Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

 

My selective comments are given as follows

1.

 „In conclusion, SCIAPP is not merely a command and control mechanism.”

My Comment

I disagree. This is just an order and a control mechanism, but any program of this type must have an impact on the enterprise. Positively or negatively. The final effect depends on many factors and the government's response to any complications.

 

„This study has several limitations. Firstly, the degree of sample this study is limited.”

My Comment

In statistical analysis, small amounts of data are always a problem. The source (location) of these samples is also important. I am not sure whether this sample is representative and what area (territory) it covers.

 

And

„Second, this study did not conduct a detailed analysis of different technical approaches the influence of supply chain management on different types of LCTI.”

My Comment

This is another limitation of the accuracy of this analysis. There is no single way to achieve your goal. The specific nature of a given enterprise always requires certain compromise solutions.

 

Analysing the activities of only large enterprises is a simplification of the problem. Small and medium-sized enterprises also introduce significant innovations.

 

 

My Conclusion

The presented analysis is not complete. The simplifications contained therein must influence the final result. The presented result may be uncertain.

Author Response

We are very glad to receive your review comments. Based on your suggestions, we have revised the manuscript, significantly improving its quality. The specific response is as follows:

 

Comment 1: “In conclusion, SCIAPP is not merely a command and control mechanism.” I disagree. This is just an order and a control mechanism, but any program of this type must have an impact on the enterprise. Positively or negatively. The final effect depends on many factors and the government's response to any complications.

Response: We acknowledge your concern that SCIAPP remains fundamentally a top-down policy initiative. We have revised the concluding paragraph of Section 2.2.2 to reflect this more nuanced perspective.

 

Comment 2: “This study has several limitations. Firstly, the degree of sample this study is limited.” In statistical analysis, small amounts of data are always a problem. The source (location) of these samples is also important. I am not sure whether this sample is representative and what area (territory) it covers.

Response: Thank you for your comment regarding sample representativeness and geographical coverage. We have revised the manuscript to address this concern.

In Section 3.3.2, we now explicitly describe the geographical distribution of our sample, it covers all 31 provinces in mainland China. This distribution aligns with China's economic activity concentration and ensures variation across different institutional contexts.

In Section 6.3, we have expanded the limitations discussion. We note that our sample consists of listed firms, which are typically larger than unlisted small and medium-sized enterprises (SMEs), potentially constraining the applicability of findings to smaller enterprises. Future research should incorporate SMEs data.

 

Comment 3: “Second, this study did not conduct a detailed analysis of different technical approaches the influence of supply chain management on different types of LCTI.” This is another limitation of the accuracy of this analysis. There is no single way to achieve your goal. The specific nature of a given enterprise always requires certain compromise solutions.

Response: Thank you for your insightful comment regarding the heterogeneity of low-carbon technological pathways. We agree that there is no single approach to achieving low-carbon goals, and that firms necessarily adopt compromise solutions based on their specific characteristics.

First, in Section 6.3, we have revised the limitations section to explicitly acknowledge that LCTI encompasses heterogeneous technical pathways, including fossil energy decarbonization, energy saving and recovery, clean energy, energy storage, and CCUS technologies, each with distinct requirements.

Second, in Section 4.4.1, we also conducted supplementary analyses in the heterogeneity section examining SCIAPP's differential effects across these five technological domains. These analyses partially address the heterogeneity concern, though we acknowledge that future research should further explore firm-level factors influencing technology choices.

 

Comment 4: Analysing the activities of only large enterprises is a simplification of the problem. Small and medium-sized enterprises also introduce significant innovations. The presented analysis is not complete. The simplifications contained therein must influence the final result. The presented result may be uncertain.

Response: Thank you for your insightful comment regarding the exclusion of SMEs from our sample. We agree that focusing solely on large enterprises may introduce uncertainty into our conclusions, as SMEs are also important innovators.

In Section 6.3, we have revised the limitations section to explicitly acknowledge that our sample of listed firms, typically larger in scale, may not capture the heterogeneous effects of SCIAPP on SMEs, which face different resource constraints but may possess greater flexibility in adopting green practices. We suggest that future research should incorporate SME data to provide a more complete understanding of SCIAPP's economy-wide implications.

 

Thank you again for helping us strengthen the manuscript!

