LAUSR

Abstract We investigate herding at the aggregate market level and for portfolios formed according to firm size, using data from the Singapore Stock Exchange. We find that herding is pronounced at the market level and in each size-portfolio. This herding is shown to be both spurious and intentional for the overall market and larger portfolios, but only intentional for the smaller portfolios. There is also significant evidence of cross-portfolio herding. The results further indicate that herding is more prevalent during rising market conditions. Lagged microstructures (liquidity and volatility) are also shown to exacerbate herding at the aggregate level and in each size-portfolio. These results are robust when we control for global market volatility and market performance. Our analysis also considers the time-varying nature of herding using a state-space model. We find that herding on the market evolves over time with changes in market events, microstructures and investor sentiment. The time-varying herding in most size-based portfolios generally behave in line with the overall market.