台灣新上市櫃銀行股利政策之實證研究

謝憲昭

《華人經濟研究》 5卷2期 (2007/09) Pp. 120-153

https://doi.org/10.30065/JCER.200709.0008

摘要

本文以事件研究法及Tobit迴歸模型研究台灣銀行業的股利政策，主要探討新上市櫃銀行的股利發放情形、股利宣告對股價的影響與決定銀行股利政策的因素，並討論銀行業與非銀行業之間的股利政策是否有不同的影響層面。新上市櫃銀行有超過85%的樣本最終都會發放股利，且在初次公開發行(Initial Public offerings, IPO)後一年內立即進行首次股利支付者高達59%。其中以股票股利和同時發放現金股利及股票股利為股利政策者居多，此結果顯示股利政策對銀行業愈趨普遍，且在IPO後的首次股利支付行為愈顯重要性。
事件研究法的實證結果顯示新上市櫃銀行在股東大會宣告股利發放之事件，對於銀行的股價有正面的價格效果，顯示與新上市公司的股利宣告效果相同，銀行股利宣告的確會傳達出某些正面有關銀行品質與未來前景的訊息。進一步的實證結果顯示，新上市櫃銀行的首次股利發放宣告所隱含的訊息內容可以吸引投資人注意，且市場對於銀行首次股利宣告的反應較非首次股利宣告者大。研究結果亦顯示，不同於市場投資人較偏好國內新上市公司發放股票股利，台灣投資人較偏好新上市櫃銀行同時發放現金與股票股利，且新上市櫃銀行只發放現金股利將帶來較大的負面股價效果。因此，銀行在IPO後，可立即使用平衡股利政策的宣告來達到正向訊息的傳遞效果，以與其他IPO銀行有所區別。
不論從現金股利、股票股利和總股利來看，迴歸分析結果顯示銀行資產報酬率均顯著正向影響銀行的股利支付，表示獲力能力佳的銀行，其配發的股利金額確實高於經營差的銀行所配發的股利金額。這個實證結果支持信號發射觀點，表示銀行管理當局可透過股利的發放來傳遞預期未來盈餘會改善之訊息。此外，實證結果亦顯示新上市櫃銀行在未來有較多成長機會時會發放較少的股利，以保留較多的盈餘以因應銀行未來的成長。

關鍵字

初次公開發行 ； 股利政策 ； 首次股利支付 ； 事件研究法 ； Tobit模型

並列摘要

Most of the empirical research has focused on established firms instead of initial public offerings (IPO) as well as on industrial firms instead of banks. The objective of this paper is to investigate the trend, the stock valuation effects as well as the factors of the dividend policy of initial public offerings of banks in Taiwan over the period from 1992 to 2005. For the IPO banks, the ratio of the dividend initiations in the first year after the IPO is 59%. Prior studies indicate that dividend initiation provides management's inside information about the quality and future earnings of the firms to shareholders. This paper thus explores the valuation effects of the dividend initiations of the IPO banks.
The major findings can be summarized as follows: Firstly, the results show that the announcement of dividend policies generates significantly positive abnormal returns. Moreover, the dividend initiations on average have lager accumulative abnormal announcement returns than subsequent dividend payments. Thus, in an environment with information asymmetries dividend initiations are an important signal to convey the quality of IPO banks. Secondly, the results show that the both cash and stock dividend initiations have the significantly positive valuation effects. On contrary, the cash dividend initiations have the significantly negative valuation effects. The results indicate that the both cash and stock dividend initiation of the IPO banks is an important characteristic that separates the other IPO banks or established banks. Finally, the regression results provide some evidences that a bank's profitability, growth opportunities, stock insider ownership and bank risk have a positive or negative impact to the total, cash and stock dividend yields. These results are consistent with the signaling hypothesis and the findings in the existing studies.

並列關鍵字

initial Public Offerings ； dividend policy ； dividend initiations ； event study ； Tobit model