilinear utility functions this family of mechanisms are supermodular games, which implies that they converge to Nash equilibrium under a wide class of learning dynamics.