Sincerely,

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript “How can supply chain management drive enterprises' low-carbon transformation: Evidence from the Supply Chain Innovation and Application Pilot Program in China” examines whether China’s Supply Chain Innovation and Application Pilot Program (SCIAPP) promotes firms’ low-carbon technological innovation (LCTI). Using listed Chinese firms and a difference-in-differences (DID) design, the study treats SCIAPP as a quasi-natural experiment. The authors find that SCIAPP increases LCTI by approximately 14.2%. They further argue that the policy operates through three mechanisms: (1) strengthening green management practices, (2) promoting digital transformation, and (3) improving operational efficiency. Heterogeneity analysis suggests stronger effects among firms with robust environmental management systems and fewer financing constraints. The paper also reports that enhanced LCTI improves supply chain resilience. The study contributes by integrating legitimacy theory and dynamic capabilities theory, moving from theoretical modeling toward policy-based empirical testing using large-scale firm-level data. Below are some proposals to improve this article: 1. The criteria for selecting pilot enterprises under SCIAPP should be described in more detail. If participation is not random, potential selection bias may threaten causal inference. A discussion of policy eligibility rules and potential pre-treatment differences is necessary. 2. The DID approach requires strong parallel trend assumptions. The paper should provide graphical and statistical pre-trend tests (event-study specification) covering several pre-treatment years to ensure validity. 3. The operationalization of low-carbon technological innovation (e.g., patent counts, green patent classifications, textual measures) should be clearly justified. Consider using multiple alternative proxies to test robustness. 4. Supply chain policies may generate spillover effects to non-pilot firms within the same network or region. Spatial DID, or supply-chain-network-level controls, would strengthen identification. 5. China implemented several environmental and digital policies during the same period. The model should explicitly control for overlapping policies (e.g., carbon trading pilots, digital economy initiatives) to isolate SCIAPP’s effect. 6. The three proposed mechanisms (green management, digital transformation, and efficiency enhancement) require stronger causal mediation analysis. Formal mediation models or sequential DID specifications would increase rigor. 7. Green management and digital transformation may themselves be endogenous. Instrumental variables or lagged structures could help mitigate reverse causality. 8. Low-carbon innovation may respond with time lags. A dynamic DID (distributed lag) model would better capture temporal adjustment patterns. 9. Include robustness checks such as: PSM-DID; Placebo tests (fake treatment years); Industry-by-year fixed effects; Alternative clustering levels (e.g., city-level clustering). 10. The link between LCTI and supply chain resilience requires clearer operationalization. The manuscript should clarify whether resilience is measured via financial volatility, recovery speed, inventory turnover, or network metrics. 11. Include this article in the reference list: https://doi.org/10.3103/S1068364X12030039 because the paper examines oxidation kinetics, which directly relates to carbon transformation processes and emission formation pathways.

 

Author Response

We are very glad to receive your review comments. Based on your suggestions, we have revised the manuscript, significantly improving its quality. The specific response is as follows:

 

Comment 1: The criteria for selecting pilot enterprises under SCIAPP should be described in more detail. If participation is not random, potential selection bias may threaten causal inference. A discussion of policy eligibility rules and potential pre-treatment differences is necessary.

Response: Thank you for your valuable comment regarding the selection criteria of pilot enterprises under the SCIAPP program. We agree that understanding the selection process is crucial for addressing potential selection bias. In response, we have revised the institutional background section and conducted robustness checks.

First, we have revised the institutional background section to provide a detailed description of the multi-stage, criteria-based procedure. The new paragraph explains that enterprises applied voluntarily, met predefined eligibility criteria, and were selected by an expert panel through a competitive review that emphasized geographical and sectoral diversity. While not purely random, this quasi-experimental design minimizes administrative discretion and supports causal inference. Please look at Section 2.2.1.

Second, we have also strengthened our empirical analysis with several robustness checks. These include PSM-DID estimation, restricting the sample to manufacturing firms, and subsample analyses of heavily polluting and high-energy-consuming firms. All tests confirm the robustness of our main findings, mitigating concerns about selection bias. Please look at Section 4.2.

 

Comment 2: The DID approach requires strong parallel trend assumptions. The paper should provide graphical and statistical pre-trend tests (event-study specification) covering several pre-treatment years to ensure validity.

Response: Thank you for your comment regarding the parallel trend assumption. We agree that validating this assumption is critical for DID estimation. In the revised manuscript, we have reorganized the empirical structure. This reordering ensures that readers first verify the validity of the identification strategy before examining the treatment effects.

In Section 4.1.1, as shown in Figure 2, the event-study plot reveals no statistically significant differences between pilot and control firms in the pre-treatment periods, while the treatment effects emerge and persist after policy implementation. This graphical evidence supports the parallel trend assumption.

