On Concurrent Solutions in Differential Games

We examine solutions in which neither player is worse off from the leadership of one in a policy maker-public game. The loop model of dynamic games is used. Outcome space is dotted with equivalence classes of solutions. The Dynamic Stochastic General Equilibrium (DSGE) results and their New Keynesian variants might represent one category. The economy is the neighborhood of a market-clearing equilibrium with Pareto-optimal properties modulo frictions. Our interest lies in the ‘old’ Keynesian genus where the representative state is one of involuntary unemployment. Two information sets are relevant. In the first case, agents look to the past and the present. In the second, they are bound by the information provided in the present. The standard analysis pertains to DSGE models under full information. We show, in contrast, that in a situation of structural disequilibrium and feedback information, all parties are better off reneging on the social compact to achieve a superior class of solutions.