Crowdsourced Earnings Forecasts: Implications for Analyst Forecast Timing and Market Efficiency

Abstract

We investigate how the arrival of Estimize, a provider of crowdsourced earnings forecasts, impacts IBES analysts’ forecast timeliness and facilitates market efficiency. We find that IBES analysts become more responsive to earnings announcements and start issuing their quarterly forecasts earlier when faced with competition from Estimize. The Estimize effect is strongest when Estimize quarterly forecasts pose a direct competitive threat to IBES — when Estimize forecasts are present within 3 days of earnings announcements (i.e., are issued early). Specifically, IBES analysts become more responsive to earnings announcements post Estimize, and issue more than 9% of their one-quarter-ahead forecasts earlier in the quarter when early Estimize coverage is present in the prior quarter. We also document that this increased responsiveness of IBES analysts facilitates market efficiency as it results in greater immediate market reaction to earnings surprises and mostly eliminates the post-earnings-announcement drift.

Keywords: crowdsourcing, analyst forecasts, post-earnings-announcement drift