緊隨盈餘宣告的分析師盈餘預測–資訊意涵課題的再釐清

Abstract

本研究針對認定分析師預測報導對市場的影響力檢測的幾項可能缺失，經由實證設計調整，探討日中的分時股價資料第一位分析師盈餘預測報導兩事件對市場的影響力，特別是本研究多了時間遞減因素的設計，藉此釐清市場反應究竟係來自公司本身或分析師，以及兩者對市場的影響力有多大，可以解釋分析師盈餘預測報導本身確實是具有實質資訊意涵。 本研究貢獻有：(1)以多事件研究法釐清了在同一天發生的公司盈餘宣告事件及第一位分析師盈餘報導對市場的影響力；(2)加入了事件影響力時間遞減因素、並控制了其他分析師報導及第一位分析師其他天期報導，提昇了統計檢定力；(3)分析師盈餘預測報導受到公司盈餘宣告及當時股票市場好壞的影響，即虛假相關及內生性問題獲得處理。主要結果有：(1)在公司盈餘宣告帶給市場衝擊後，分析師盈餘預測報導確實會帶給市場另一個邊際顯著的衝擊。(2)對於訊息難解讀、財報難懂、資訊量少的公司，市場投資人更需要分析師的報導。(3)當市場關注或較仰賴分析師解讀時，分析師因應速度愈快；會計資訊解讀難度愈大時，分析師因應速度愈慢；(4)盈餘宣告時間相同時，市場關注及亟需分析師解讀的公司，分析師會優先報導。

The study extends the literature for informativeness of analyst forecasts by improving the effectiveness of test measures for the magnitude of market reactions triggered by analyst earnings forecasts revisions through adjusting the empirical designs. We reexamine the informativeness of the first analyst earnings forecast revision by examining abnormal stock market returns in every ten-minute intervals during the trading hours on the earnings announcement day, especially adding the time decaying factors. Moreover, our study both measure the magnitudes of the market reactions immediately after companies’ earnings announcements and analysts’ earnings forecasts revisions. Thereby, we could clarify whether the market reactions result from analysts revisions instead of corporate announcements. Overall, our results are consistent with notion that investors perceive analyst earnings forecasts accompanying earnings announcements as being informative. This study has the following contributions. First, we adopt multi-event settings to distinguish the market reactions triggered by analysts’ earnings forecasts revisions from those caused by companies’ earnings announcements at the same day. We resolve the spurious relationships and endogeniety problems by taking control of companies’ earnings announcements and market performance then which could impact on the analysts’ earnings forecasts revisions. Second, the statistical power is improved after adding the time decaying factors and controlling for noises: for instance, FY1 revisions of analysts other than the first one, and other revisions except FY1 of the first analyst. Our findings add to the existing literature. First, after earnings announcements take shocks on market, analyst earnings forecasts also marginally lead to market reaction. Second, the investors appear to reply more on analyst forecasts when their companies’ financial statements and information are more difficult to interpret or they typically have smaller amount of information available to the market. Third, when companies belong to those market interested or more dependent on analysts, analysts react quickly to earnings announcements. In contrast, for the firms whose accounting information is difficult to interpret, analysts react slowly to the announcements. Fourth, among companies announcing earnings at the same time, when companies are with greater assets, with more analysts following them, with greater earnings volatility, with greater discretionary accruals or with greater temporary in earnings and more dependent on analysts, analysts report them first.