本篇文章利用NYSE，AMEX與NASDAQ上市公司之資料，併同時利用CRSP與COMPUSTAT兩個資料庫來進行分析，礙於資料年限問題，我們選取自1984-2003年共643個股利初次發放的樣本公司，利用有別與以往的方法來觀察股利初次發放後5年公司風險與相關因子變化之情況，另外再運用產業與規模最相近的比較公司來觀察是否存在產業因子的干擾，而使結果產生偏誤情況的發生，故我們的結果可以避免此問題的發生。我們的結果顯示，當公司在股利初次發放以後會有總權益風險的下降，而比較公司未有顯著風險下降的情況。我們再將總權益風險分成系統風險與非系統風險，我們可以發現系統風險在股利初次發放後會有暫時性的下降，但是非系統風險確有顯著的持續性下降，也因此我們將總風險的下降歸因於非系統風險的下降。另外系統風險也可以再分為資產風險與財務風險，我們發現資產風險有顯著的下降。每股盈餘在股利初次發放後會有短暫的增加，股權周轉率在股利初次發放後在股利初次發放後也有下降的情況，而外部股東人數似乎對於股利發放與否的影響力不高。我們利用兩種方法來討論負債與股權之關係，發現兩種結果皆指出在股利初次發放以後負債比率會下降，也就是說明負債與股權存在替代的關係。 This paper examines the risk structure changes and the relation between debt financing and dividend payment. Our finding presents that the decrease in total equity risk is followed by the decrease in the residual variance, and there is also a decrease in equity beta after dividend initiation. Dividend payout is the main factor to reduce firms’ total equity risk and the most part of the decrease comes form the decrease in the residual variance. This study also identifies the influence of dividend initiation in earning per hare, turnover rate and outside shareholders. The result indicates that the sample firms only experience a significant increase in earning per share in year +1 but a decrease from year +2 to +5. We further identify the relation between debt and dividend with short-term and long-term method. We find that the firms with dividend increasing would likely choose lower debt to equity ratio, so the relation between debt and dividend is so-called substitution effect. Moreover, we implement the long-term method to justify the relation, and the result also shows that a substitution effect between debt and dividend, consistent previous result, but it is a transitory effect.
