Abstract

The non-tradable share reform beginning in 2005 has been called an important milestone mark on the capital markets in recent years, and since then, the separation of tradable shares and non-tradable shares in the past capital market has gone, even the state-owned shares are tradable. All shares are freely circulate, which also leads to some phenomena rising and developing in the capital market, and major shareholders injecting assets into listed companies is one of those phenomena. Many scholars has found that on the one hand, the Chinese government encourages major shareholders of listed companies to inject high-quality assets into companies to achieve overall listing and promote listed companies to become bigger and stronger. On the other hand, certain equity value-added benefits can be made by asset injections of major shareholders. Therefore, asset injections are upsurged both in state-owned and private-owned listed companies. In recent years, many studies have argued on the motives and the economic consequences of major shareholders' asset injections. Some studies show that asset injections generate synergy effects, enhance the integration of industrial chains, and also reduce the related party transactions which are“supports”from major shareholders. But more researches find that in the long run, the asset injection is a measure for major shareholders to“tunneling”. However, this work notes that the local state-owned listed companies constitute a procession of“main force”in the asset injection. Besides, major shareholders of local state-owned listed companies are usually local SASAC or local state-owned conglomerate firms, and these major shareholders are essentially local government's“front-stage voice”, thus the local government can intervene in local state-owned listed companies through their economic activities more conveniently. In addition, in a transition economy like China, government intervention in the economy and the enterprises is a very common phenomenon. When the market still doesn't fully effectively play a decisive role in the allocation of resources, the government intervention is an essential alternative and complementary mechanism that can generate some“supporting”functions, but in reality, the“offside”government intervention in macro-economic and micro-economic activities causes excessive imbalances, the non-standardized market order and business inefficiencies, and makes enterprises not achieve business objectives properly. These problems are equally thought-provoking, and are concrete reflections of government“grabbing hand”. Few existing studies discuss from the perspective of government intervention to study whether the local government intervention in the asset injection is the“supporting hand”or“grabbing hand”, which is based on the characteristic that the local government is the actual controller of local state-owned listed companies. Therefore, the first question of this work is whether the government intervention improves or reduces the local state-owned listed companies' performance after the asset injections, and whether there exists the motivation of“tunneling”by the local government. Analyzing this is sue would enrich the studies of the government intervention and the asset injection, and provide a reference for further clarifying the relationship between the government and enterprises in China.

In addition, this work also notes that our local state-owned listed companies are generally controlled by the state-owned shareholders through the pyramid ownership structure, and the pyramid ownership structure is established and is developing gradually accompanied by the reform of China's state-owned enterprises. The main purpose of the introduction of the pyramid ownership structure in state-owned enterprises is to optimize the relationship between the government and enterprises by means of the characteristics of the pyramid shareholding pattern, and then gradually realize the separation of the government's ownership function and community administrative function. Moreover, some researches have confirmed that the pyramid ownership structure helps to achieve the ultimate controllers'separation of ownership and control and effective allocation of the Groups'internal resources, and alleviate internal corporate financing constraints. In addition, one important feature of the pyramid ownership structure is that it can defense the external forces' intervention on enterprises, such as the government. Existing studies have not specifically discussed the pyramid ownership structure's defensive role on local government intervention on local state-owned listed companies' asset injections, and few researches have studied the pyramid ownership structure's defensive role from its multi-layer and multi-chain feature. Therefore, the second question of this work is that if the“grabbing hand”of the government intervention exists in local state-owned listed companies'asset injections, whether the pyramid ownership structure can significantly alleviate the“grabbing”effect, and what inherent impacts the pyramid ownership structure's layers and chains would have on restraining the government intervention. Discussing this issue would help enrich the studies of the pyramid ownership structure's defensive role, and provide some empirical evidence in order to improve ownership structures and corporate governance mechanisms of China's stateowned listed companies.

