【作者】 于静；

【导师】 陈工孟；

【作者基本信息】 上海交通大学 ， 金融学， 2010， 博士

【摘要】 股权分置是困扰我国证券市场发展的重大制度问题,它破坏了上市公司利益机制一致性的基础,导致流通股股东和非流通股股东的利益发生冲突。股权分置改革的完成使大小股东的利益趋于一致,大股东的利益取向理论上回归于公司价值本身,然而股权分置改革达到这种成效了吗?这是学术界和实务界普遍关心的问题。目前学术界对于股改成效方面的研究还不多,也不够深入。本文的落脚点放在股权分置改革对上市公司利益输送和公司绩效的影响上。在回顾了中国股票市场的历史进程后,并对利益输送、公司绩效和股权分置改革的相关文献进行梳理后,分别建立了股权分置改革前和股权分置改革后的利益输送和公司绩效模型,分析了影响上市公司利益输送和公司绩效的关键因素,采用规范和实证相结合的研究方法,从理论和经验两个方面分析和检验了股权分置改革对利益输送和公司绩效的影响,并对后股改时代的公司治理提出了一些意见和建议。借鉴LLSV(La Porta, Lopez-de-Silanes, Shleifer, Vishny)模型的基础上,针对我国证券市场的实际情况,结合股权分置改革前后控股股东的特征,分别建立了利益输送和公司绩效模型。建模后发现股权结构和治理结构对利益输送和公司绩效的影响方式没有发生变化。即,现金流权越大,利益输送越小,公司绩效越大;控制权越大,利益输送越大,公司绩效越差;两权分离度越大,利益输送越大,公司绩效越低;公司治理越好,利益输送越少,公司绩效越好。虽然影响方式与股改前没有差别,但是影响程度都有所增强。股权分置改革后,现金流权和公司治理水平对利益输送和公司绩效的影响都呈现正向增强,控制权和两权分离度对利益输送和公司绩效的影响都呈现反向增强。即对于相同的现金流权、控制权、两权分离度和公司治理水平,股权分置改革后,利益输送程度显著降低,公司绩效显著提高。通过模型分析我们认为,股权分置改革改善了大股东利益输送行为的程度,提高了上市公司的业绩。股改前后利益输送和公司绩效模型的建立,为后文股权分置改革对利益输送和公司绩效影响的实证研究打下了很好的理论基础。本文利用股利变量,首先证实股利的利益输送假说,再检验股权分置改革对上市公司利益输送的影响。笔者利用2003-2008年上市公司样本,证实了股利的利益输送假说。在中国,大股东凭借其超强控制力大量掠夺上市公司财富,而不顾上市公司的发展和投资机遇,尤其是第二大股东往往是第一大股东的合谋者,协助第一大股东侵占小股东的利益。第一大股东和第二大股东的股权性质相同注定其行为方式一致,支持现金股利反对股票股利,因为发放现金股利对其极为有利,股权分置改革前大股东的股利收益率高出小股东好多倍,是对小股东赤裸裸的掠夺,而股票股利与其原持有的股票属性相同流通仍然受到限制,自然会受到大股东的排斥。政府作为大部分上市公司的实际控制者,急需资金进行经济建设,渴望上市公司大额派现,因此对上市公司的利益输送起到了推波助澜的作用。股权分置改革的完成,转变了大股东的股权性质,使其由原来的非流通股转变为限售股、流通股,这一改革具有革命性的意义,从根本上改变了上市公司的治理结构,使大股东和小股东的利益趋于一致,大股东的利益取向回归于公司价值本身,因此股权分置改革降低了上市公司的利益输送行为。根据分样本的回归结果,股权分置改革对不同的样本影响程度不同,对国家股的样本、低股权制衡的样本和商业样本影响更显著。选用主营业务利润资产收益率为公司业绩的代理指标,利用2003-2008年中国上市公司样本全面检验了股权分置改革、大股东行为等因素对公司绩效的影响。结果发现,控制权性质为国有股和法人股时,股改后公司业绩显著增加,从两权分离度角度来看,低两权分离度的样本,股改之后公司业绩显著上升,从行业角度来看,隶属于公用事业和工业的样本股改之后业绩显著增加。通过回归分析我们发现,股权分置改革提高了上市公司的经营业绩,无论是对全样本进行回归分析,还是对分样本进行回归检验。股权分置改革对国家股样本、两权分离度低的样本和工业样本影响更大。股权分置改革之所以提高了上市公司的经营业绩原因在于改善了公司治理效率。大股东行为对我国上市公司的影响总体来说是正向的,主要是利益协同效应在我国上市公司中起了主要作用。当我们把大股东变量和股权分置改革变量进行交叉研究时发现,随着股权分置改革的完成,对于同样的控制权、所有权、两权分离度、第一大股东持股比例、第二大股东持股比例、第一大股东与第二大股东持股比例差,股改后公司业绩都显著增加,只是当控制权性质为国家时或者低两权分离度的样本,对于同样的第二大股东持股比例,股改后公司业绩出现了下降。可见,我们的研究至少有以下发现:股权分置改革提高了公司治理效率,使大小股东的利益趋于一致,大股东的利益取向回归于公司价值本身。股改后,大股东利益输送行为降低,公司绩效提高,当股权分置改革变量与大股东变量进行交叉时,对于相同的大股东变量,股改后公司业绩更好。本文的创新之处有:(1)构建了符合我国实际情况的利益输送和公司绩效模型。LLSV模型是一个普遍通用的利益输送和公司绩效模型。针对我国证券市场的实际情况,分别建立了股权分置改革前和股权分置改革后上市公司利益输送和公司绩效模型,并对两个模型进行了对比。研究发现,股权结构和治理结构对利益输送和公司绩效的影响方式没有发生变化,但是影响程度都有所增强。也就是说,本文首先从理论角度验证了股权分置改革对利益输送和公司绩效的影响。(2)首次运用大样本验证了股权分置改革对利益输送行为的影响。利益输送行为是当今公司金融领域度量的难点。用资金占用和关联交易来度量,我们很难区分什么是正常交易什么是利益输送,而且从2006年开始国家规范和整治了上市公司的资金占用和关联交易行为。笔者利用用友软件的案例和实证研究的方法验证了红利的利益输送假说,并以此为手段检验了股权分置改革对利益输送行为的影响,同时上市公司发放红利并不是出于单一的目的,因此还考虑了自由现金流假说、结果模型、替代模型、股利的生命周期等视角,并对其进行了控制。为了更深入的检验股权分置改革对利益输送行为的影响,还从控制权性质、股权制衡和行业角度进行了分样本研究,并验证股权分置改革降低上市公司利益输送行为的真正原因。因此,本文关于股权分置改革对利益输送行为影响的检验深入而细致,并具有开创性。(3)首次运用大样本检验了股权分置改革对公司绩效的影响。前面学者对于股权分置改革对上市公司绩效影响的研究主要存在如下几个问题:检验方法过去简单,仅仅对股改前后两年或两个季度进行对比就得到股改是否改善上市公司绩效的结论;实证研究的样本太少,不足以对所研究的问题进行支撑;研究过于粗略,不够细致深入。针对上述问题,本文的研究样本囊括了所有的上市公司,通过对比分析、回归分析和分样本回归分析全面而深入地检验了股权分置改革对公司绩效的影响。同时还研究了金字塔控制结构对公司绩效的影响,并将股权分置改革变量和大股东变量进行交叉,检验股权分置改革对大股东行为和公司绩效的影响。更多还原

【Abstract】 Split share structure is a major institutional problem in China’s securities market development. It undermines the consistent basis of the interests of listed companies and leads to the conflict of tradable shares shareholders and non-tradable shares shareholders. The completion of non-tradable share reform makes the interests of large and small converge. The interests of major shareholders return to the company’s value-oriented theoretically. However, has the non-tradable share reform achieved the goal? The academic and practical concerns the matter. At present, the research on the effectiveness of non-tradable share reform is not enough and thorough. This paper concentrates on the impact of non-tradable share reform on tunneling and corporate performance. After reviewing the historical process of China’s stock market and summarizing the relevant literature of tunneling, corporate performance and non-tradable share reform, we established the models of tunneling and corporate performance before and after non-tradable share reform and analyzed the key factors in tunneling and corporate performance. Using a combination of normative and empirical research methods, we test the impact of non-tradable share reform on tunneling and corporate performance from theory and experience. Finally, we give some suggestions on corporate governance of post-reform era.Based on LLSV(La Porta, Lopez-de-Silanes, Shleifer, Vishny) model, we established the tunneling and corporate performance models