*4.3. Descriptive Statistics*

The means (M), standard deviations (SD), and correlations of all variables are reported in Table 3. The mean values (standard deviations) of environmental corporate social responsibility and green innovation performance were 4.650 (0.949) and 4.042 (1.019), demonstrating that Chinese firms have shouldered environmental corporate social responsibility more actively in recent years. However, there was heterogeneity in terms of green innovation performance among these firms. The mean value of shared vision capability was 4.060, indicating that employees tended to identify with firms that actively shouldered environmental corporate social responsibility. The mean value of resource slack was 4.058 and the standard deviation was 1.201, which showed that there was a difference in resource slack between firms. Moreover, as the results show, environmental corporate social responsibility was positively associated with green innovation performance (r = 0.343, *p* < 0.01) and shared vision capability (r = 0.361, *p* < 0.01). The results also confirm that shared vision capability was significantly related to green innovation performance (r = 0.347, *p* < 0.01). Moreover, resource slack was found to be positively associated with green innovation performance (r = 0.394, *p* < 0.01), and also with shared vision capability at work (r = 0.338, *p* < 0.01).

#### *4.4. Hypothesis Testing*

We used a hierarchical regression analysis to examine the proposed hypotheses in SPSS 25.0 [69]. Table 4 shows the results of the hypothesis testing. Firstly, environmental corporate social responsibility had a significantly positive impact on green innovation performance (β = 0.360, *p* < 0.01). Therefore, Hypothesis 1 was confirmed.


**Table 4.** Hierarchical regression analysis results.

Note: <sup>+</sup> *p* < 0.10; \* *p* < 0.05; \*\* *p* < 0.01.

Hypothesis 2 proposed that shared vision capability mediates the relationship between environmental corporate social responsibility and green innovation performance. Referring to the traditional testing methods recommended by Baron and Kenny [70], we evaluated the mediating role of shared vision capability: (1) controlling for gender, age, education, job grade, job tenure, firm type, firm size, and firm age, and environmental corporate social responsibility positively affected shared vision capability (β = 0.277, *p* < 0.01); (2) environmental corporate social responsibility had a positive impact on green innovation performance (β = 0.360, *p* < 0.01); (c) shared vision capability had a positive influence on green innovation performance (β = 0.463, *p* < 0.01); and (d) after taking shared vision capability into account, the impact of environmental corporate social responsibility on green innovation performance became smaller and significant (β = 0.264, *p* < 0.01), which represented partial mediation. To confirm the robustness of the mediating role of shared vision capability, we used PROCESS, an SPSS macro (95% CI, 1000 bootstrap resamples) to examine it. The results indicated that environmental corporate social responsibility had a positive influence on green innovation performance through shared vision capability (Indirect effect = 0.096, CI (0.048,0.154)), supporting Hypothesis 2.

Hypothesis 3 predicted that firms with high levels of slack resources can strengthen their green innovation performance even more via shouldering environmental corporate social responsibility than firms with low levels of slack resources. As shown in Table 4, the results demonstrated that the interaction between environmental corporate social responsibility and resource slack had a statistically significant and positive impact on green innovation performance (β = 0.126, *p* < 0.01). In order to further demonstrate the moderating role of resource slack, we drew interaction diagrams on the basis of a procedure suggested by Aiken and West [71]. Figure 2 depicts how the positive impact of environmental corporate social responsibility on green innovation performance is stronger with high levels of resource slack (1 SD above the mean) than with low levels of resource slack (1 SD below the mean). Moreover, to test Hypothesis 3, we adopted a simple slope analysis. The simple slope test showed that environmental corporate social responsibility was more strongly correlated to increased green innovation performance with high levels of resource slack (slope = 1.042, t = 4.711, *p* < 0.01) than with low levels of resource slack (slope = 0.665, t = 5.119, *p* < 0.01). Thus, Hypothesis 3 was supported.

**Figure 2.** The moderating effect of resource slack.

For the sake of confirming the findings about the moderating role of resource slack, we adopted PROCESS, an SPSS macro (95% CI, 1000 bootstrap resamples), to examine Hypothesis 3. As shown in Table 5, environmental corporate social responsibility had a significant and direct influence on green innovation performance with high levels of resource slack (Effect = 0.408; CI (0.249, 0.567)). However, the moderating effect was smaller with low levels of resource slack (Effect = 0.119; CI (0.004, 0.234)). Hypothesis 3 was supported again.

**Table 5.** The impact of environmental corporate social responsibility on green innovation performance on different levels of resource slack.


#### **5. Discussion**

#### *5.1. Theoretical Implications*

Firstly, extant studies indicated that there is a positive impact of corporate social responsibility on firms' outcomes [43,72]. However, few studies have explored the environmental corporate social responsibility–green innovation performance relationship, and the correlation mechanism is still unclear. On the basis of social identity theory, we confirmed that environmental corporate social responsibility is positively associated with green innovation performance via shared vision capability. This finding indicates that a future research direction might be exploring corresponding mechanisms by which corporate social responsibility connects to firms' outcomes [73]. Moreover, there are conflicting views on the impact of corporate social responsibility on innovation outcomes [74,75]. This finding can also help to clarify the inconclusive results of previous studies by suggesting that a mediating role of shared vision capability exists between environmental corporate social responsibility and green innovation performance.

Secondly, most prior studies have confirmed that green customer and supplier integration [7], green absorptive capacity [45], and organizational green learning [10] can be drivers of green innovation performance. The study provides new insight into the antecedents of green innovation performance. We scrutinized the environmental corporate social responsibility–green innovation performance link and the shared vision capability– green innovation performance link in Chinese firms. The results of the data analysis indicate that environmental corporate social responsibility and shared vision capability are positively associated with green innovation performance, which expands the existing research on the antecedents of green innovation performance. The finding also responds to the call of Du et al. [7], who put forward future research directions that deepen the research on the antecedents of green innovation performance.

Finally, we found that resource slack played a positive moderating role in the relationship between environmental corporate social responsibility and green innovation performance. Previous studies showed that resource slack can moderate the corporate social responsibility–firms' outcomes relationship. Alshorman et al. [76] empirically confirmed that resource slack moderated the influence of corporate social responsibility on firm market value by investigating 95 nonfinancial Jordanian firms. Xie [77] found that slack resources can improve the positive link between corporate social responsibility and green technology innovation. Our finding extends the knowledge pool by linking the environmental corporate social responsibility, resource slack, and green innovation performance literature [7,11,76]. Meanwhile, the finding about the moderating role of resource slack verifies the proposition of resource-based theory [29]. High levels of resource slack can provide firms with adequate human, material, and financial resources to innovatively make manufacturing processes greener and lower carbon in response to an environmentally

responsible strategy. Thus, we further clarified the boundary conditions of the effect of environmental corporate social responsibility on green innovation performance, which deepens the research conclusions of Mo et al. [73] and gives insight into the moderating role of resource slack on the corporate social responsibility–firms' outcomes relationship.