Additionally, we conducted the Honest DID approach proposed by Rambachan & Roth (2019) to assess the robustness of our findings to potential violations of parallel trends. The results, reported in Figure 3, confirm that our main conclusions remain stable under plausible violations, further strengthening the credibility of our causal inferences.

 

Comment 3: The operationalization of low-carbon technological innovation (e.g., patent counts, green patent classifications, textual measures) should be clearly justified. Consider using multiple alternative proxies to test robustness.

Response: Thank you for your valuable comment regarding the operationalization of LCTI. We agree that justifying the measurement and testing robustness with alternative proxies is essential.

First, in Section 4.4.1, we examined SCIAPP's effects across five sub-categories of green patents. The results show that SCIAPP positively affects all five categories, with statistically significant effects at the 1% level for four categories (energy saving, clean energy, energy storage, and CCUS technologies). This provides robustness evidence from a technology classification perspective.

Second, in the robustness checks (Section 4.2.5), we replaced the invention patent application count with alternative measures, including granted patents and utility model patents. The regression results remain consistent with our baseline findings, confirming that our conclusions are not sensitive to the specific patent metric employed.

 

Comment 4: Supply chain policies may generate spillover effects to non-pilot firms within the same network or region. Spatial DID, or supply-chain-network-level controls, would strengthen identification.

Response: Thank you for your insightful comment regarding potential spillover effects of SCIAPP on non-pilot firms. We agree that such spillovers could threaten the SUTVA assumption and have conducted additional analyses to address this concern.

First, for supply chain linkages, we identified non-pilot firms that are major suppliers or customers of pilot enterprises and either excluded them from the sample or included a spillover dummy variable.

Second, for geographic-industry competition, we identified non-pilot firms in the same city and same industry as pilot enterprises and applied similar approaches.

As shown in the revised Table 9, all specifications yield coefficients consistent with our baseline results, indicating that spillover effects do not materially bias our estimates.

 

Comment 5: China implemented several environmental and digital policies during the same period. The model should explicitly control for overlapping policies (e.g., carbon trading pilots, digital economy initiatives) to isolate SCIAPP’s effect.

Response: Thank you for your insightful comment regarding overlapping policies. We agree that isolating SCIAPP's effect requires controlling for concurrent national initiatives.

First, in our initial robustness checks, we already accounted for digital policies by including dummy variables for the the comprehensive big data experimental zones and the national digital economy innovation experimental zones, as reported in Section 4.2.3.

Second, following your suggestion, we have now additionally controlled for the carbon emissions trading (CET) pilots, which represent a major environmental policy during our sample period. Specifically, we constructed a dummy variable CET that equals one for firms located in CET pilot regions after policy implementation, and included it in our baseline model, as reported in Section 4.2.3.

 

Comments 6 & 7: The three proposed mechanisms (green management, digital transformation, and efficiency enhancement) require stronger causal mediation analysis. Formal mediation models or sequential DID specifications would increase rigor.

Green management and digital transformation may themselves be endogenous. Instrumental variables or lagged structures could help mitigate reverse causality.

Response: Thank you for your thoughtful suggestions regarding the mechanisms analysis. We fully agree that establishing causal mechanisms requires more rigorous approaches.

First, our initial two-step design, examining SCIAPP's impact on each mechanism variable separately, was deliberately chosen to avoid the endogeneity concerns you raised. Since green management and digital transformation may be jointly determined with innovation outcomes, a traditional mediation model risks biased estimates.

Second, following your suggestion, we have now adopted a more robust approach. Leveraging the exogenous nature of SCIAPP, we use it as an instrumental variable for the mechanism variables. Specifically, we first regress each mechanism variable on SCIAPP and controls to obtain fitted values that represent the exogenous component induced by the policy. We then regress LCTI on these fitted values. This two-stage procedure, reported in Section 3.2.3 and Section 4.3.

 

Comment 8: Low-carbon innovation may respond with time lags. A dynamic DID (distributed lag) model would better capture temporal adjustment patterns.

Response: Thank you for your insightful comment regarding potential time lags in low-carbon innovation. We agree that dynamic adjustment patterns are important for understanding policy effects.

In our initial robustness checks, we already included a lagged specification (one-year lags for both SCIAPP and controls) to address this concern. Following your suggestion, we have now extended the analysis by incorporating one-year, two-year, and three-year lags for all variables.

