Modern macro even Paul Krugman will love
An open question: How can the results in this paper be reconciled with results by John Cogan, Tobias Cwik, John Taylor, and Volker Wieland, who seem to perform a similar policy simulation in a similar model but reach a very different conclusion? Are there subtle differences in the models? Or subtle differences in the policy experiments? Or did one team simply make a mistake of some sort?
Figuring out why these two prominent teams of researchers come to opposite conclusions about fiscal policy multipliers, and which conclusion is more applicable to actual policy, would be a good paper topic for an ambitious grad student.