Market states and institutional herds: evidence from the Taiwan stock market

This study examines whether institutional herds, as defined by Chung and Kim, influence future excess returns across different market states in Taiwan stock market. Using specialized trading data aggregated from three major institutional investor groups, we focus on institutional herding behavior during market collapses and booms. Fama–MacBeth regression results show that overall institutional herding positively contributes to future excess returns during market downturns and across the full sample, but negatively during market booms. By dividing stocks into quintiles based on herding intensity, we find that concentrated herding (top 20%) tends to reduce returns during downturns, whereas broader herding helps stabilize the market. In booms, leading herding may enhance returns, while widespread herding tempers excessive gains. Portfolios with the highest institutional herding scores also yield higher cumulative returns than those with the lowest scores (bottom 20%). These findings underscore the critical role of institutional herding, particularly high-intensity herds in shaping market dynamics under extreme conditions, providing a comprehensive understanding of how institutional herds influence the Taiwan stock market.This study offers empirical insights into how institutional herding, particularly under extreme market conditions, influences future excess returns in Taiwan’s stock market. By distinguishing herding intensity and market phases, the findings reveal that herding behavior enhances return predictability during downturns but dampens performance during booms. These results contribute to a deeper understanding of institutional dynamics and their implications for market efficiency and asset pricing.