“…To do this, for the sake of simplicity and following the literature (see, for example, Wilson, 1979;Obstfeld, 1983;Turnovsky, 1997Turnovsky, , 2000Schubert and Turnovsky, 2006), we abstract from any uncertainty about the duration of the shock and assume that the demand shock occurs at time t = 0 and expires at time t = T and that all agents know the duration of the shock. Hence, only at time t = 0 does the shock constitute news, whereas at time T, the removal of the shock is no news, as all agents have taken it into account.…”