지배주주의 존재가 기업경영에 미치는 영향: 소유구조와 임금의 관계

Abstract

This paper analyzes the role of controlling shareholders in corporate wage decisions. Even though existing theories have argued for the positive role of controlling shareholders who own substantial portions of voting shares, there have not been many theoretical or empirical papers that specifically identify or confirm their roles in corporate management. If any, existing papers analyze the relationship between managerial ownership and firm value or firm performance. However, the causal relationship is not easily testable, and the results are indeterminate. This is partly due to the endogeneity problem between ownership and firm value in addition to the mixed effects of managerial ownership on firm value. In theory, larger managerial ownership positively affects firm value by aligning the incentives of managers with those of outside shareholders, but managers would also hold larger numbers of shares if they find the shares of their companies are undervalued. Such an endogeneity issue makes it extremely difficult to test empirically the effect of managerial ownership on firm value. In addition, a larger size of managerial ownership entrenches managers, which has a negative effect on the firm value, leading to indeterminate conclusions on the issue. In this paper, we identify wage bargaining as a corporate decision process in which managers or controlling shareholders can get involved. We set up a theoretical model to predict the relationship between managerial ownership and wage decision, and empirically test the hypotheses based on the model. We use a cooperative bargaining model in which a controlling shareholder or a manager who represents the share holder bargains with employees to decide on employees’ wage. Unlike other production factors such as financial capital or intermediate materials to be out- sourced from competitive factor markets, wage provides a good candidate for confirming the role of controlling shareholders in corporate management since wage is negotiable in the bargaining process between the firm and its employees. We derive theoretical predictions on the effect of controlling ownership and the personal benefits controlling share holders gain from controlling the wage of the firm’s employees. As a result, we could empirically confirm the disciplinary role of controlling ownership in wage decision in firms. The theoretical model predicts that controlling ownership is negatively correlated with the wage of employees as long as the profitability of the firm is not too high or employees’ stock ownership is large enough. This is because the relative importance of cash flow rights of a controlling shareholder increases with his stock ownerships while the relative importance of his private benefit of control decreases. Our intuition dictates that the more shares the controlling shareholder owns, the higher wage will be incurring more cost to his share value, providing an incentive for him to lower the wage. Thus, the condition of low profitability is needed since the controlling shareholder may concede too much in a wage bargaining if he has too much to lose when the profitability of the firm is very high and the bargaining fails. The condition for the reasonable level of employees’ ownership also implies that employees also need to have something to lose if the bargaining fails. Empirical analyses confirm the theoretical predictions and show that there is a negative relationship between the two variables. We also show in the theoretical model that the size of the private benefit of control is positively correlated with employees’ wage. We conjecture that controlling shareholders concede to the employees in the wage bargaining since the private benefit can be lost if the bargaining fails. The data used in this paper for empirical analyses is obtained from the TS-2000 of KLCA, and covers the listed companies on the Korea Stock Exchange during the years from 1999 to 2002. The empirical analyses confirm the theoretical predictions and show that there is a negative relationship between controlling ownership and wage level. When we include an interaction variable between low profitability dummy variable and controlling ownership, the coefficient is negative and significant at the 1% significance level. The result confirms our conjecture that the monitoring role of controlling shareholders will be stronger when the firm does not perform well. When we divide the sample firms based on their employees’ ownerships and profitability, we again confirm our theoretical results that the monitoring role of controlling shareholders would be stronger in those companies with higher employees’ ownership and lower profitability. The over all results suggest that a professional manager with lower stock ownership would collude with his employees and allow them higher wages in order to secure his control over his firm. The paper also shows that the disciplinary role of a controlling shareholder is more prominent when the performance of the company is poorer. On the other hand, any increase in the ownership by employees or their participation in a labor union tends to have an increasing effect on their wage level. The theoretical and empirical results of the paper enhance our understanding of the role of a controlling shareholder as a monitor of corporate management. 본 논문은 Adrian and Franzoni(2005)가 개발한 학습 자본자산가격결정모형(learning CAPM)을 재고찰한다. Adrian and Franzoni(2005)의 모형은 변수오차(errors-in- variable), 생략변수(omitted-variable) 문제로 인한 모형의 내생성을 무시하였는데, 본 논문에서는 설명변수와 오차항 사이에 상관관계가 존재할 경우에 적용할 수 있는 Kim (2006)의 방법론을 대안으로 제시하여 모형 설명력의 개선 여부를 고찰한다. 본 연구의 실증분석 결과는 아래와 같다. 첫째, Hausman(1978)의 내성성 테스트에서 시장초과수익률과 오차항 사이에 뚜렷한 상관관계가 존재하므로 Kim(2006)의 방법론을 정당화한다. 둘째, 제시된 모형에서 소형-가치주는 1.62, 대형-성장주는 0.94로 시장베타의 평균이 각각 추정되어 소형-가치주에 대해 체감하는 위험이 대형-성장주에 비해 상대적으로 높았으며 이는 소형-가치주의 높은 기대수익률을 정당화한다. 셋째, 제시된 모형의 설명력이 Adrian and Franzoni(2005)의 모형에 비해 전반적으로 개선되었으며, 특히 가중가격오차(CPE)가 약 49% 감소하였다.