Social learning expectations: microfoundations and a Dynare toolbox

Social learning (SL) is a behavioral model in which expectations and the resulting aggregate dynamics stem from the interactions of a large amount of heterogeneous agents. Nonetheless, this framework has so far lacked micro-foundations and a general-solution method. This paper bridges these two gaps with: (i) a micro-founded New Keynesian model with social learning expectations; (ii) a general solution method that we implement in a Dynare toolbox that solves any linear state-space model with SL expectations. The resulting framework provides a self-contained tool to contrast policy analysis under SL and rational expectations. As an illustration, optimal monetary policy rules are studied under the two expectation regimes.