TY - UNPB
T1 - Information Diffusion As Reflected In Analyst Forecast Revisions
AU - Schrand, Catherine M.
AU - Taylor, Daniel J.
AU - Walther, Beverly Rayburn
AU - Wang, Clare
PY - 2014/10
Y1 - 2014/10
N2 - In this paper, we examine how earnings information is diffused to capital markets by investigating the analyst’s decision about when to issue a revised earnings forecast following an earnings announcement. At the earnings announcement date, the analyst chooses either to forecast immediately with the opportunity for subsequent revisions, or to wait and revise at any time before the announcement of the subsequent quarter’s earnings. This choice sets in motion in the potential paths for information diffusion. We propose that the analyst’s choice depends on a trade-off between accuracy and speed, and hypothesize three categories of factors associated with that choice: the ease of the forecasting task, analysts’ incentives, and the ability to learn from other analysts (“social learning”). Our findings suggest that the ease of the forecasting task (as proxied by earnings predictability, the informativeness of the press release, and analyst ability) and analysts’ incentives (as proxied by brokerage house size, firm size, and the analysts’ workload) are significantly associated with the decision to forecast immediately. We find little evidence, however, of analysts’ delaying their release of a revised earnings forecast in anticipation of potentially learning from other analysts.
AB - In this paper, we examine how earnings information is diffused to capital markets by investigating the analyst’s decision about when to issue a revised earnings forecast following an earnings announcement. At the earnings announcement date, the analyst chooses either to forecast immediately with the opportunity for subsequent revisions, or to wait and revise at any time before the announcement of the subsequent quarter’s earnings. This choice sets in motion in the potential paths for information diffusion. We propose that the analyst’s choice depends on a trade-off between accuracy and speed, and hypothesize three categories of factors associated with that choice: the ease of the forecasting task, analysts’ incentives, and the ability to learn from other analysts (“social learning”). Our findings suggest that the ease of the forecasting task (as proxied by earnings predictability, the informativeness of the press release, and analyst ability) and analysts’ incentives (as proxied by brokerage house size, firm size, and the analysts’ workload) are significantly associated with the decision to forecast immediately. We find little evidence, however, of analysts’ delaying their release of a revised earnings forecast in anticipation of potentially learning from other analysts.
M3 - Working paper
BT - Information Diffusion As Reflected In Analyst Forecast Revisions
ER -