according to the actual situation of China’s securities market and the features of controlling shareholder before and after non-tradable share reform. We find that the effect of share structure and governance structure on tunneling and corporate performance do not change in manner. That is, the greater the cash flow right, the smaller the tunneling, the greater the firm performance; the larger the control right, the larger the tunneling, the worse the firm performance; the larger the separation of ownership right and control right, the larger the tunneling, the better the firm performance; the better the corporate governance, the less the tunneling, the better the firm performance. Although there is no difference in manner, but strengthen to some extent. After non-tradable share reform, the impact of cash flow rights and corporate governance on tunneling and corporate performance are positively enhanced; the effectiveness of control right and separation of control right and ownership right on tunneling and corporate performance are inversely enhanced. For the same cash flow right, control right, separation of control right and ownership right and corporate governance, tunneling reduced and corporate performance improved after non-tradable share reform. Through model analysis, we believe that non-tradable share reform can reduce tunneling and increase corporate performance. The construction of model of tunneling and corporate performance before and after non-tradable share reform laid a good theory foundation for research on the effect of non-tradable share reform on tunneling and corporate performance.We use dividend variable to confirm tunneling, and then test the effect of non-tradable share reform on tunneling. Using 2003-2008 data of listed companies, we confirm tunneling of dividends. In China, large shareholder plunder the wealth of companies by superior control right, regardless of the development of listed companies and investment opportunities. Especially, the second-largest shareholder is the accomplice of the largest shareholder, assisting in expropriation of minority shareholders of the largest shareholder. If the shares nature of the largest shareholder and the second-largest shareholder are the same, they are destined consistency in their behavior. They will support cash dividends against the stock dividend, because cash dividend is in its favor. Cash dividend yield is many times higher than the minority shareholders’, an expropriation of minority shareholder. The attribute of stock dividend is the same as its original’s. Circulation right remains restricted. The stock dividend is naturally rejected by large shareholders. Government, as the actual controller of most listed companies, were in urgent need of funds for economic development, so they hope cash dividends from listed companies. They play an important role in fueling tunneling. The completion of non-tradable share reform changes the equity nature of large shareholders, which makes them from non-tradable share to tradable share and limited share. This reform is revolutionary, which changed corporate governance structure and make the interests of large and small shareholders converge. So the interests of large shareholders return to company. Non-tradable share reform reduces tunneling. According to sub-sample regression results, non-tradable share reform works differently, which influences national share sample, low equity balance sample and commercial sample more.Selecting the income from principal operations return on assets to evaluate corporate performance, we do a comprehensive test of the effect of non-tradable share reform and large shareholders’behavior on corporate performance. We find that corporate performance increase after non-tradable share reform, when the nature of control right are state-owned shares and legal person shares. Low separation of ownership and control right sample increases firm performance after non-tradable share reform. The public utilities and industrial samples improve corporate performance after non-tradable share reform. Through regression analysis, we find that non-tradable share reform increase corporate performance, whether in the whole sample regression or in the sub-samples. Non-tradable share reform influences national sample, low separation of ownership and control right sample and industrial sample more. The reason why non-tradable share reform increases corporate