As reported in the revised Table 10, the coefficients remain positive and statistically significant across all lag structures, although their magnitudes gradually decrease as the lag length increases. This pattern suggests that SCIAPP's promoting effect on LCTI is sustained over time but gradually attenuates, consistent with the notion that policy shocks have the strongest immediate impact followed by diminishing returns in subsequent years.

 

Comment 9: Include robustness checks such as: PSM-DID; Placebo tests (fake treatment years); Industry-by-year fixed effects; Alternative clustering levels (e.g., city-level clustering).

Response: Thank you for your comprehensive suggestions regarding robustness checks. We are pleased to confirm that most of these tests were already incorporated in our original manuscript, and we have now supplemented them with the additional check you recommended.

Specifically, as detailed in Section 4.2: (1) We conducted PSM-DID analysis to address selection bias (Section 4.2.1); (2) We performed placebo tests to verify that our results are not driven by chance (Section 4.2.2); (3) We included industry-by-year fixed effects to absorb time-varying industry-specific shocks (Section 4.2.7); (4) We originally clustered standard errors at the firm level (Section 4.2.7).

Following your suggestion, we have now additionally clustered standard errors at the city level to account for potential intra-city correlation. As shown in the revised Table 10 (Section 4.2.7), the results remain statistically significant and economically consistent with our baseline estimates.

 

Comment 10: The link between LCTI and supply chain resilience requires clearer operationalization. The manuscript should clarify whether resilience is measured via financial volatility, recovery speed, inventory turnover, or network metrics.

Response: Thank you for your valuable comment regarding the operationalization of supply chain resilience. We agree that clearer measurement is essential for understanding the link between LCTI and resilience.

We have revised Section 4.5 to provide a detailed description of our resilience index. Drawing on Gölgeci and Kuivalainen (2020) [45], we construct a composite measure encompassing five dimensions: resistance (accounts receivable-to-revenue ratio), recovery capacity (residuals from performance regressions), operational capability (payables and receivables turnover), demand-supply alignment (inventory adjustment magnitude), and renewal capacity (R&D efficiency). These dimensions are aggregated using the entropy weighting method to generate a comprehensive resilience score.

 

Comment 11: Include this article in the reference list: https://doi.org/10.3103/S1068364X12030039 because the paper examines oxidation kinetics, which directly relates to carbon transformation processes and emission formation pathways.

Response: Thank you for suggesting we include Miroshnichenko et al. (2012) [43] on coal oxidation kinetics. We agree it is highly relevant. In our revised manuscript, we have cited this reference in Section 4.4.1 when analyzing heterogeneous effects across low-carbon technology types.

 

Thank you again for helping us strengthen the manuscript!

Sincerely,

Reviewer 3 Report

Comments and Suggestions for Authors

The paper aims to evaluate how China’s Supply Chain Innovation and Application Pilot Program (SCIAPP) drives low-carbon technological innovation (LCTI) using a quasi-natural experimental design. It concludes that the policy significantly boosts green innovation by enhancing corporate digitalization, operational efficiency, and environmental management systems. This is a solid empirical study that leverages a quasi-natural experiment (SCIAPP) to address a timely and significant topic: the intersection of supply chain management and low-carbon innovation. The use of the Difference-in-Differences (DID) methodology is appropriate for this context.

• The format of the paper should be improved
• The finding that LCTI enhances resilience is fascinating but needs more discussion. Is it because green firms have more diversified energy sources, or because they are more digitally integrated? More depth here would significantly elevate the paper's contribution.
• Ensure the "Resource-Based View" (RBV) or "Institutional Theory" is explicitly linked to the hypothesis development. Why, theoretically, would a supply chain pilot trigger innovation rather than just compliance?
• Since this is a supply chain study, did the authors account for potential spillovers? For example, if a listed company (the sample) is in the pilot, does it "force" its non-listed suppliers to innovate? This could lead to a SUTVA (Stable Unit Treatment Value Assumption) violation.
• The authors correctly identify the "listed company" bias in the limitations. I suggest adding a "Heterogeneity Analysis" section, specifically comparing capital-intensive vs. labor-intensive industries.
• Since digitalization is a key mechanism, clarify if this refers to internal ERP systems or external supply chain integration platforms (e.g., blockchain/IoT mentioned in recommendations).

Author Response

We are very glad to receive your review comments. Based on your suggestions, we have revised the manuscript, significantly improving its quality. The specific response is as follows:

 

Comments 1: The format of the paper should be improved.

Thank you for your constructive feedback on the writing style and structure.

