10－K 보고서의 텍스트 분석에 근거한 경쟁변수가 재무분석가의 이익예측 속성에 미치는 영향

[Purpose]The purpose of this study is to examine the relation between the text－based competition measures and properties of analysts’ earnings forecasts. [Methodology]The research uses the frequency of competition－related words suggested by Li et al.(2013) as a proxy for competition measure. There are two expectations on the results. On the one hand, as competition among firms are more competitive, information environment gets worse and analysts’ forecasts become less accurate due to lack of corporate information. This is because firms are concerned about the proprietary costs. On the other hand, as competition among firms are more competitive, information environment gets better and analysts’ forecasts become more accurate due to the increased corporate disclosure. This is because managers try to reduce information asymmetry between them and owners. [Findings]The results are as follows： First, as suggested by Li et al.(2013), frequency of competition－related words is higher, in other words, competition among firms are more competitive, the properties of analysts’ forecasts are improved. Their forecasts are more accurate, less biased, and less dispersed. Second, these relations are pronounced in such firms with better information environment as bigger firms, firms with a larger number of analyst following, and firms with low R&D expense. [Implications]This study is the first study to examine the relation between the text－based competition measures and properties of analysts’ forecasts and shows that the relation is more pronounced in firms with better information environment. It sheds light on the needs of text－based research on annual reports in Korea.