Forecast Walk-Downs and Strategic Incentives for Bias

Date
2018-08-24
Authors
Patatoukas, Panos
Yezegel, Ari
Zeng, Jieyin
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Journal Title
Journal ISSN
Volume Title
Publisher
Volume
Number/Issue
Starting Page
Ending Page
Alternative Title
Abstract
Are strategic incentives for bias among forecasters a necessary condition for forecast walk-downs? We identify the group of professional macro forecasters affiliated with the Federal Reserve System as a setting where forecasters are free from strategic incentives for bias. Unlike sell-side analysts, who have incentives to curry favor with management when forecasting firm-level profits, this group of professional macro forecasters faces strong incentives to produce unbiased forecasts of the U.S. economy. Remarkably, however, we document a walk-down in their GDP growth forecasts, even after excluding the corporate profit component of U.S. output. Whereas most prior explanations for forecast walk-downs are conditioned on preexisting incentives for bias, we find that asymmetrically high forecasting difficulty in downturns relative to upturns of the U.S. economy can explain the macro walk-down even in the absence of such incentives. Overall, our paper contests the traditional view of forecast walk-downs as de facto evidence of strategic incentives for bias. Time-series and cross-sectional tests further illustrate the relevance of our findings for firm-level studies. One overarching implication is that research on sell-side analysts’ forecasts should consider strategic incentives for bias alongside information about the state of the U.S. economy and heterogeneity in the cyclical exposure of individual firms to macro fluctuations.
Description
Keywords
Forecast walk-downs, Professional macro forecasters, Sell-side financial analysts
Citation
Extent
Format
Geographic Location
Time Period
Related To
Rights
Rights Holder
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.