First Impressions and Analyst Forecast Bias

Research Retrieved: July 2019
Retrieve the paper from the SSRN

Issue

The authors investigate whether analysts are subject to first impression bias, causing them to let first impressions of companies influence their forecasts of performance. First impression bias causes unreasonably optimistic (pessimistic) forecasts of future performance if a firm does particularly well (poorly) in the year before the analyst follows it. The authors assess this on the basis of field data, using a sample of 1,643,089 firm-announcement-analyst observations spanning from 1983 until 2016.

Key Findings

  • Analysts’ first impressions influence performance forecasts: If an analyst has a positive (negative) first impression of a firm, the analyst tends to make optimistic (pessimistic) forecasts of earnings per share and set optimistic (pessimistic) price targets relative to other analysts covering the same firm at the same time. A first impression of a firm is determined by the stock return performance 12 months before the analyst first issues the forecast. The first impression is positive if that stock was in the top 10th percentile of its industry and as negative if it was in the bottom 10th percentile.
  • Analysts’ first impressions influence recommendations: A positive first impression makes analysts more likely to recommend buy, while a negative one makes analysts less likely to recommend buy. The opposite is true for sell recommendations.
  • Negative first impressions last longer: While the positive impression effect on earnings-per-share-forecasts lasts for 24 months, the negative impression effect on these forecasts lasts at least 72 months.
  • Analysts’ first impressions influence coverage length: Analysts who have a positive first impression of a firm tend to cover that firm one month longer than neutral analysts. Furthermore, analysts who have a negative impression are more likely to stop following the firm approximately five months sooner than neutral analysts.
  • Analysts’ first impressions influence accuracy: Previous studies show that analyst forecasts of earnings per share are, on average, overly optimistic. The authors assess whether first impressions affect the accuracy of these forecasts. For that purpose, they measure analysts’ mean absolute forecasting errors relative to other analysts forecasting the same firm at the same time. While positive first impressions have no effect on accuracy, negative first impressions lead analysts to be more accurate. This indicates that negative first impressions counteract the optimism of the average analyst.
  • Investors account for the first impression bias: Investors seem to be able to understand the first impression effects and react accordingly. The market reacts less positively when analysts who have positive first impressions issue positive revisions, and more negatively when they issue negative revisions. Similarly, the market reacts more positively when analysts who have negative first impressions issue positive revisions, and less negatively when they issue negative revisions.

Relevance for Practice

The study shows that even professional analysts, who are sophisticated and highly motivated market participants, fall prey to the first impression bias. It is hence reasonable to assume that the findings hold for a wide range of investor types. Further, when assigning analysts to follow a particularly successful or unsuccessful firm, an investor may benefit from taking into account the first impression biases their analysts may have in future forecasts, price targets, and recommendations.