庫藏股宣告與後續績效之關係

Abstract

本研究旨在探討公開市場股份買回宣告與後續長期營業績效、股票報酬之關係。本文主要目的為檢測庫藏股決策的資訊內涵，進而探求公司從事庫藏股活動時可茲預期之發展。文獻上理論皆預期市場對股份買回宣告會有正向反應，然而，僅就觀察初始、短期市場證據，實難檢視庫藏股宣告的資訊內涵。因此，本文擬追蹤探討庫藏股宣告後的長期績效，以瞭解公司啟動買回股份決策背後的經濟性動機。 文獻上有多種假說解釋公司宣告買回庫藏股之動機，其中以資訊信號假說與自由現金流量假說二者獲得較多實證支持。為進一步分辨此二種競爭性假說，本文不僅探討公司庫藏股宣告後的長期營業績效與投資支出之變動，並探討庫藏股宣告後的長期異常股票報酬。本文假設若庫藏股宣告後買回公司績效有所改善，且績效的改善係源自成長機會帶動的投資支出增加，則信號假說之預期獲得實證支持。反之，若公司於庫藏股宣告後縮減其投資支出與現金準備，肇因於公司未來成長機會匱乏，則自由現金流量假說獲得實證支持。 本研究以宣告買回庫藏股之台灣上市櫃公司為樣本，實證發現，庫藏股宣告的前後年度，相較於控制公司，買回公司營業績效皆呈現下跌，且買回宣告的後續三年，買回公司之營業績效並未改善，實證結果與信號假說之預期不符，本研究證據未能支持庫藏股宣告係傳達公司未來獲利改善之訊息。相反地，本研究提供部分證據顯示公司買回庫藏股乃是缺乏投資機會，因而藉由買回股份將多餘的現金流量發還給股東，然而，實證未發現買回公司於庫藏股宣告後顯著減少現金準備。因此，本文實證僅提供有限證據支持自由現金流量假說。 為檢視庫藏股宣告後的長期股票報酬，本文採用三種衡量方法：累積異常報酬、買進持有異常報酬及Fama-French 三因子模式。實證顯示，公司於庫藏股買回宣告期間會有正的異常報酬，然而，買回宣告的後續三年，買回公司並未呈現顯著正的異常報酬，顯示當公司從事庫藏股活動時，並不意謂其有較佳之長期異常績效，就長期市場績效而言，本研究實證證據亦不符合信號假說之預期。 整體而言，相較於未買回公司，買回公司於庫藏股宣告的後續長期年度呈現較差的營業績效與股價報酬，買回公司並未展現較佳的前景，顯示未能支持以資訊信號假說解釋公司宣告買回庫藏股之動機。

In this study, we examine the long-term operating performance and abnormal stock returns subsequent to open market share repurchase announcements. The major purpose of this essay is to examine the information content of repurchase decisions and to find out what can be predicted when a firm is involved in share repurchase activities. Given that the theories all predict the same stock reaction to repurchase announcements, obviously, it is difficult to test the information content of repurchases by simply looking at the initial, short-term evidence. Thus, by examining the long-term performance following share repurchases, we may better understand the economic motivations behind the decision to launch an open market repurchase program. The two most widely accepted explanations of the motivation behind share repurchases are the information signaling hypothesis and the free cash hypothesis. In order to differentiate between the two competing hypotheses, the essay not only investigates long-term changes in operating performance and investment expenditures following share repurchases but also investigates long-term abnormal stock returns in the post-repurchase period. We hypothesize that if repurchasing firms experience an improvement in performance and the improve performance is a result of an increase in investment expenditures due to growth opportunities after the announcement, then the prediction of the information signaling hypothesis has empirical support. On the other hand, if repurchasing firms reduce their investment expenditures and cash reserves after the repurchase announcements because these firms do not have many investment opportunities, then the free cash flow has empirical support. Using a sample of open market share repurchase announcements in Taiwan, we observe a decline in operating performance for repurchasing firms before and after the announcement. Contrary to the conventional wisdom, there is no evidence supporting the claim that repurchasing firms are signaling better prospects. In contrast, this study provides partial evidence supporting the claim that firms with few investment opportunities will distribute the excess cash flows by share repurchases. However, we find no evidence that firms reduce their cash reserves after the share repurchase announcements. Overall, this study provides limited evidence to support for Jensen’s (1986) free cash flow hypothesis. To gauge long-run stock returns in the post-repurchase period, we use three different approaches, namely, the CAR approach, the buy-and-hold approach, and the Fama- French three-factor model. We find some evidence of positive abnormal performance surrounding the announcement period. However, we do not find any evidence of abnormal returns for repurchasing firms in the three-year post-announcement period essentially. The repurchasing firms do not exhibit strong superior abnormal performance over long horizons when they make open market share repurchase. Our main empirical finding seems to be inconsistent with the most important predictions of signaling hypothesis. Collectively, the results reveal that repurchasing firms experience the comparatively lower post-announcement operating performance and stock returns. Taken together, our findings do not support the implication of the signaling hypothesis.