关键词: 高管薪酬, 创新, 委托代理理论, 公司治理

Abstract: Technological innovation is a heterogeneous activity characterized by high risk, unpredictability, long term, multi-stage and labor intensity. Therefore, it needs the efforts of executives and poses severe challenges to the design of incentive contracts for executives. The theoretical model of Manso (2011) shows that the standard pay for performance scheme is not conducive to innovation, and the optimal executive incentive plan should show substantial tolerance (or even reward) for short-term failures and reward long-term success. 
Based on the theoretical analysis of Manso (2011), this paper attempts to examine the impact of short-term managerial compensation on technological innovation of listed companies. This paper argues that the short-term managerial compensation has a positive impact on the technological innovation of listed companies. High short-term pay is a tolerance for early failure, even a reward in a sense. Higher short-term salary inhibits the risk aversion tendency of senior executives and encourages them to invest more in research and development based on the long-term value of the enterprise. Therefore, higher short-term salary will promote the innovation input of the enterprise. Assuming that the probability of R&D success is stable, more R&D input will naturally generate more R&D output (such as invention patents), so higher short-term salary will also promote the innovation output of listed companies. Not only that, higher short-term salary can eliminate the worries of senior executives to some extent, which is not only conducive to more R&D activities of senior executives, but also conducive to the focus of senior executives to strengthen the management of innovation, thus improving the innovation efficiency of enterprises.
Taking the A-share listed companies in Shanghai and Shenzhen from 2008 to 2016 as samples, this paper finds that the short-term executive compensation has a significant positive impact on the future technological innovation of enterprises. This is embodied in three aspects: first, short-term executive compensation increases the investment in technological innovation. According to the availability of data, this paper defines the short-term executive compensation as the natural logarithm of the total amount of the top three short-term executive compensation (i.e. salary and bonus, excluding allowances). Technological innovation input is defined as the ratio of R&D input to total assets. Control variables include total assets, the ratio of fixed assets to the total number of employees, the annual stock return, the ratio of market value to book value, the ratio of total leverage, the tenure of CEO, the age of CEO, the education of CEO, the year and industry dummy variables. In order to control the influence of autocorrelation and heteroscedasticity on regression results, the standard deviations of regression coefficients were clustered at the enterprise level. Second, short-term executive compensation increases the technological innovation output of enterprises, which is measured by the number of invention patents granted. Third, short-term executive compensation improves the efficiency of technological innovation of enterprises.
In view of the above empirical results, this paper makes a large number of robustness tests. First, this article uses instrumental variables. Although the previous regression equation contains a large number of control variables, there may still be endogenous problems between short-term executive compensation and technological innovation. One possibility is that the characteristics of an enterprise cause the enterprise to set higher short-term compensation incentives and carry out more technological innovation activities, thus leading to the spurious regression between the short-term executive compensation and the technological innovation of the enterprise. To avoid this problem, this article use instrumental variable regression. Appropriate instrumental variables should satisfy both relevance and exclusion. With this in mind, this article provides two instrumental variables for 2SLS estimation. The first instrumental variable is the average of short-term executive compensation of the companies in the same industry. The second instrumental variable is the average of short-term executive compensation in the same province. The results show that the short-term executive compensation has a significant positive impact on the input, output and efficiency of technological innovation. Second, this paper controls other variables. According to the existing literature, this paper takes the duality of CEO, the board size, the independence of board, the degree of accounting conservatism, whether the property right of the company is state-owned, and whether the audit institution belongs to the big four firms as additional control variables. The empirical results show that the short-term executive compensation still has a significant positive impact on the input, output and efficiency of technological innovation.
The prediction of the positive relationship between short-term executive compensation and enterprise technology innovation is based on the conflict of interest between shareholders and managers, which is easy to induce managers to implement opportunistic behavior. However, in a stronger regulatory environment, the opportunistic behavior of executives is more likely to be curbed, and the positive impact of short-term executive compensation on technological innovation of enterprises will be more obvious. Therefore, this paper further studies the moderating effect of external regulatory intensity. Consistent with Kim et al. (2011), this paper uses the number of analysts covered and the shareholding ratio of institutional investors as proxy indicators for external supervision. The larger the number of analysts or the higher the proportion of institutional investors, the greater the intensity of external regulation. The empirical results of this paper show that: external supervision has a significant moderating effect on the relationship between the short-term executive compensation and the technological innovation of enterprises; external supervision is more powerful, and the incentive effect of short-term executive compensation is more obvious.
This paper has three policy implications: first, it shows that enterprises can motivate their executives to become better innovators by properly designing short-term compensation. Second, the research results of this paper also show that the board needs to pay special attention to the time dimension of executive compensation in order to motivate innovation. Third, the research results of this paper also indicate that China should actively disclose the executive compensation plans of listed companies in the future.

Key words: managerial compensation, innovation, agency theory, governance