a1 firstname.lastname@example.org, Department of Finance, University of North Carolina at Chapel Hill, Chapel Hill, NC 27514
a2 email@example.com, California Institute of Technology, Division of the Humanities and Social Sciences, Pasadena, CA 91125
a3 firstname.lastname@example.org, Department of Accounting, University of North Carolina at Chapel Hill, Chapel Hill, NC 27514
a4 email@example.com, Department of Accounting, Rice University, Houston, TX 77005.
We examine how analysts respond to public information when setting stock recommendations. We model the determinants of analysts' recommendation changes following large stock price movements. We find evidence of an asymmetry following large positive and negative returns. Following large stock price increases, analysts are equally likely to upgrade or downgrade. Following large stock price declines, analysts are more likely to downgrade. This asymmetry exists after accounting for investment banking relationships and herding behavior. This result suggests recommendation changes are “sticky” in one direction, with analysts reluctant to downgrade. Moreover, this result implies that analysts' optimistic bias may vary through time.