Zero Bound on Nominal Interest Rates and Ex Ante Positive Inflation: A Cost Analysis

Summers (1991) proposes that a central bank, in conducting monetary policy, should pursue a small but positive ex ante inflation rate even before nominal interest rates hit the zero bound. He insists that the central bank can thus reduce the social costs brought about by negative shocks to the economy. This line of argument, however, does not explicitly consider the social costs arising from the positive inflation rate itself, but emphasizes only the benefits of a reduced risk of hitting the nominal interest rate bound. In this paper, I consider both the benefits and costs of maintaining a positive ex ante inflation rate as a way of circumventing the discomfort imposed by zero bound constraints. I show that this trade-off between the costs and benefits broadly determines the desirable rate of positive ex ante inflation that the central bank should pursue. By using simple model simulations, I conclude with these two points. Firstly, there in fact exists a desirable rate of positive ex ante inflation whose benefits exceed its costs. By parameterizing the model with average values from past Japanese experience, this finding is also shown to have been applicable to Japan. Secondly, this desirable level of the inflation rate largely depends on the degree of forward-lookingness in the economy, and the size and persistence of shocks generated.