台灣上市櫃公司『家族控制程度』變化與原因之研究

Abstract

本研究主要貢獻為拉長樣本研究期間以探討台灣上市櫃公司家族性的變化，創造家性族控股指標作為判定家族程度高低的依據，進而探討績效、最終控制者特性、公司特性與台灣上市櫃家族控股公司變動以及家族控股指標之影響。樣本取自1996年底存在上市櫃公司做篩選，參考La Porta et al.（1999）最終控制權股東之衡量方法，判定篩選出226家上市櫃公司，進而分辨企業至2016年最終控制者未變動之公司。進一步探討，家族企業、企業家族性變動(家族控股公司轉變為非家族性公司)、家族性指標3個主要的被解釋變數(虛擬變數)探討所有解釋變數與其之關聯性研究。 實證結果顯示應變數為家族控股公司時，當經理人內部化比率越大時，董事長為家族成員的可能性越高。並運用家族控股變動作為應變數，發現董事席次控制比率顯著為負，顯示當公司董事席次控制比率下降，董事長越有可能由家族成員轉變為非家族成員擔任。在被解釋變數不變之下，將解釋變數設定為前一期的變數資料，發現在最終控制股東的控制權與現金流量請求權偏離差越大及經理人內部化比率越小，董事長職務越有可能轉變由非家族成員所擔任。 在被解釋變數為家族控股指標之下，現金流量權越大則公司的家族性控股指標則越大，並發現公司控制權與現金流量權越貼近則家族控股指標則會越高。實證結果顯示，最終控制者特性變數對於家族控股企業的型態有顯著的影響。

The main contribution of this study is to study the changes of Taiwanese family firm during the study period, and to create the family index as the basis for judging the family level, and then explore the performance, the ultimate controller characteristics and the characteristics of the company for Taiwanese family changes and family indicators of the impact. Data from the end of 1996 there are 298 listed companies to do screening, refer to La Porta et al.(1999)the final control of the determination of the use of the ultimate control of the control of shareholders to determine the screening of 226 listed companies, and then distinguish the company which did not change the final controller to 2016. And use three major dependent variables (dummy variables): family firms, changes of degree of family control, family index to explore all of explanatory variables and its relevance. Our empirical results show that when dependent variable was family controlled companies, the greater the manager's internalization ratio, it will increase the probability of chairman is family member. Using the change of family controlled companies as the dependent variable, it was found that board of directors ratio was significantly negative, indicating that when board of directors was reduced, the more likely the chairman who was changed from familial to non-familial member. Under the same dependent variable the explanatory variables is set as the variables data of the previous period, it is found that the greater the deviation between the voting rights and cash flow rights and the smaller the manager's internalization ratio that the chairman is more likely to be changed to non-family member. In addition, when dependent variable is family controlled index, the greater the cash flow right will make greater family controlled index. While the smaller the deviation between the voting rights and cash flow rights, the higher the family controlled index will be. The empirical results show that the final controller characteristic variables have a significant effect on the degree of family control.