Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Using both a linear regression method and a probability-based method, we find that on average, analysts place larger than efficient weights on (i.e., they overweight) their private information when they forecast corporate earnings. We also find that analysts overweight more when issuing forecasts more favorable than the consensus, and overweight less, and may even underweight, private information when issuing forecasts less favorable than the consensus. Further, the deviation from efficient weighting increases when the benefits from doing so are high or when the costs of doing so are low. These results suggest that analysts' incentives play a larger role in misweighting than their behavioral biases. Copyright 2006, Oxford University Press.