We examine how well implied volatility forecasts future stock market volatility. If markets are efficient and the option pricing model is correct, the implied volatility calculated from option prices should be an unbiased and informationally efficient predictor of future volatility; that is, it should correctly impound all available information, including the asset's price history. However, numerous studies have found that the implied volatility forecast is biased and/or is not informationally efficient. We re-examine this issue using S&P500 futures options data, a more active market which is less susceptible to measurement error than the OEX options market considered in most previous studies. Our findings are that, first, implied volatility has strong predictive power and generally subsumes the information in historical volatility; second, prediction and efficiency results are quite sensitive to the forecasting horizon and whether the data period covers the October 1987 stock market crash; and third, at least in the S&P500 options market, measurement error due to such factors as bid–ask spreads, non-synchronous price observations and discrete prices does not significantly affect the results.