Based on the above ideas, after reviewing related domestic and foreign studies, the work introduces the regulation background of the government intervention in local state-owned enterprise at first, and then analyses motives, means, types and outcomes of the government intervention. Next, the work studies motives, means and consequences of the government intervention on asset injection activities. Last, from the features and the functions of the pyramid ownership structure, accompanied with the institutional background, the work analyses how the pyramid ownership structure improves local state-owned listed companies' performance after asset injections by restraining the government intervention. On the basis of theoretical analysis, this work proposes the corresponding research hypothesis, uses asset injection events of Chinese local state-owned listed companies from 2006 to 2011 as the sample, and tests the relationship between the government intervention and asset injection performance as well as the relationship among the government intervention, the pyramid ownership structure and asset injection performance. Finally, based on the empirical analysis, the results and the conclusions of this work are summarized. With that, the work proposes several related policy recommendations.

This work gets the following conclusions：

（1）Compared with developed countries, due to the incompleteness and imperfectness of market-oriented economy, it is impossible to make the market fully play its role in distributing resources in China. Thus, the government intervention still plays a key role in the development of China's current economy and it could be a substitutive mechanism of the market. Meanwhile, it is common that the government and government officials take precedence to fulfill their own political goals or society governance goals with the sacrifices of operating goals of companies. In order to quickly reach their goals and transfer the burdens, and get promotion under the idea of achieving their official careers, government officials are urgent to intervene in companies through different ways. On the one hand, with the methods of the government intervention on companies, mergers and acquisitions of properties can quickly satisfy the government and officials. On the other hand, with the ownership of listed companies, local state-owned listed companies are usually controlled by the local State-owned Assets Supervision and Administration Commission（SASAC）or state-owned conglomerate firms, and their actual controller is the local government. This makes the government intervention on local state-owned listed companies is more intensive and frequent. The asset injection is preferred as a merger and acquisition model after the reform of non-tradable shares. Since the local government has the ability and the motivation to inject low-quality assets into local state-owned listed companies as a controller, asset injctions help the local government to enlarge scales of local state-owned listed companies and get rid of burdens of non-profitable state-owned assets. But those assets cannot bring any margin profits. What's worse, they may be irrelevant to companies'primary business and not generate positive effects of resource integration, which would be a disadvantage for the future perform ance. It is more likely to be a“tunneling”effect on local state-owned listed companies in the future with the“forced marriage”style of asset injections under the government intervention. The government is not the substantial shareholder of private-owned listed companies, and due to the reason of protection of state-owned assets, it is unlikely to force private-owned listed companies to take over assets which controlled by the government. Private-owned listed companies will consider more about the effect on its progress of performance and shareholders' wealth after asset injections. Based on the analysis above, it is possible to conclude that the performance of local state-owned listed companies would be worse than privateowned listed companies after asset injections with considering of the ownership. This conclusion has been proved by the empirical research of this work.

（2）Economy development, infrastructure construction, living standard and even the government governance degree between different provinces（cities）have substantial differences due to the effect of location, historical change and government policy. In general, the marketization degree has a decreasing trend from the eastern coastal regions to the western inland regions. Although a huge gap still exists from developed countries, in China, compared with the low marketization regions, in the high marketization regions, the degree of market-oriented economy, the ability of absorbing and using capitals and talents, the system construction and enforcement, the regulation and punishment have all developed to a mature and complete level. The inverse relationship between the degree of the marketization and the government intervention shows that the increase of the degree of market-oriented economy means the improvement of the domi nant role of market on the economy development and the resource distribution. Accordingly, the effect of the government intervention as a substitute of marketization would be shortened or reduced, and vice versa.

