Monetary policies in a world of biased agents

Our perceptions and our forecasts are often systematically biased. These biases affect economic activity.

Professor Paul de Grauwe has co-authored a new discussion paper with Professor Yuemei Ji for the Centre for Economic Policy Research, which models some aspects of these biases.

Abstract

"Our perceptions and our forecasts are often systematically biased. These biases affect economic activity. In this paper we model some aspects of these biases. To do so we use a behavioural macroeconomic model. In this model agents have cognitive limitations, preventing them from having rational expectations. Instead, they use simple forecasting rules (heuristics). Rationality is introduced by assuming that these agents learn from their mistakes and are willing to switch to different rules if these are found out to perform better. We apply this model to analyse how systematic optimistic and pessimistic forecasts of the output gap affect the economy and the transmission of monetary policies. A key insight provided by this analysis is that an increased polarisation between optimistic and pessimistic forecasts has the effect of creating agnosticism about the true value of the equilibrium output gap. This also leads to a dominance of purely extrapolative forecasting by economic agents who “fail to see the light”, which in turn enhances the volatility of the business cycle and of inflation. It also reinforces the need for the central bank to stabilize output movements, beyond the need for inflation stabilisation."