標題: 上市公司研發支出財富效果與負債比率關係之研究
An Empirical Test of the Relationship between Shareholder Wealth Effects of R&D Announcements and Debt Ratio in Public Companies
作者: 胡怡萍
Yi-Ping Hu
王克陸
Dr. Keh-Luh Wang
經營管理研究所
關鍵字: 負債監督假說;股東財富效果;debt-monitoring hypothesis;shareholder wealth effect
公開日期: 1999
摘要: 大部分實證研究均肯定研發支出增加的宣告顯著引起股價上漲(異常報酬)。本篇研究是從負債監督假說出發,探討台灣上市公司研發宣告股東財富效果與負債比率之關係。高負債比例控制了管理者的自主能力(managerial discretion),降低了管理者可獲之自由現金流量,且迫使管理者工作更加努力、較少補貼性消費、做較好投資決策,要不然會有破產風險。所以,以代理成本為基礎的資本結構模型隱含了市場可藉由觀察宣告廠商的負債比率,來對難以觀察的研發宣告品質作推論。因此,宣告研發支出增加的企業,其異常報酬應正相關於負債比率。研究設計採二階段(1)縱斷面:以事件研究法探討研發宣告效果(2)橫斷面回歸分析:控制管理者持股、企業規模、研發密度後,探討負債比率是否會影響研發宣告效果。實證結果發現:(1)整體上市公司或電子業在研發宣告日、宣告日前一日累積至宣告日皆有顯著正異常報酬,即研發宣告的確為股東帶來異常財富效果。(2) 整體上市公司或電子業宣告效果與負債比率顯著呈現二次式凹口向下關係,即負債比率較低時,負債監督假說成立,高負債比具有監督效果,研發支出增加宣告的股價反應可端視股票市場之前關於此宣告廠商的資本結構訊息;但當負債比率達到一定比時(本實證研究結果為33﹪),負債監督假說反不成立。
Based on the debt-monitoring hypothesis, we would like to examine the relationship between shareholder wealth effect of R&D announcements and debt ratio for the public companies in Taiwan. Two steps research designs are used, the first is serial analysis, which uses event-study method to measure abnormal returns to announcing firms, the latter is cross-sectional analysis, which uses ordinary-least-squares regression while controlling for several potentially influential variables to determine whether a relation exists between the abnormal return at announcement and the announcing firm’s capital structure. The empirical results are organized as two parts. The first is , on average, there is statistically significant positive abnormal return during period (0,1) and announcement date whether in the full sample or electronic sample. The other suggests a quadratic relation between the debt ratio and the R&D induced abnormal stock returns. That means when the debt ratio is low, the empirical result provides support to the debt-monitoring hypothesis.
The stock price reaction to plans to increase R&D expenditures could depends on the stock market’s prior information about the capital structure of the announcing firms. But when the debt ratio is high (our research find is 33%), we don’t have enough evidence to support the debt-monitoring effect.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT880457073
http://hdl.handle.net/11536/66020
Appears in Collections:Thesis