Therefore, it would be helpful to deepen observations on the government intervention on the performance of local state-owned listed companies after asset injections, with quantizing the degree of the government intervention and the degree of the market development where local state-owned listed companies are located. The higher the degree of the government intervention on local state-owned listed companies, the more convenience the government has to give more pressure and transfer official goals and burdens toward companies' asset injection activities. And the motivation and the effect of“tunneling”and self benefit transferring of the government through forcing companies' asset injections would be stronger. The result would be different at the region with higher level of market development. Since the market dominant effect on the resource distribution is stronger, and better legal regulation and enforcement would be restrictions to such infringement behavior and rent-seeking, a certain restraint on“tunneling”of the asset injection under the government intervention would emerge. Therefore, this work points out that the stronger the degree of the government intervention and the lower the degree of the market development of an area, the worse the performance of local state-owned listed companies would be after asset injections. Based on the analysis above, this work adopts the index of“Chinese marketization index”to measure the degree of the government intervention and the degree of the market development of the province（city）where local state-owned listed companies are located, and the empirical study has proved the conclusion of the theoretical analy sis.

（3）The pyramid ownership structure of local state-owned listed companies has developed with the promotion of the reform of state-owned enterprises. The original intention of introducing the pyramid ownership structure is to progress the separation of the government and enterprises step by step and strengthen the enterprise's independent operation and principle statue in the market, without changing the ownership of state-owned properties. More and more state-owned enterprises have become listed by the promotion of the government and conglomerate firms with the formation and growing of Chinese stock market. The problem of the owner's absence has been solved with the establishment of central and local SASAC, and meanwhile, the three-level pyramid model of state-owned listed companies which includes SASAC, the state asset management company and the state-owned enterprise（including the state-owned listed company）has also been radicated. The pyramid ownership structure increases the cost of the government intervention and restrains the government intervention on local state-owned listed companies by inserting many different levels of intermediate companies. It could be an important substitutive mechanism of preventing the government intervention and reducing the“tunneling”effect when the market is imperfect and the legal protection is weak. Based on the analysis above, this work stresses that the performance of local stateowned listed companies with the pyramid ownership structure is better than those without the pyramid ownership structure, as the pyramid ownership structure has the function of restraining the government intervention. This conclusion has been proved by the empirical research of this work.

（4）Multi-layer and multi-chain is a notable feature of the pyramid ownership structure and it is the main reason that makes government intervention difficult to take effect. With the increasing of vertical layers, there would be more serious agency problems for those agents who are also client. Thus, the information transfer would be delayed and there would be more agency costs which should be taken by the government. Finally, the government would further“decentralize”after balancing costs and benefits and the government intervention would be reduced. With the increasing of horizontal agency chains, the capacity of company amount would be enlarged at the interlayers, and then the government intervention will get more interference from the companies and entities on the chains, which increases the information asymmetry for the government. Finally, agency costs and supervision costs would be additionally increased for government intervention and encroachment activities, thus the pyramid structure could be an effective way to control government intervention on local state-owned listed companies.

Thus, the more layers the pyramid structure has, the more conflicts and disturbances from the agency relationship during local state-owned listed companies' asset injections would emerge. Agency costs, information transferring costs and supervision costs would increase during the government intervention, for the government is a controller under such situations. When the costs exceed the expected returns from asset injections, the rational choice of decreasing the asset injection activities by the government intervention would be chosen. Thus, the degree of government“tunneling”effect in asset injections would decrease and the performance of companies injected would increase. Based on the analysis above, this work proposes that the more layers and chains the pyramid ownership structure has, the more restraining effects on the government intervention during asset injections in local state-owned listed companies would emerge. And the performance would be better after asset injections. Based on the analysis above, the number of layers and chains of the pyramid ownership structure would be classified based on medians of layers and chains. The above conclusions have been confirmed by the empirical research of this work.

The main innovations of this work are as follows:（1）The work explains the government intervention on enterprises with another angle by attributing local state-owned listed companies' worse performance after asset injections to the government intervention.（2）The work gives a further confirmation of the pyramid ownership structure's restraining effect on the government intervention with the event of asset injection.（3）At the part of specific variable and indicator designing, we conduct a further study of the restraining effect of the pyramid structure on government intervention based on the characteristics of vertical layers and horizontal agency chains, together with the empirical test, which deepens the study of the restraining effect of the pyramid ownership structure.

Key words: Government Intervention Pyramid Ownership Structure Asset Injection Corporate Performance Local State-Owned Listed Company