Risk Management for Monetary Policy at the Zero Lower Bound

As labor markets improve and projections have inflation heading back toward target, the Fed has begun to contemplate lifting the federal funds rate from its zero lower bound (ZLB). Under what conditions should the Fed start raising rates? We lay out an argument that calls for caution. It is founded on a risk management principle that says policy should be formulated taking into account the dispersion of outcomes around the mean forecast. On the one hand, raising rates early increases the likelihood of adverse shocks driving a fragile economy back to the ZLB. On the other hand, delaying lift-off when the economy turns out to be resilient could lead to an unwelcome bout of inflation. Since the tools available to counter the first scenario are hard to implement and may be less effective than the traditional tool of raising rates to counter the second scenario, the costs of premature lift-o exceed those of delay. This article shows in a canonical framework that uncertainty about being constrained by the ZLB in the future implies an optimal policy of delayed lift-o. We present evidence that such a risk management policy is consistent with past Fed actions and that unconventional tools will be hard to implement if the economy were to be constrained by the ZLB after a hasty exit.