This dissertation comprises two essays on earnings forecasting accuracy. Chapter 2 focuses on how management forecasting accuracy is affected by managers behavioral biases over time and Chapter 3 addresses how analyst portfolio design choices affect cross-sectional differences in analyst forecasting accuracy.In particular, Chapter 2 examines how CEOs overconfidence and their exhibited self-serving attribution biases affect how they adjust their future earnings forecasts when they receive feedback concerning their prior forecasts. I find that overconfident CEOs respond to feedback by improving their future forecasting accuracy, but they do so more slowly than their less confident peers. I also find that overconfident CEOs learn to improve their future forecasting accuracy only when feedback is less ambiguous in the form of forecasting errors. In contrast, managers who are less confident respond to both less ambiguous forecasting errors and more ambiguous market feedback concerning their prior forecasts.Chapter 3 examines analysts supply chain coverage portfolio design and their forecasting accuracy. I define the relation between a firm and one of the firms major customer firms as a supply chain relation. Further, I define an analyst who issues forecasts for both the firm and one or more of the firms major customers in the same year as a supply chain analyst. I classify all firms followed by the supply chain analyst into one of the three categories in a given year: focal firms, i.e., firms for which the analyst also covers one or more of the firms major customers, major customers of a focal firm for the analyst, and other firms which include all remaining firms in the analysts portfolio.I find that analysts who follow both a focal firm and one or more of the firms major customers issue significantly more accurate earnings forecasts for both the focal firm and the firms major customers than the same analysts issue for other firms in the analyst portfolio. I also find that these analysts are more accurate in their forecasts for the focal firm and the firms major customers than other non-supply chain analysts following the same firm, but not the firms supply chain. I show that the superior forecasting accuracy for supply chain analysts for both the focal firm and the firms major customers is achieved at the cost of reduced forecasting accuracy for other firms in the analyst portfolio. In explaining the relative importance of forecasting accuracy for the firms in the analyst portfolio, I find that focal firms and the firms major customers are more likely to generate more profitable trading commissions and operate in an industry segment that has a greater number of other peer firms than other firms in the same analyst portfolio. This evidence is consistent with earlier studies and suggests when analysts have stronger incentives to generate trading commissions from the stock of a firm, the analysts are more likely to spend effort to acquire more precise information about the firm (Hayes, 1998). When covering the firms supply chains helps analysts produce more precise information about the firm, analyst forecasting accuracy improves accordingly.
