Crowdsourced Earnings Forecasts, Analyst Forecast Timing, and the Speed of News Recognition

• In Event: Financial Analysts: Forecast and Revision Behavior

Abstract

We combine theory with a unique empirical setting to investigate how the introduction of Estimize, an online aggregator of crowdsourced earnings forecasts, impacts IBES analysts’ forecast timeliness and the speed at which new information is impounded into stock prices. Consistent with forecast timing theory, our study finds that IBES analysts start issuing forecasts earlier for firms for which Estimize coverage commences. This shift in forecast timing is driven by experienced IBES analysts and is generally accentuated by increased availability of public information. We also find that firms for which crowdsourced forecasts become available start reflecting bad news faster in their stock prices. Surprisingly, we also find that the recognition of good news becomes delayed post Estimize. Results are robust to winsorization, changes in analyst coverage, management forecasts, and fixed effects allowing for varying trends of potentially diverse populations, as well as to matching estimators.

Authors

• Rajiv D Banker, Temple University

• Joshua Khavis, Temple University

• Han-Up Park, Temple University