Page 257 - Ebook HTKH 2024
P. 257

(2)  TEC2  is  defined  as  the  amount  of  STD  funds  divided  by  the  firm’s  sales
                  revenue.

                         Since only a small number of companies have established STD funds, we found
                  that 32 out of 523 companies contributed to the STD fund, representing 10% of the total.
                  To  address  the  impact  of  numerous  zero  values,  we  employ  a  Tobit  model  in  our
                  regression analysis.
                        3.3. Control variables
                        We  include  a  host  of  firm-specific  attributes  that  are  good  predictors  of  firm
                  technologization based on prior literature (Gan and Xu, 2019; Tong et al., 2018; Yang,
                  Xu, and Lai, 2021). At the firm level, we control for the following variables: (1) Size is
                  the natural logarithm of a firm’s market value, (2) Leverage is the ratio of long‐term

                  debt over total assets, (3) Cash is the ratio of cash over total assets, (4) ROA is the ratio
                  of net income over total assets, (5) ROE is the ratio of net income over shareholder’s
                                                                          %
                  equity. We will normize all continuous variables at 1  and 99% levels. The descriptive
                  statistics of these firm-level control variables are presented in Panel A of Table 1.
                        4. Empirical results
                        4.1. Descriptive statistics
                        Panel A of Table 1 reports the descriptive statistics of the main variables used in

                  the study. There are 1,886 variables in total from 2019 to 2021. The mean value of TEC1
                  is  1.24, and the mean value of TEC2 is 0.0003. At the provincial level, the mean value
                  of Environmental governance is 3.487.
                        Panel B reports the Pearson correlation analysis of the variables. As shown in Panel
                  B, the pair-wise correlation Environmental governance with other variables is generally
                  low, as expected, indicating that no serious multicollinearity problem exists when we
                  include these variables in the regression model.

                                                   Table 1. Descriptive statistics
                        Panel A: Summary statistics
                              Variable        N        mean       sd        p25          p50       p75
                           Environmental
                     governance               1886     3.487      0.695     2.904      3.317     3.687

                           TEC1               1886     1.079      4.756     0          0         0
                           TEC2               1886     0.0003     0.0027  0            0         0
                           LEV                1886     0.509      0.479     0.251      0.478     0.683
                           SIZE               1886     26.599     1.755     25.537  26.627       27.687
                           ROA                1886     0.041      0.079     0.009      0.036     0.076
                           ROE                1886     0.088      0.177     0.024      0.085     0.159
                           CASH               1886     0.077      0.085     0.017      0.049     0.104













                                                                                                         249
   252   253   254   255   256   257   258   259   260   261   262