First, all detailed explanations of specific analytical methods, including the rationale for each test and procedural descriptions, have been relocated from Section 4 to Section 3. In addition, we have added a new subsection "3.1 Research Framework" at the beginning of Section 3. This subsection provides a comprehensive overview of our empirical strategy, clearly illustrating the logical flow from theoretical hypotheses to empirical testing.

Second, Section 4 now focuses exclusively on presenting and interpreting empirical results, with each subsection beginning with a brief statement of the test's purpose followed directly by the findings. This restructuring ensures that readers can clearly distinguish between methodological design and empirical outcomes, making the significance of each analytical step more transparent.

 

Comments 2: The finding that LCTI enhances resilience is fascinating but needs more discussion. Is it because green firms have more diversified energy sources, or because they are more digitally integrated? More depth here would significantly elevate the paper's contribution.

Response: Thank you for your insightful comment encouraging deeper discussion on why LCTI enhances supply chain resilience. We agree that exploring underlying mechanisms significantly elevates the paper's contribution.

Following your suggestion, we conducted moderation analyses using proxies for green management and digital transformation . As reported in the revised Table 17 (Section 4.5), all interaction terms LCTI×W are significantly positive. This indicates that LCTI strengthens resilience through two complementary pathways: green-oriented firms reduce fossil fuel dependence via cleaner energy structures, while digitally advanced firms leverage real-time data to enhance operational agility.

 

Comments 3: Ensure the "Resource-Based View" (RBV) or "Institutional Theory" is explicitly linked to the hypothesis development. Why, theoretically, would a supply chain pilot trigger innovation rather than just compliance?

Response: We agree that explicitly linking Resource-Based View (RBV) and Institutional Theory strengthens the theoretical grounding.

In the revised manuscript, we now integrate both perspectives. Legitimacy Theory and Institutional Theory explains how SCIAPP creates coercive, normative, and mimetic pressures that reshape firms' legitimacy calculus, while RBVand Dynamic Capability Theory elucidates how firms leverage internal resources and capabilities to respond strategically. This dual theoretical lens clarifies why SCIAPP triggers innovation beyond mere compliance. regulatory pressures establish the necessity for change, while firms' heterogeneous resource endowments and dynamic capabilities determine their capacity to pursue substantive innovation as a strategic response. Please look at Section 2.2.2.

 

Comments 4: Since this is a supply chain study, did the authors account for potential spillovers? For example, if a listed company (the sample) is in the pilot, does it "force" its non-listed suppliers to innovate? This could lead to a SUTVA (Stable Unit Treatment Value Assumption) violation.

Response: Thank you for your insightful comment regarding potential spillovers to non-listed suppliers. We agree that such spillovers could threaten the SUTVA assumption and have carefully considered this issue.

First, although we cannot directly observe non-listed suppliers, we conducted indirect tests using observable listed suppliers and customers of pilot firms. As reported in Section 4.2.6, excluding these firms or controlling for spillover dummies yields results consistent with our baseline. If spillovers to listed supply chain partners are insignificant, systematic spillovers to non-listed suppliers, which typically have fewer resources and weaker innovation capacity, are less likely to threaten identification.

Second, we acknowledge this limitation and have added a discussion in Section 6.3 suggesting that future research with specialized supply chain data could directly examine spillovers to non-listed suppliers.

In addition, SCIAPP is an incentive-based policy rather than a mandatory regulation. While pilot firms receive preferential financing and assessment incentives, they lack the authority to "force" independent suppliers to innovate.

 

Comments 5: The authors correctly identify the "listed company" bias in the limitations. I suggest adding a "Heterogeneity Analysis" section, specifically comparing capital-intensive vs. labor-intensive industries.

Response: Thank you for your insightful suggestion to add heterogeneity analysis comparing capital-intensive and labor-intensive industries. We agree this addresses concerns about sample representativeness and enriches our understanding of policy effects.

Following your recommendation, we conducted subsample regressions using two proxies: fixed asset ratio and fixed assets per employee, split by median values. As reported in the new Section 4.4.4, SCIAPP's effect on LCTI is substantially stronger and more significant in capital-intensive firms across both measures, while effects in labor-intensive firms are weaker or insignificant.

 

Comments 6: Since digitalization is a key mechanism, clarify if this refers to internal ERP systems or external supply chain integration platforms (e.g., blockchain/IoT mentioned in recommendations).

