Analysts' Superiority in Processing Public Information: Evidence from Recommendation Revisions

Abstract

In this paper, I study analysts' superiority over the market in processing publicly disclosed earnings information by examining a sample of recommendation revisions issued subsequent to annual earnings announcements within a short period of thirty trading days. The main findings of this study are as follows: First, I provide strong evidence that these recommendation revisions convey valuable information to the market for clarifying the long term implications of recently released earnings. These revisions significantly alter the market's belief about the value implications of announced earnings, suggesting that analysts do have superiority over the market in processing pubic information. Also, the extent of this superiority is positively related to analysts' performance in picking stocks and forecasting earnings. Recommendation revisions issued by analysts with superior performance can make the market revise its assessment about the value implications of previous earnings to a much greater extent than those issued by analysts with moderate performance. Moreover, the extent of this superiority increases with the level of information complexity of earnings signals. Analysts' information is even more valuable to the market for reevaluating previous earnings when the earnings information is more difficult to analyze. Lastly, on average, the extent of this superiority declines after Regulation Fair Disclosure, but still remains significant, suggesting that analysts do not solely rely on inside information from the management to interpret public information. Actually, the decline in the extent of superiority is more likely due to a great increase in the number of revisions issued by analysts whose expertise is not in processing public information.

Prior studies document that investors also use subsequent earnings announcements to adjust their estimate of the value implications of previous earnings. This study finds initial evidence that when analysts' information and subsequent earnings announcements provide consistent predictions on how previous earnings is misinterpreted, subsequent earnings announcements become less useful to investors for updating their beliefs regarding the implications of previously released earnings. This paper also compares the extent of analysts' superiority in processing publicly released earnings information across industries and find that analysts exhibit a greater degree of superiority for firms in the manufacturing and retail industry.