Do Social Ties between Audit Committee Members and CEO Affect Earnings Management?

審計委員會之設立是目前公司治理的焦點。而為了確保審計委員會的有效性，我國對於委員們的獨立性更是嚴格要求。但即使委員們合乎法律規範與企業管理階層間無親屬或財務利益關係，仍可能經由其他的社群連結與管理階層有連繫。本研究將焦點放於親屬以及財務以外之社群連結，期望探討委員與執行長間之社群連結是否會影響到財務報導品質。 本研究以手動搜尋CEO與審計委員會的經歷資料，分析兩者是否具社群連結，再將社群連接分成教育連結與職場連結，並以異常應計絕對值作為盈餘管理之代理變數，檢視社群連結與財報品質間的關連。本研究發現委員與CEO間之社群連結與異常應計絕對值雖然方向與預期相符為正向關係但沒有顯著影響，將社群連結細分成教育及職場連結後，教育連結對於異常應計絕對值有正向且顯著影響，職場連結雖與預期方向相同，但並不顯著；當同時具有職場及教育連結，對異常應計絕對值有正向且顯著影響。To make sure that the audit committee can provide sufficient oversight over financial reporting system, the regulators require that all the audit committee members should be independent. Although all the audit committee members appear to be independent, still they can be connected to CEO through their social networks. The purpose of this thesis is to explore whether the social ties between the audit committee members and CEO will affect the quality of financial reporting. Using hand-collected data from 2011-2013, I examine the relationship between social ties and abnormal accruals, proxy of earnings management. I didn’t find that these social ties will increase abnormal accruals. But when I further subdivide these social ties into education ties and employment ties, I find a significant positive relation between education ties and abnormal accrual, which suggests that the audit committee member who has education ties between CEO is more likely to trust CEO and thus may weaken their oversight over financial reporting system. When both education and employment ties exist between CEO and audit committee members, the results still hold