Noise traders and the rational investors: a comparison of the 1990s and the 2000s

Purpose – The purpose of this paper is to compare and contrast the effect of individual and institutional sentiments on the US stock returns during a prolonged bull phase of ten years in the 1990s compared to shorter boom and bust cycles of the 2000s. The study is focussed on a set of stocks that are prone to sentiments and speculations. Design/methodology/approach – To compare the dynamic interaction of individual and institutional sentiments and stock returns, the authors use the vector autoregression (VAR) approach. The VAR model has proven to be especially useful for describing the dynamic behavior of economic and financial time series because it does not impose a priori restriction on the structure of the system. Using impulse response function, the authors determine how stock returns respond over time to a shock in institutional and individual sentiments. Findings – The authors find that sentiments of individual investors can affect returns mostly when there is a prolonged upward trend in stock prices, while sentiments of institutional investors can impact the returns when stock market is more volatile. Originality/value – This paper compares the effect of noise traders and rational investors’ sentiment on stock returns during the persistent period of positive abnormal returns of the 1990s and the more volatile stock returns of the 2000s.