1. Introduction
Sustainable supply chain management (SSCM) is the embodiment of sustainable concepts in supply chain management, which is driven by the demands of stakeholders to manage the material flow, information flow, and capital flow in supply chain. It also enables managers to strategically and transparently integrate and realize the social, environmental, and economic goals of an organization [
1]. SSCM, which integrates environmental and social concerns into the supply chain, has been widely implemented by firms to face challenges in global sustainability. Elkington [
2] first put forward the triple bottom line for enterprises to encourage the firms to increase their economic leverage through protecting the environment and improving social performance. Environmental challenges place competing demands on enterprises who extend their green effort across their supply chains [
3]. Companies and their extended supply chains are a holistic system, and their management method at the supply chain level is important [
4]. Recent studies have shown that some companies and their suppliers develop sustainable strategies in order to reduce energy consumption and improve materials reuse [
5].
Most of the literature on firm size shows that large enterprises are more conducive to the company’s economic performance and productivity [
6]. While, in fact, as a firm expands, the firm consciously fulfills its corporate and social responsibilities. Such firms have a richer understanding of the corporate environment and social responsibility, and thus use better resources to fulfill their corporate social responsibility. Similarly, without a regulatory framework of green supervision, most SMEs have no incentive to pay heed to sustainability in the supply chain. However, SMEs represent an important sector of many countries, both economically and socially. For instance, in China, SMEs provide more than 80% of employment in cities and towns, and have become the main channel of employment. Indeed, more than 20 million SMEs registered with China’s Business Administration Council in 2015, which realized an increase in contribution of profits and taxes. In the industrial sector, more than 365,000 SMEs account for 97.4% of the total number of enterprises; these SMEs contributed taxes of 2.5 trillion RMB, accounting for 49.2% of the total tax revenue; their profits of 4.1 trillion RMB accounted for 64.5% of the total industrial profit. The development of China’s SMEs also creates many jobs and supports the community.
As the largest developing country, China’s sustainable development strategy and countermeasures are of great significance to the choice of sustainable development path in developing countries. The Chinese government has announced that the CO2 emission per unit of GDP will declines by 10–20%, compared with 2005 in the Doha Amendment. Moreover, the Paris Agreement also provides external institutional framework for sustainable development in China, which adds external pressures and impetus, and brings new opportunities for the transformation of economic structure and green development in China.
Previous studies on SSCM mainly focused on enterprise economic and environmental performance and very rarely embraced social dimension [
7,
8]. Further, empirical studies on SSCM on developing economy enterprises and SMEs are scant. Enterprises need a framework to help them to identify and implement their sustainability development schemes [
9]. The purpose of this paper is to verify the moderating role of firm size on sustainable supply chain management (SSCM) practices and performance, and, furthermore, to reveal the different impacts of SSCM practices on the performance of different firm size categories in China.
4. Analysis and Results
The hypotheses proposed in this paper were examined through six steps, using hierarchical regression analysis [
48]. First, the control variable of firm age and the control variable of industry type are involved in the regression. Second, the two independent SSCM practice factors are involved in the regression. Third, firm size as a moderating variable was included. The fourth to sixth layers of the model added the interaction between SSCM practices and firm size, respectively, to the test.
Table 4 shows the results of the regression analysis on economic performance. Model 1 in
Table 4 shows that the control variables do not have a significant impact on the enterprise environmental performance. Next, adding internal SSCM practices and external SSCM practices in Model 2 leads to a significant change in
R2 (Δ
R2 = 0.18,
p < 0.001), and a negative beta value for economic performance (
β = −0.11,
p < 0.05), which indicates that in order to improve the internal SSCM practices, the firm may need to reduce the economic benefit. Hypothesis 1a was thus not supported. The external SSCM practice shows a significant positive relationship with economic performance (
β = 0.19,
p < 0.01), thus supporting Hypothesis 2a. Although the internal SSCM practices will lower economic performance, good external SSCM practices will offset this effect. Model 3 showed that the addition of firm size does not improve the predictive ability of the regression model on economic performance (Δ
R2 = 0.24,
p < 0.001). The coefficient of firm size was statistically positive and significant (
β = 0.22,
p < 0.001). Hypothesis 3a was thus supported.
