Abstract:     The separation of ownership and control in modern enterprises leads to information asymmetry between management and shareholders. It is costly or difficult for shareholders to monitor management. Management has incentives to implement behaviors that maximize the benefit of themselves rather than the profit of shareholders. Consequently, the agency problem arises, which will affect management investment behavior, especially R&D investment behavior. In the case of serious agency problems, management may reduce investment in high-risk investment projects, such as research and development. There are two main reasons why this paper analyzes the impact of the implementation of equity incentive plan on R&D investment of enterprises. First, technological innovation is an important driving force for economic growth, which is considered to be the input of human and financial resources derived from R&D. Second, with high levels of uncertainty and information asymmetry, R&D investment is a powerful tool for testing governance measures such as equity incentive. The research in this paper is helpful to give an insight into the endogenous mechanism of management equity incentive to promote R&D investment, and it is of great significance to further improve the equity incentive mechanism and promote enterprise R&D activities to enhance the innovation capability of enterprises.
    The standard principal-agent model ignores the endogenous problems of R&D activities and equity incentives, which leads to inconsistent conclusions of empirical tests. Some scholars (Barker and Muller (2002), Wu and Tu (2007), Chen Linrong (2018), Chen Xiaodong and Zhou Jianan (2014)) believe that equity incentives can promote R&D investment of enterprises. Some scholars research shows that there is no clear correlation between CEO equity incentives and R&D investment (Balkin et al. (2000)). O′Connor and Rafferty (2012) believe that R&D investment is affected by many factors, in addition after controlling variables, the relationship between corporate governance and R&D investment is weakened. Shen Liping, Huang Qin (2016), Tang Qingquan (2011), Ghosh (2010) and other scholars believe that there is a non-linear relationship between executive shareholding level and R&D investment. The inconsistency of the existing literature results shows that it is difficult to empirically test the relationship between equity incentive and R&D investment, because the relationship between R&D investment and equity incentive is endogenous. The contradictory results in the research literature are precisely the reflection of this endogeneity. Moreover, most of the literature takes management shareholding as an alternative variable of equity incentive, which lacks guiding significance for the establishment of equity incentive mechanism to promote enterprise innovation in China. Since the implementation of the Equity Incentive Management Measures for Listed Companies (Trial Implementation) in 2006, up to August 2016, 924 equity incentive plans issued by listed companies and approved by the shareholders′ meeting have involved 739 listed companies, of which 145 listed companies have implemented two or more equity incentive plans in the form of stock options, restricted shares and stock appreciation rights, which provides a realistic basis for empirical testing of this endogenous problem and a test condition for establishing a reasonable equity incentive system to promote innovation.
    This paper takes companies that implemented equity incentive during the period of 2006-2013 as the research sample, with all companies in this industry as the control sample, the inverse Mills ratio as the management private information measurement variable, and the Heckman method for regression analysis. The results show that the management private information is significantly related to the R&D investment of the enterprise, that is, the enterprise R&D investment and the equity incentive are endogenous; on the basis of controlling endogeneity, implementation decision of the equity incentive is related to the R&D investment of the enterprise in the next three years, that is, the management of the company that implements equity incentive tends to increase R&D investment in the next three years. However, combined with the results of the sample selection model, companies with high shareholding ratio of executives and high compensation of executives tend to implement equity incentive, indicating that the implementation of equity incentive is the result of insider control. The results of this study show that the increase of R&D investment cannot be regarded as the alleviation of agency problem. The essence is the result of agency problem. That is to say, the positive correlation between incentive effect and R&D investment is the result of agency problem, not the alleviation of agency problem. Therefore, listed companies should be cautious in implementing equity incentives to avoid making equity incentives a tool for executives to make profits.
   The contribution of this paper lies in the following three aspects: (1) The existing literatures focus on alleviating the endogenous problem of R&D investment and equity incentive, and obtaining a better estimation coefficient of R&D investment and equity incentive, whereas this paper argues that it is more meaningful to explore the endogenous mechanism of R&D investment and equity incentive itself. The root of endogenous problem of equity incentive lies in the joint decision-making of the implementation of equity incentive and the later R&D investment, that is, the implementation of the equity incentive determines the level of later R&D investment. In addition to the economic factors such as the compensation of management, shareholding ratio and ownership concentration, the implementation of equity incentive is inevitably influenced by the management′s private information (that is, the management′s judgment on the future output of R&D input (likelihood ratio)). This kind of private information will also affect the R&D investment after the implementation of equity incentive. (2) A large number of literatures have studied the principal-agent relationship between shareholders and management, most of which are related to the adverse selection before the signing of the management compensation contract and moral hazard afterwards. Few literatures have analyzed the moral hazard caused by the equity incentive contract. This paper analyses the moral hazard model of R&D investment selection after the signing of the equity incentive contract, and incorporates the implementation of the equity incentive contract into the decision-making process of R&D investment selection. The endogenous mechanism of equity incentive and R&D investment is put forward; (3) This paper proposes to use the inverse Mills ratio to measure management′s private information, which affects the implementation of corporate equity incentive and later R&D investment. Therefore, inverse Mills ratio can be constructed in the implementation model of equity incentive to measure private information of management.

Key words: equity incentive, R&D investment, agency problems, endogeneity