performance is that it improve the corporate governance efficiency. The impact of large shareholders’behavior on listed companies is positive. Convergence-of-interests effect plays a major role. When we put large shareholders variable and non-tradable share reform variable cross, we find that for the same control right, ownership right, separation of ownership and control right, the first shareholder, the second shareholder, the difference between the first shareholder and the second shareholder, corporate performance increases after non-tradable share reform. Only when the nature of control right is country or the separation of ownership and control right is low, corporate performance reduces after non-tradable share reform for the same the second shareholder.Now, our study can find that non-tradable share reform improves corporate governance efficiency, and that make the interests of large and small shareholders converge, and that let large shareholders company’s value oriented. After non-tradable share reform, tunneling reduces, and corporate performance increases.Now, our study found that at least the following: split share reform to improve the efficiency of corporate governance, so that the interests of small shareholders in line, the interests of major shareholders in the company’s value orientation of their return. After the share reform, large shareholders reduce the transmission behavior, corporate performance increases, while the share reform variable and the variable cross-largest shareholder, the shareholder for the same variables, better company performance after the share reform. When we put large shareholders variables and non-tradable share reform variable cross, for the same large shareholders variables, corporate performance is better after non-tradable share reform.In this dissertation, the innovations are: (1) we construct tunneling and corporate performance model according to Chinese actual situation. LLSV model is a general model for tunneling and corporate performance. In view of the actual situation of the stock market, we establish tunneling and corporate performance model before and after non-tradable share reform and compare them. We find that the effect of share structure and governance structure on tunneling and corporate performance do not change in manner, but strengthen to some extent. That is, this paper confirms the effect of tunneling and corporate performance in theory first. (2) We first use a large sample to test the influence of non-tradable share reform on tunneling. To measure tunneling is difficult in corporate finance. If we use fund occupation and related party transactions to measure tunneling, we cannot distinguish normal transactions from tunneling. From 2006, the state begins to formulate fund occupation and related party transactions. We use UFIDA’s case and empirical method to confirm dividend tunneling, and to test the effect of non-tradable share reform on tunneling. In the same time, the distribution of dividend is not for a single purpose, so we control free cash flow hypothesis, outcome model, substitute model and dividend life cycle. In order to examine the effect of non-tradable share reform on tunneling in depth, this article does sub-sample research from control right nature, equity balance and industry. We also verify the real reason why the non-tradable share reform can reduce tunneling. Therefore, this article on the research of the effect of non-tradable share reform on tunneling is thorough and innovative. (3) This paper first utilizes a large sample to test the effect of non-tradable share reform on corporate performance. Past scholars’researches have several questions. First, the test method is too simple, just comparing two years or two quarters before and after non-tradable share reform. Second, the empirical sample is too short to verify the research problem. Their researches are too general and not detailed enough. In response to these problems, this study includes all the listed companies. We test the effect of non-tradable share reform on corporate performance through comparative analysis, regression analysis and sub-samples regression analysis. We also study the effect of the control structure of the pyramid on corporate performance. We put large shareholders variable and non-tradable share reform variable cross to examine the effect of non-tradable share reform on large shareholders and corporate performance.更多还原