Average inflation targeting when agents are learning

Abstract

Many central banks consider the change of the approach to the monetary policy. One of the options is average inflation targeting. Under rational expectations this policy leads to a tangible increase in social welfare. In this paper, we demonstrate that in the case of adaptive expectations, on the contrary, it is more likely to lead to a decrease in welfare. In particular, considering the case of adaptive learning, we show that attempts by the central bank to stabilize average inflation lead to unnecessarily high output volatility. Moreover, even if the central bank includes average inflation in its loss function, it still would not optimally alternate periods of low and high inflation. Thus, the implementation of average inflation targeting seems to be inappropriate. We consider the case of the absence of zero lower bound, which allows us to formulate conclusions that are relevant for developing and developed countries which have experienced a prolonged acceleration of inflation in recent years.