애널리스트 낙관주의와 불투명성 및 주가급락에 관한 연구

We examine whether the optimistic bias of analysts can increase firms’ opaqueness and thur affect the firms’ stock price crash risk. Analysts’ optimistic bias could be more problematic to investors in the emerging market countries (Chen and Hameed, 2006). In this paper, using Korean manufacturing companies listed in the KRX KOSPI for the sample period spanning from 2002 to 2012, we find the following results. First, we find that analysts’ optimistic bias can significantly increase firms’ opaqueness, proxied with R-squared in the market model type regressions (Morck et al., 2000; Jin and Myers, 2006). We show that our analyst coverage proxies significantly increase opaqueness. Further, we show that the results are driven by coverage by optimistic analysts and the ratio of optimistic reports. In contrast, we do not find significant evidence that coverage by conservative analysts do not significantly improve the opaqueness. Combined together, our findings could dispute the validity of analyst coverage as a proxy for a firm’s transparency. The question would be more important in developing countries as Korea in that the information by analysts might be both biased and less informative. Second, we find that optimistic bias can significantly increase the three crash risk proxy variables: 1) occurrence of negative extreme return (CRASH), 2) negative conditional skewness (NCSKEW), and 3) down-to-up volatility (DUVOL) (Hutton et al., 2009; Kim et al., 2011; Chen et al., 2001). We provide evidence that optimistic analyst coverage, proxied by number of optimistic reports and number of optimistic analysts, can significantly increase the stock price crash risk. These findings support Xu et al. (2013). Further, our question of whether analyst coverage is a valid proxy for the transparency could be complimented in that firms with greater analyst coverage might have greater stock price crash risk as well. Lastly, we explore more by applying two additional optimistic bias related variables: best analyst dummy (BEST) and private information weight dummy (WEIGHT). Best analysts (All-star analysts in the U.S.) are known to be less optimistic than non-best analysts. In addition, analysts who exaggerate their private information are known to be more optimistic in Korea (Kang and Kim, 2012). We find that coverage by best analysts can significantly attenuate both opaqueness and CRASH and DUVOL. We could not find significant relation between opaqueness and analysts’ weight on private information. However, as we conjectured, WEIGHT increases CRASH and DUVOL and the relations are significant. Our results could contribute to the existing literature at least for two folds. First, we add another evidence to the literatures on the effect of analysts’ optimistic bias on opaqueness and crash risk after Xu et al. (2013). Moreover, it is the first evidence in Korea and our results elucidate that investors need to be cautious when they use the information provided by analysts because the information could be biased.