Answers are grounded in this paper and cite their sources.
We examine the association between whisper forecast activity and analysts' forecast biases, and provide a rationale for why whisper forecasts occur when analysts' forecasts already exist. We contrast an ex-post measure of analysts' forecast biases with two conditional measures of analysts' ex-ante forecast biases, distinguishing between pessimistic and optimistic biases in analysts' short-term quarterly earnings forecasts, controlling for a number of firm characteristics known to associate with analysts' forecast biases and whisper forecasts. We find evidence in support of the conjecture whisper forecasts occur as a signal to market participants that analysts' forecasts are biased, and that whisper forecasts assist investors in undoing some of the biases in analysts' forecasts. The ability to predict the likelihood of a whisper forecast occurring varies depending on whether one uses ex-post or ex-ante measures of analysts' forecast biases, the model used to estimate forecast biases (forecast revision or bandwagon model), and the direction of the measured bias (pessimistic versus optimistic). The sensitivity of our results to these variations underscores how empirical research using either forecast errors or forecast revisions to identify news or measure bias may yield conclusions that are not generalizable.
No takes yet. Share an insight, caveat, or question.
Teitel et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: