The evolution of herding behavior in stock markets: Evidence from a smooth time-varying analysis

Abstract

We apply a new, nonparametric approach to study time-varying herding behavior. Using this approach to compare herding behavior between the A-share, Hong Kong, and the U.S. stock markets over the past two decades, we present several new findings. First, before the Global Financial Crisis, the A-share stock market exhibited a pro-longed yet weakening herding behavior, which was mainly driven by non-fundamental factors. Second, periods with no herding and periods with adverse herding alternated between 2008 and 2021. Adverse herding, mainly driven by fundamental factors, intensified during market turbulence. Third, both the Hong Kong and the U.S. stock markets displayed adverse herding behavior persistently. In sum, our approach provides some new evidence on time-varying herding in both emerging and developed markets.