We added the interaction term of SSCM practices and small enterprises in Model 4, and found that the regression significantly increased the predictive power (Δ
R2 = 0.32,
p < 0.001). Such results indicate that moderation effects do exist. The coefficient of the interaction between the internal SSCM practices and small firms has no significant effect (
β = −0.09). However, the coefficient of the interaction between the external SSCM practices and small firms was statistically positive and significant (
β = 0.14,
p < 0.05). In Model5, we study the impact of SSCM practices on the economic performance of the mid-sized enterprises. It has a significant change in
R2 (Δ
R2 = 0.28,
p < 0.001). The internal SSCM practices have no significant impact on the economic performance of mid-sized enterprises (
β = −0.06), but external SSCM practices have a significant positive impact on the economic performance of mid-sized enterprises (
β = 0.15,
p < 0.01). Additionally, the negative beta for an interaction variable appeared in Model6 (
β= −0.13,
p < 0.05), indicating an opposite moderation effect. Further examination indicates that the coefficients of the interaction between the external SSCM practices and large enterprises are positive and significant (
β = 0.16,
p < 0.01). Thus, from Models 4, 5, and 6 in
Table 4, Hypothesis 4a was supported, but Hypothesis 5a was not supported. The results of
Table 4 show that the external SSCM practices are associated with economic performance improvement. The larger the firm size, the greater the influence of external practices on economic performance. While internal SSCM practices and economic performance are negatively correlated, large firm size can weaken this correlation.
Model 1 in
Table 5 shows that the control variables do not have a significant impact on the enterprise environmental performance. Adding internal SSCM practices and external SSCM practices in Model 2 leads to significant changes in
R2 (Δ
R2 = 0.81,
p < 0.001), which indicates that SSCM practices have a significant positive impact on environmental performance (for internal SSCM practices,
β = 0.41,
p < 0.01; for external SSCM practices,
β = 0.89,
p < 0.001). Hypotheses 1b and 2b were supported. External SSCM practices have a higher impact on a firm’s environmental performance. We inserted firm size into Model 3, and found that firm size has a significant positive relationship with environmental performance (Δ
R2 =0.81,
p < 0.001;
β = 0.11,
p < 0.01), thus supporting Hypothesis 3b.
Next, we added the interaction term of SSCM practices and small enterprises in Model 4, and found that the regression increased the predictive power significantly (Δ
R2 = 0.82,
p < 0.001). This suggests that moderation effects do exist. The result showed that the coefficient of the interaction between SSCM practices and small enterprises was positive and significant (for internal SSCM practices,
β = 0.24,
p < 0.01; for external SSCM practices,
β = 0.29,
p < 0.01). We discuss the impact of SSCM practices on the environmental performance of mid-sized enterprises in Model 5. The result showed that Model 5 has a significant change in
R2 (Δ
R2 = 0.83,
p<0.001). Both internal SSCM practices and external SSCM practices have a significant positive impact on the environmental performance of mid-sized enterprises (for internal SSCM practices,
β = 0.39,
p < 0.01; for external SSCM practices,
β = 0.49,
p < 0.01). Further examination indicates that the coefficients of the interaction between SSCM practices and large enterprise are positive and significant in Model 6 (for internal SSCM practices and large enterprises,
β = 0.76,
p < 0.001; for external SSCM practices and large enterprises,
β = 0.63,
p < 0.001). Thus, according to Models 4–6 of
Table 5, Hypotheses 4b and 5b were supported.
Table 5 shows that SSCM practices are associated with environmental performance improvement. The larger the firm size, the more significant the positive impact of internal management on environmental performance.
Model 1 in
Table 6 shows that the control variables do not have a significant impact on enterprise environmental performance. Model 2 has a significant change in
R2 (Δ
R2 = 0.76,
p < 0.001), the results showed that SSCM practices have a significant and positive impact on social performance (for internal SSCM practices,
β = 0.48,
p < 0.001; for external SSCM practices,
β = 0.47,
p < 0.001). Hypotheses 1c and 2c were supported. Firm size in Model3 has a significant impact on the
R2 of the regression model (Δ
R2 = 0.77,
p < 0.001), and firm size has a significant positive relationship with social performance (
β = 0.11,
p < 0.01). Hypotheses 3c was thus supported.
Model 4 in
Table 5 shows a significant change in
R2 (Δ
R2 = 0.80,
p < 0.001). Such results indicate that moderation effects do exist. The coefficient of the interaction between SSCM practices and small enterprises has a significant impact on social performance (for internal SSCM practices,
β = 0.41,
p < 0.01; for external SSCM practices,
β = 0.31,
p < 0.01). Next, on the impact of SSCM practices on social performance in mid-sized enterprises in Model 5, there is a significant change in
R2 (Δ
R2 = 0.78,
p < 0.001). SSCM practices have a significant positive impact on the social performance of mid-sized enterprises (for internal SSCM practices,
β = 0.43,
p < 0.01; for external SSCM practices,
β = 0.41,
p < 0.01).
Further examination indicates that the coefficients of the interaction between SSCM practices and large enterprise are positive and significant in Model 6 (Δ
R2 = 0.79,
p < 0.001; for internal SSCM practices and large enterprises,
β = 0.66,
p < 0.001; for external SSCM practices and large enterprises,
β = 0.91,
p < 0.001). The effect of external SSCM practices on social performance in large enterprises is more significant than that of internal SSCM practices. From Models 4, 5, and 6 in
Table 6, Hypotheses 4c and 5c were supported. The SSCM practices are associated with social performance improvement. The larger the firm size, the more significant the positive impact of internal management on social performance.