在零和競爭環境下經環境調整與統計噪音調整後之證券商管理效率

Environment and Statistical Noise-adjusted Efficiency under Zero-sum Gains with an Application to Securities Firms

研究資料包絡分析的既存文獻大都沒有同時考慮有限制產出、環境變數與統計誤差對效率值衡量之影響。本篇研究的貢獻在於同時考量在上述限制與影響下，以發展在零和競爭環境下，將環境因素與統計誤差項排除後之效率衡量模型。本研究採用產出調整跨期追蹤模型，以台灣地區金控公司成立時期為分界點，資料涵蓋金控公司成立前之2001年之139家券商至2003年之110家，與後金控之2004年（103家券商）至2006家（92家券商）。實證結果顯示：外商投資之證券子公司其效率值優於本土證券公司，專業經紀商雖然被排除在金控的證券子公司資格中，但其效率值卻優於本土綜合券商。金控法的施行明顯限制了專業經紀商的生存環境，造成專業經紀商的家數持續減少。本論文的實證結果顯示台灣的證券商發展環境直接被法令所影響。未來研究可延伸此篇論文以股票市占率極大化為首要單一目標之設定至包含其他之目標，並使用共同邊界分析法，以考量不同類型證券商間之差異。

Most existing studies using data envelopment analysis (DEA) do not take into account the restriction of 100% market share and the effects of statistical noise and environment on efficiency scores.This paper simultaneously combines and extends Lins et al. (2003) and Fried et al. (2002) to compute environmental and statistical noise-adjusted efficiency under zero-sum gains (ZSG) via output slack adjustments.A panel dataset of securities firms in Taiwan from pre-FHC periods during 2001 (139 firms) - 2003 (110 firms) to post-FHC during 2004 (103 firms) - 2006 (92 firms) is used for empirical study.Empirical findings confirm that a continuous stream of foreign-affiliated securities firms offer better practice than domestic ones.Small and specialized domestic brokerage firms are relatively efficient and hence should not be legally prohibited from being subsidiaries of financial holding companies in Taiwan.This paper specifically shows that regulation directly influences the securities industry in a small open economy.The future research may extend the single stock market share maximization in this paper to include other outputs as well as use the mefafrontier analysis to account for differences in security firm types.