為釐清經理人異動對公司日常營運決策及績效之影響及資訊內涵，本研究透過分析國內近五年來上市櫃公司樣本後發現，大多數發生經理人異動之公司多為一次且自願性之異動，而不是經常性之活動，此為一家公司經營權輪替之正常行為。至於其他發生非經常性異動之公司則多在5年內有高於2次以上之異動行為，此和該類公司經營績效不佳及集團內部有爭權行為有很大之關係。關於經理人異動事件宣告效果方面，本研究實證結果顯示，無論是何種異動原因，市場似乎皆呈現負向反應，且以非自願性異動事件負向反應程度為大。 至於經理人異動對公司投資及融資比例之影響方面，本實證結果顯示，不同之異動原因對公司投資、融資比例之影響亦不同。以自願性異動事件來說，事件發生前之固定資產成長率呈現遞增的現象，顯示經理人異動前其投資決策普遍較同業為積極，而事件發生後卻轉而成為較同業保守，顯示新上任者可能因為經驗之不足而不敢採行太積極的策略，而其負債比例在事件發生前逐年遞減，發生後則呈逐年遞增的現象。至於非自願性異動事件，本研究發現其在事件發生前即開始採行較保守的投資策略，推論可能和此類公司基本面較差，或是經理人進行資產誤置之行為所導致，異動後新繼任者之投資轉趨積極；而其於事件發生前負債比例增加，發生後則呈現減少的現象。 至於經理人異動後公司之績效情況，本研究結果顯示無論是會計指標或是市場績效指標，自願性異動樣本之結果在事件發生後公司經營績效皆較之前為差，和市場對事件宣告之反應相符合；但以非自願性異動樣本來說，事件發生後公司經營績效之指標皆較之前為佳，顯示當初事件發生時市場之反應似乎太過悲觀，但這部分的結果可能存在有存在偏誤(survival bias)之問題。 上述的實證結果使得台灣公司經理人異動和市場宣告反應、經營決策及績效首次獲得連結，對投資人在未來從事投資時可作為參考依據。 / In order to understand the information content of CEO turnover to corporate operating decision and performance, from analyzing the listing companies between 1997-2002 in TW, this research found that the most CEO turnover is once and voluntary activity, this is normal in the daily operation. As for the other involuntary turnover activity, which often happened in the company with bad performance or takeover. As for the announcement effect about CEO turnover, the article found that no matter what reasons, the effect is negative, and more significant in involuntary turnover. About the CEO turnover influence on corporate investing and financing decision, this article shows that different turnover reason would result in different decision. Take the companies with voluntary turnover for instance, the fixed asset growth rate grew before turnover but fell afterward. I assume that’s because new CEO takes more conservative investment decision than original one. And the leverage rate in this kind of companies falls before the event but grows afterward. As for the companies with involuntary turnover, this article found that the original CEO took more conservative investment decision than new CEO, this might be because of the worse fundamental situation in this kind of companies or the asset misallocation of the original CEO. And the leverage rate of involuntary turnover grows before the event but falls afterward.

本文主要換討SOX之後宣告重編公司之經濟後果。探討如下議題:(1)信用評等是否/如何反應公司的重編資訊;(2)重編內涵與經理人異動之關聯性;(3)經理人異動、信用評等改變以及重編嚴重性三者之關聯性。 / Following the passage of the Sarbanes-Oxley Act of 2002 (SOX), the increasing occurrence of accounting restatements has drawn considerable attention concerning the financial statement quality and adverse consequences of accounting restatements from investors, regulators, auditors and business communities. The primary purpose of this research is to investigate the economic consequences of accounting restatements announced after SOX based on their relations with credit ratings and management turnover. To examine the following research issues of (1) whether and how the credit rating reacts to companies’ restatements, (2) whether restatement characteristics are associated with management turnover, and (3) what the association among management turnover, credit ratings, and restatement severity is, I gather data on 1,838 companies that restated financial statements between 1997 and 2005. In the first part of the study, my results provide empirical evidence consistent with the conjecture that higher severity restatements are more likely to be followed by subsequent unfavorable ratings. Furthermore, rating agencies using accounting-based measures to predict the probability of bankruptcy perform better in assessing ratings in the post-SOX period. Moreover, rating agencies give auditor changes a “fresh-look” after SOX. In the second part of the study, the likelihood of CEO/CFO turnover significantly increases for companies with higher restatements of severity, and a CEO is more likely to be terminated if the company credit rating following restatements is downgraded. The results show that there is no “cop a plea” effect and when restatements are prompted by companies, management turnover appears to be more concerned with the dollar amount of overstatement on income and/or restatements affecting core earnings. In addition, when executives window-dress earnings to portray a more favorable earnings picture, they are more likely to be terminated following subsequent financial restatements. Moreover, results also indicate that after SOX companies seem more likely to blame their auditors for restatements of higher severity and dismiss their auditors afterwards to maybe avoid the replacements of management.