Returns to buying upward revision and selling downward revision stocks: evidence from Canada.

Purpose: The purpose of this paper is to investigate the role of earnings forecast revisions by equity analysts in predicting Canadian stock returns Design/methodology/approach: The sample covers 420 Canadian firms over the period 1998-2009. It analyses investors’ reactions to 27,271 upward revisions and 32,005 downward revisions of analysts’ forecasts for Canadian quoted companies. To test whether analysts’ earnings forecast revisions affect stock return continuation, forecast revision portfolios similar to Jegadeesh and Titman (2001) are constructed. The paper analyses the returns gained from a trading strategy based on buying the strong upward revisions portfolio and short selling the strong downward revisions portfolio. It also separates the sample into upward and downward revisions. Findings: The authors find that new information in the form of analyst forecast revisions is not impounded efficiently into stock prices. Significant returns persist for a trading strategy that buys stocks with recent upward revisions and short sells stocks with recent downward revisions. Good news is impounded into stock prices more slowly than bad news. Post-earnings forecast revisions drift is negatively related to analyst coverage. The effect is strongest for stocks with greatest number of upward revisions. The introduction of the better disclosure standards has made the Canadian stock market more efficient. Originality/value: The paper adds to the limited evidence on the effect of analyst forecast revisions on the returns of Canadian stocks. It sheds light on the importance of analysts’ earnings forecast information and offers support for the investor conservatism and information diffusion hypotheses. It also shows how policy can improve market efficiency