Detecting and Measuring Financial Cycles in Heterogeneous Agents Models: An Empirical Analysis

Abstract

This paper proposes a macroeconometric analysis to depict and measure possible financial cycles that emerge due to the dynamic interaction between heterogeneous market participants. We consider two-type heterogeneous speculative agents: Trend followers tend to follow the price trend while contrarians go against the wind. As agents’ beliefs are unobserved variables, we construct a state-space model where heuristics are considered as unobserved state components and from which the conditions for endogenous cycles can be mathematically derived and empirically tested. Further, we specifically measure the length of endogenous financial cycles. The model is estimated using the equity price index for the 1960–2020 period for the UK, France, Germany, Italy, Ireland, and the USA. We find empirical evidence of endogenous financial cycles for all countries, with the highest frequencies in the USA and the UK.