Byoung-Hyoun Hwang
Cornell University – Dyson School of Applied Economics and Management; Korea University – Department of Finance

Baixiao Liu
Florida State University

Dong Lou
London School of Economics & Political Science (LSE)

May 5, 2014

Abstract:

We hypothesize that analysts with a bullish stock recommendation have an interest in not being subsequently contradicted by negative firm-specific news. As a result, these analysts report downward-biased earnings forecasts so that the company is less likely to experience a negative earnings surprise. Analogously, analysts with a bearish recommendation report upward biased earnings forecasts so that the firm is less likely to experience a strong positive earnings surprise. Consistent with this notion, we find that stock recommendations significantly and positively predict subsequent earnings surprises, as well as narrow beats versus narrow misses. This predictability is concentrated in situations where the motivation for such behavior is particularly strong. Stock recommendations also predict earnings-announcement-day returns. A long-short portfolio that exploits this predictability earns abnormal returns of 125 basis points per month.

Professional forecasters play an integral role in financial markets. They collect, process, and transmit information to market participants, who, in turn, use these reports when making their investment decisions (e.g., Stickel 1995; Womack 1996; Barber et al. 2001, 2003; Kothari 2001). While significantly altering market expectations, these reports may not reflect forecasters’ true beliefs, however. Researchers find, for example, that sell-side analysts sacrifice forecast accuracy and report biased forecasts in order to stimulate trading (e.g., Hayes 1998), obtain access to management (e.g., Lim 2001), and/or generate investment banking business.1In this study, we propose and test for a different source of bias in earnings forecasts. We propose that analysts have an interest in not having recommendations subsequently contradicted by important firm-specific news. Consider, for example, a sell-side analyst with a bullish stock recommendation, i.e., an analyst who signals to the market that she believes that a firm is currently undervalued. If the firm subsequently misses its consensus earnings forecast and experiences a negative earnings surprise, this could be construed as contradicting the analyst’s bullish view on the company and might raise questions about her competency. Similar concerns could arise when a bearish stock recommendation is followed by a strong positive earnings surprise.

We suspect that, to avoid such perceptions, analysts with bullish recommendations report downward biased earnings forecasts, so that the companies are less likely to experience negative earnings surprises. Relatedly, analysts with bearish recommendations report upward biased earnings forecasts, so that the companies are less likely to experience positive earnings surprises. Our idea is related to a large body of literature in economics and psychology suggesting that people avoid disappointment by strategically altering their expectations about desired outcomes (e.g., Bell 1985; Van Dijk et al. 2003).

We begin our analysis by assessing the assumption that market participants partially judge an analyst’s ability by whether her recommendations are followed by firm-specific news that is consistent or inconsistent with her overall view of the company. In particular, we look at analysts whose recommendations are at variance with subsequent earnings surprises, and we test whether these analysts subsequently experience negative career outcomes. Consistent with this notion, we find that after controlling for earnings-forecast accuracy and stock-return-based measures of recommendation performance, analysts in the bottom quintile with respect to the fraction of “consistent earnings surprises” are 1.6% (p-value<0.00) more likely than other analysts to leave the analyst sample. For comparison, the corresponding impact of being in the bottom quintile with respect to earnings-forecast accuracy is 1.5%. Analysts with a high fraction of inconsistent earnings surprises are also less likely to be named an Institutional Investors’ All-Star analyst.

Prior literature provides evidence that the effect of forecast accuracy on our measure of analyst career outcomes is highly nonlinear (e.g., Hong et al. 2000, 2003). That is, while being in the 4th or bottom quintile in terms of earnings-forecast accuracy is of great relevance for the analyst’s career, being in the 3rd or 4th quintile has no meaningful effect. This nonlinearity suggests that the cost of giving up forecast accuracy is limited in certain situations, providing a justification for why analysts sometimes choose to boost their earnings surprise consistency by issuing biased earnings forecasts and, essentially, sacrificing forecast accuracy.

‘Consistent’ Earnings Surprises via SSRN