We present a model where heterogeneous districts choose both whether to experiment and the policies to experiment with. Since districts learn from each other, the first-best requires that policy experiments converge so that innovations are useful also for neighbors. However, the equilibrium...