Different compositions of aggregate sentiment and their impact on macroeconomic stability

Authored by Frank Westerhoff, Reiner Franke

Date Published: 2019

DOI: 10.1016/j.econmod.2018.07.022

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Mathematical description Other Narrative

Model Code URLs: Model code not found


In recent times a number of agent-based models have been put forward that specify an aggregate sentiment as the difference between optimists and pessimists and let the agents endogenously switch between the two attitudes. The present paper extends this stylized framework by adding a third category, which may be viewed as neutrality. On this basis it then formulates a dynamic three-dimensional Goodwinian model with a special focus on multiple long-run equilibrium positions, which may emerge from just one and very natural nonlinearity in the switching process. The equilibria exhibit the same difference between optimists and pessimists and thus give rise to the same aggregate rate of growth, so that they cannot be distinguished at the macroeconomic level. The feature in which they nevertheless differ is the share of neutral agents. Remarkably, this affects stability. In particular, the trajectories may converge to one of two locally stable equilibrium points, or alternatively to a uniquely determined limit cycle. Coexistence of these attractors is absent in a two-state sentiment dynamics. Generally, the results may also be of interest to empirical business cycle research.
Dynamics herd behavior Goodwin cycles Multiple attractors herding Three-state sentiment dynamics Agent-based modelling