An Analysis On The Dynamic Effect of Enterprise’s Environmental Governance On Green Technology Innovation: New Explanation From The Perspective of Knowledge And Technology Transfer

: Environmental governance ( EG ) and green technology innovation ( GTI ) are 26 important means to promote the construction of ecological civilization of all countries 27 around the world. Past scholars focused on the impact of EG on GTI based on the static 28 perspective usually, but ignored the impact of the dynamic development law of 29 enterprises. This study differentiates, takes enterprise’s life cycle stage as the 30 breakthrough point, and analyzes the dynamic effect of EG on GTI at the first time. 31 Further, considering the important of information interaction among different enterprise 32 in the background of collaborative innovation, this study reveals the evolution trend of 33 enterprise’s knowledge and technology transfer ( KTT ) in different life cycle stage, and 34 explains the internal mechanism of the dynamic effect mentioned above. The theoretical 35 model finds that for enterprises in different life cycle stage, the impact of EG on GTI 36 depends on abatement cost and innovation compensation effect two aspects. The 37 development of enterprise’s KTT helps to strengthen the incentive effect of EG on GTI, 38 thus causing the differentiated GTI effect among different enterprises under the restrict 39 of EG . The empirical research results based on the micro data of Shanghai and 40 Shenzheng A-share listed firms from 2013 to 2018 in China confirm the theoretical 41 inference. EG has a positive role in promoting GTI , however, compared with the 42 enterprises in growth and mature stage, the positive innovation effect does not hold for 43 the enterprises in recession stage. The statistical results show that the enterprise’s KTT 44 in growth and mature stage is significantly better than that of enterprises in recession. 45 Moreover, the empirical analysis results confirm that enterprise’s KTT have a positive 46 moderating effect on the relationship between EG and GTI . Combined with the above 47 conclusions, this study puts forward several useful management implications for 48 improving the designing of environmental governance system, optimizing the 49 cooperation of GTI and the EG decision-making under the background of collaborative 50 innovation. 51


55
Due to the deviation between traditional economic model characterized by 56 environment resource consumption and social needs of sustainable economic 57 development, environmental governance (EG) is one of the major direction for the 8 study dynamically analyzes the impact of EG on enterprise's GTI in different life cycle 153 stages, and expounds the formation mechanism of dynamic effect mentioned above 154 from the perspective of KTT at the first time.  promote the evolution of GTI ability, and quickly reduce the cost of GTI denoted by The main concern of this study is whether EG can promote GTI. Due to the 234 measurement of GTI is the number of applications for environmental patents, so, this 235 paper constructs the following panel counting model.
where, the subscripts i and t denotes the firm and year, respectively. GTI indicates the 238 level of firm's green technology innovation, EG represents the environmental 239 governance. X represents the control variables, mainly includes the age of enterprises, 240 the current ratio, the return on assets, the turnover rate of assets and the enterprise where, a1 and a2 reflect the direct and indirect effects of the EG on the GTI, respectively.  Winsorize treatment was performed on the 1% level of all continuous variables. Table   261 1 shows the description analysis results.

284
The return on assets is the percentage of a company's profit divided by its net assets, 285 which reflects the ability to obtain income based on its own assets (Xin et al.,2010). 286 The growth of return on assets can enhance the financial support for green investment 287 and promote GTI. 288 The Current ratio is the ratio of current assets to current liabilities, which reflects 289 the ability of current assets to be realized and used to repay liabilities. The higher the 290 value is, the stronger the short-term solvency is, which is helpful to increase GTI 291 investment, otherwise it is weak.  The higher the asset turnover rate of an enterprise, the stronger its sales ability, the 300 higher the efficiency of asset utilization, and the lower the principal-agent cost. this study measures the KTT by the enterprises "Whether there is a technology 310 disclosure/transfer" and "whether there is a patent disclosure /transfer", where "yes" is 311 1 and "no" is 0. The value of VIF is much lower than 10, and indicates that there was no problem 315 with multicollinearity. The results of correlation analysis listed in the Table 2 shows 316 that the correlation coefficient of each variable is less than 0.7.

317
Insert Table 2  Based on the equation (12) and employed the static fixed effect model, the 320 benchmark effect of EG on GTI is tested. As can be observed from the results listed in 321 the second column of the Table 3, the coefficient of EG is 0.05 and passes the 322 significance test at 1% level. In order to guarantee the reliability of the empirical results, 323 each control variables are added step by step. As shown in columns from (3) to (7) of Lower EG is hardly to affect the enterprise production and management activities and   333 The impact of EG on GTI may be endogenous, which leads to inconsistent  According to the research of Arellano and Bover (1995) and Blundell and Bond 344 (1998), the System-GMM model is employed to estimate the results. As shown in the both of them passes the significance test at 1% level. It must be noted that the above 347 conclusion is based on the premise that GMM estimates are effective. The AR (1) , AR

348
(2) and Sagan test statistics reported in Table 4, which proved that the tool variables 349 selected in system-GMM estimation are valid.

350
Insert Table 3 here

382
Based on the previous analysis, we find that EG has different effects on GTI of        The numbers in parentheses are t-statistic; * p<0.1. ** p<0.05. *** p<0.01.  Year The scale of KTT