Discussion

Mark Gertler and Martin Eichenbaum followed up on comments made by one of the discussants, BenBernanke. Gertler noted that the zero lower bound (ZLB) on the nominal interest rate might have played its biggest role from late 2008 to 2010, before quantitative easing and forward guidance produced their full effects. This period was also characterized by a rise in aggregate uncertainty, he argued, which might have interacted with the binding ZLB. Gertler identified two channels through which increaseduncertainty couldhave operated. First, increasedprecautionary savings further reduce the neutral interest rate, bringing it closer to the ZLB. Second, higher uncertainty raises the spread between shortand long-term interest rates, because the ZLB creates an option value by putting a lower bound on future short-term rates. Gertler encouraged the authors to quantify these channels using theirmodel. Eichenbaumpointed out that the binding ZLB took place during a period of large changes in fiscal spending.According to theHutchins Center Fiscal ImpactMeasure, fiscal policywas strongly expansionaryuntil 2011, he argued, before turning very contractionary until 2014. Eichenbaum wondered whether the authors could investigate the interplay between fiscal policy and the ZLB by repeating their empirical exercise using the preand post-2011 subsamples. He admitted that theremight be power issues with such a short sample. The authors acknowledged that strongly countercyclical fiscal policy could in theory explain the empirical irrelevance of the ZLB that they document in the paper. However, this explanation is hard to reconcile with their evidence on the response of long-term interest rates. Fiscal policy alone could not explain why the behavior of these rates did not change during the ZLBperiod, they argued. This is suggestive of a dominant role for monetary policy in the form of quantitative easing and forward guidance, according to the authors.