The Lure of the Slant: Analyst Optimism and Asset Prices
National University of Singapore (NUS) - Department of Finance
October 6, 2012
AFA 2012 Chicago Meetings Paper
24th Australasian Finance and Banking Conference 2011 Paper
This paper studies the effect of analyst optimism on asset prices. A common measure used to study analyst influence is ex-post earnings-forecast bias, which is endogenous in a traditional pricing model. Bias not only reflects optimism, which should not affect prices if investors are rational, but also conveys information bias, which affects prices. To identify the optimism effect, I use two instrumental variables: forecast staleness and size-adjusted analyst tenure. Analyst optimism increases with tenure because of career concerns; analyst optimism increases with forecast staleness because of the “walk-down” hypothesis. However, there should be no systematic relation between information bias and tenure or between information bias and forecast staleness. Using these instruments, I identify a conditional price response to analyst optimism. If an optimistic analyst makes an upward revision, then investors respond by paying higher prices for stocks because the analyst is optimistic (in addition to the response to favorable information). However, if an analyst makes a downward revision, then analyst optimism has no effect on prices. Investors respond to the downward revision by paying lower prices for stocks because of unfavorable information only. In addition to the short-term price impact, analyst optimism affects asset prices over a longer term. Robust to trading costs, a zero-investment analyst tenure portfolio is significantly correlated with investor sentiment and earns an average abnormal return of 69 basis points per month for stocks that are likely influenced by good news only.
Number of Pages in PDF File: 60
Keywords: Analyst Optimism, Asset Prices, Earnings-Forecast Bias, Investor Sentiment
JEL Classification: G12, G17, G29working papers series
Date posted: January 24, 2011 ; Last revised: October 7, 2012
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo1 in 0.531 seconds