本研究的主要目的係評估對於台灣中小企業與金融機構這兩個獨特的樣本之董監事最低持股成數規定是否允當，並檢測代理理論與管家理論對於內部人持股與公司績效間的關係之解釋性，以及利益收斂效果與管理據守效果是否同時存在。以1999年到2006年間在證券交易所及櫃檯買賣中心有掛牌的中小企業為第一樣本，一共取得62家中小企業擁有八年完整資料的496筆觀察值的最終樣本；以2002年到2006年間在證券交易所及櫃檯買賣中心有掛牌的金融機構為第二樣本，一共取得41家中小企業擁有五年完整資料的205筆觀察值的最終樣本。採用並改良Hansen (1999)所提出來的門檻廻歸模式的估計與檢測的方法，改進以往文獻在模式設定與估計方法上的缺失，此為本文最重要的貢獻之ㄧ。 實證結果發現，門檻廻歸方法可以得到比傳統研究方法更高適合度的模型，避免過去常用的多次方模型或主觀的分段模型所造成的偏誤。其次，產業效果存在，不同產業樣本有著大同小異的實證結果。對台灣掛牌交易的中小企業及金融機構而言，其股權與績效之關係是非線性的，呈現在內部人持股水準為5.4%處(中小企業)及3.94%處(金融機構)轉折的先負後正(中小企業)及顯著正到不顯著正(金融機構)的實證結果。顯示利益收斂效果與管理據守效果的確同時存在，且主管機關對於中小企業及金融機構的董監事最低持股成數規定雖有存在之必要，但目前規定的下限水準過高。政府主管機關應該調降董監事最低持股成數的水準，進而提高董事會的獨立性，讓公司邁向較佳的治理環境。而其他新興市場國家也可仿效台灣經驗，設定適合的董監最低持股成數規定。再者，特別是中小企業，可對其內部人採取股票選擇權或績效配股方式以達到顯著的激勵效果，降低權益代理問題。最後，外部投資人與外國投資者也可藉由內部人持股數所釋放的訊息，降低資訊不對稱的不利影響。 此外，本研究發現，管家理論在台灣的環境中頗為適用；內部人往往自詡為忠誠的管家，而非只謀私利的據守者。因此，建議未來的相關研究應對於管家理論在公司治理領域的影響著墨更多，相信更能研擬出較好的公司策略。 This study aims to evaluate the suitability of Taiwan’s minimum shareholding requirements for small- and medium-sized enterprises (SMEs) and banking firms, and to examine whether stewardship theory and both incentive and entrenchment effects can characterize the ownership-performance relationship simultaneously. We apply new empirical methods to both balanced panel datasets including 496 firm-level observations of 62 Taiwanese SMEs publicly traded from 1999-2006 and 205 firm-level observations of 41 Taiwanese banking firms publicly traded from 2002-2006. We use and refine the threshold regression methods proposed by Hansen (1999) to avoid the distortion of specifications and estimations adopted in past research. This is one of important contributions in this study. The results firstly show that our methods obtain a better specification with higher goodness of fit that prevents spurious relations obtained from traditional methods. Secondly, industry effects do exist. For SMEs sample, there is a nonlinear ownership-performance relationship, first negative then positive, with a single threshold value of insider ownership (5.4%). For banking firms sample, there is a nonlinear ownership-performance relationship, steeply increasing then insignificantly positive, with a single threshold value of insider ownership (3.94%). This indicates that both convergence-of-interest and entrenchment effects combine and that the current regulated minimum level of insider shareholdings for public SMEs and banking firms may be too high. We suggest that Taiwanese authorities should lessen ownership obligations thereby leading to improved corporate governance by increasing the independence of the boards of directors. Thirdly, other emerging markets can learn from the Taiwan experience and then formulate their own regulations. Fourth, for publicly traded SMEs, firms could offer compensation to insiders such as stock options and performance shares. Finally, asymmetric information problems often experienced by outside stockholders and foreign investors can be alleviated by these signals of insider ownership. In addition, our evidence supports that the stewardship theory does work in Taiwan. Insiders behave as loyal stewards rather than motivated by desire for entrenchment. Our empirical evidence suggests an avenue of future research, specifically to form better strategies for those involved in corporate governance.
