DescriptionThis dissertation consists of two essays on financial analysts’ stock recommendations. The first essay examines the relationship between corporate social responsibility (CSR) report and the value of financial analysts’ stock recommendation revisions. We find that the value of stock recommendations for socially responsible companies is lower than non-socially responsible companies. Also, we show that there is an inverse relation between the level of information on CSR strengths and concerns and the value of financial analysts’ stock recommendations. Furthermore, when we focus on the sensitivity of change of CSR ratings, our result indicates that the value of stock recommendations is negatively associated with a firm’s improvement on CSR score. As a firm experiences more change in social responsibility strengths and concerns, the value of analysts’ stock recommendations decreases. Our results imply that the value of recommendation is a function of firms’ CSR ratings and the amount of information on CSR strengths and concerns. In the second essay, we hypothesize that a perception of higher ability is implicit when an analyst makes a bold recommendation, and that this self-assessment is more likely to be correct when there are few analysts covering the firm. Consistent with our hypothesis, we find that it is highly profitable to trade based on bold recommendations for firms with low analyst coverage (risk-adjusted return of 30% per year), but not bold recommendations for high coverage firms. Herding recommendations, whether for firms with low or high coverage, are not profitable. The profit from this trading strategy is related to the news released during earnings announcements.