Contrast Effects and Analyst Forecasts
Contrast effect occurs when decision makers unconsciously interpret a signal by contrasting it with the preceding signal. Using the setting of analyst earnings forecast revisions in response to firms’ earnings announcements on consecutive days, we find evidence of contrast effects only if announcing firms on both days are covered by the analyst . The effects are driven by less experienced or skill ed analyst s and by scenarios where the benchmark earnings news on day t 1 likely receive s more attention, such as larger earnings surprise or larger announcing firm on day t 1. While prior research finds that investors tend to contrast earnings news against the preceding day’s earnings news from bellwether firms, analysts do not appear to benchmark against these firms if they are not covered by the analysts . We also find that
contrast effects adversely impact the analyst forecast accuracy and analysts seek to correct initial errors with subsequent revisions, consistent with contrast effect as a psychological bias rather than an intentional strategy. Additional analyses suggest that our results are not driven by mechanical relation s between earnings news on consecutive days such as information transfer and are exacerbated when analyst s suffer from limited attention or decision fatigue