SHORT-TERM HERDING EFFECT ON MARKET INDEX RETURNS

The study analyzes the predictability of stock market returns based on the previous day's cross-sectional market-wide herd behavior. Assuming that herding may lead to stock price overreaction and result in subsequent price reversals, I suggest that daily stock market returns may be higher (lower) following trading days characterized by negative (positive) market returns and high levels of herding. Analyzing the daily price data for S&P 500 Index and all its constituents and employing two alternative market-wide herding measures based on cross-sectional daily deviation of stock returns, I document that the days of both positive and negative market returns tend to be followed by price reversals (drifts), if the market-wide levels of herding are high (low). The herding effect on the next day's stock market returns is found to be more pronounced following the days when the sign of the market return corresponds to the direction of the longer-term stock market tendency and the days characterized by relatively large stock market movements. The effect also remains significant after accounting for the specific numerical value of the market return.