Determinants of Earnings Forecast Error, Earnings Forecast Revision and Earnings Forecast Accuracy

Earnings forecasts are ubiquitous in today's financial markets. They are essential indicators of future firm performance and a starting point for firm valuation. Investors heavily rely on forecasts when making investment decisions. The extremely inaccurate and overoptimistic forecasts during the most recent financial crisis have raised serious doubts regarding the reliability of such forecasts. In addition, the academic literature attributes systematic biases in forecasts to analyst incentives other than providing truthful earnings forecasts. These incentives result from analysts serving market participants with different needs.

Today's capital markets are characterized by an active and close network of interpersonal relationships. Market participants interacting in this network pursue a number of different objectives leading to potential conflicts of interest. The interactions and interdependencies as well as the differing objectives of the different groups of market participants also impact the information content and quality of earnings forecasts.

Over the last four decades, numerous empirical studies have provided evidence that analyst forecasts are optimistically biased at the beginning of a fiscal year. However, analyst incentives to bias the forecast might change over the course of the year because the initial incentive to bias a forecast optimistically will be dominated by other incentives such as to be the most accurate forecaster at the end of the year. To improve forecast accuracy, analysts will therefore revise their forecasts for upcoming news, to account for the impact that news has on reported earnings and thus the initial forecast error. Moreover, analysts need to correct their forecasts, so as to compensate for the initial optimism bias. Since forecast revision is equal to the difference between two forecast errors, I argue that revisions are driven by the same determinants as forecast errors. In addition to the intuitive impact of news on revisions, I argue that a second major driver of revisions is the change in analyst incentives to systematically bias their earnings estimates. In this chapter, I draw on the literature of forecast error to derive a revision model that provides a new understanding of the drivers of forecast revisions.

Accounting literature offers several explanations for the variation in analysts' forecast precision (e.g. differences in forecast ability, experience or effort). In this chapter, I identify two new determinants of forecast accuracy, namely general forecast effort and the provision of investment advice. The identification of these determinants significantly contributes to existing literature. The results show that investors can use aggregate forecasts based only on those issued in conjunction with investment advice to formulate more accurate estimates on a firm's future earnings performance. Moreover, emphasizing the importance of general forecast characteristics such as general forecast effort in addition to firm-specific characteristics enhances the understanding of the major determinants of forecast accuracy.

This thesis deals with the determinants of forecast errors, revisions and accuracy. More specifically, the thesis answers the following questions: 1) How do analyst incentives lead to forecast errors? 2) How do changes in analyst incentives lead to forecast revisions?, and 3) What factors drive differences in forecast accuracy?