Do risk disclosures in annual reports improve analyst forecast accuracy?. Issue 4 (2nd October 2017)

• Record Type: Journal Article
• Authors: Wang, Xiongyuan; Li, Yanqiong; Xiao, Min
• Is Part Of: China journal of accounting studies. Volume 5:Issue 4(2017)
• Page Start: 527
• Page End: 546
• Publication Date: 2017-10-02
• Subjects: information disclosure -- risk disclosure -- forecast accuracy; Accounting -- Periodicals; Accounting -- China -- Periodicals; 657.05
• DOI: 10.1080/21697213.2017.1438288
• Languages: English
• ISSNs: 2169-7213

Abstract

Abstract: Risk information disclosed in annual reports may have positive impacts on analyst earnings forecast by improving information quality, or may have negative impacts by increasing the risk perception of analysts. In this paper, we quantify risk information disclosures in annual reports using textual analysis, and examine the impact on analyst forecast accuracy. The results at firm-level show that (1) collectively analyst forecast accuracy increases with more risk disclosures; (2) this effect is more significant in the group of non-SOEs, higher auditing quality, higher earnings quality, or better corporate governance, suggesting that good internal and external governance mechanisms provide a guarantee for the information function of risk disclosures and for analyst forecast accuracy; (3) the conclusion are robust to firm inherent risks, and they are favourable for increasing forecast accuracy when disclosing risks higher than the inherent risks. The results at analyst individual-level show that (1) individually, analyst forecast accuracy increases with more risk disclosures; (2) this effect is more significant in the group of non-star, less industrial expertise and less follow experience. It suggests that the weak heterogeneity of risk disclosures in annual reports is favourable to increase analyst forecast accuracy in China. The conclusions support the information argument, and contribute to the emerging risk disclosure literature and the literature on analyst forecasts.