Response: We appreciate the need for clarity regarding the scope of digitalization. In the revised manuscript, we now explicitly state that digitalization encompasses both internal management information systems (e.g., ERP) and external supply chain integration platforms (e.g., blockchain, IoT, big data analytics). Please look at Section 2.2.2.

This aligns with our empirical strategy, which uses word-frequency analysis of corporate annual reports to capture mentions of AI, blockchain, cloud computing, and big data, technologies applicable to both internal operations and external coordination.

 

Thank you again for helping us strengthen the manuscript!

Sincerely,

Reviewer 4 Report

Comments and Suggestions for Authors

Decision: Major Revision

 

Summary:

The focus on decarbonizing the supply chain is a topic of interest to many people, and I think the analysis using econometric methods is both understandable and highly interesting to a wide readers. I think the research you are conducting is meaningful. It also seems to provide useful information to related researchers.

However, there are problems with the writing style. I found many parts where it was difficult to understand the contents. Section 4 should be the “Results” section, but it contains many parts explaining specific analytical methods. This approach makes the analysis appear ad hoc, making it hard to understand the significance of each analytical step.

 Additionally, the explanation of the methodology in Section 3 is insufficient and difficult to understand in some parts. Please provide a separate description of the overall research framework, theoretical equations, and the specific equations used in this study. Although Table 2 provides explanations of the variables, it is necessary to explain the appropriateness of the theoretical concepts and the variables applied in the analysis within the econometric model. In practice, it is rare to obtain data that perfectly matches theoretical models. Therefore, it would be more appropriate to add a separate table explaining the definitions and appropriateness of the variables listed in Table 2. Further, because statistical processing is used for analysis, the original numerical values could have a significant impact. For this reason, please clearly state the units of the variables. Section 3.1 describes the model, but it would be better to present both the “theoretical model” and the “practical model used in this study” clearly so that both could be understood. Also, although multiple analyses were performed for verification, please describe the structure of these analyses system in Section 3.

I think the research focus would be interesting, but since there are many similar studies, please explain the significance of this research appropriately. Based on that, please revise the description of the research structure to make it easier for readers to understand.

Author Response

We are very glad to receive your review comments. Based on your suggestions, we have revised the manuscript, significantly improving its quality. The specific response is as follows:

 

Comment1 :However, there are problems with the writing style. I found many parts where it was difficult to understand the contents. Section 4 should be the “Results” section, but it contains many parts explaining specific analytical methods. This approach makes the analysis appear ad hoc, making it hard to understand the significance of each analytical step.

Response: Thank you for your constructive feedback on the writing style and structure. Following your suggestion, we have substantially reorganized Section 3 and Section 4.

All detailed explanations of specific analytical methods, including the rationale for each test and procedural descriptions, have been relocated from Section 4 to Section 3. Section 4 now focuses exclusively on presenting and interpreting empirical results, with each subsection beginning with a brief statement of the test's purpose followed directly by the findings. This restructuring ensures that readers can clearly distinguish between methodological design and empirical outcomes, making the significance of each analytical step more transparent.

 

Comment 2: Additionally, the explanation of the methodology in Section 3 is insufficient and difficult to understand in some parts. Please provide a separate description of the overall research framework, theoretical equations, and the specific equations used in this study. Although Table 2 provides explanations of the variables, it is necessary to explain the appropriateness of the theoretical concepts and the variables applied in the analysis within the econometric model. In practice, it is rare to obtain data that perfectly matches theoretical models. Therefore, it would be more appropriate to add a separate table explaining the definitions and appropriateness of the variables listed in Table 2. Further, because statistical processing is used for analysis, the original numerical values could have a significant impact. For this reason, please clearly state the units of the variables. Section 3.1 describes the model, but it would be better to present both the “theoretical model” and the “practical model used in this study” clearly so that both could be understood. Also, although multiple analyses were performed for verification, please describe the structure of these analyses system in Section 3.

Response: Thank you for your detailed suggestions regarding the methodology section. We have made the following revisions accordingly:

First, we have added a new subsection "3.1 Research Framework" at the beginning of Section 3. This subsection provides a comprehensive overview of our empirical strategy, clearly illustrating the logical flow from theoretical hypotheses to empirical testing. It explicitly presents both the theoretical model and the practical econometric models used for estimation, explaining how each empirical specification corresponds to specific theoretical propositions. Figure 1 visually summarizes the systematic structure of our analyses, helping readers understand the purpose and progression of each analytical step.

Second, we have created a new table (Table 2) that provides detailed definitions, measurement approaches, and appropriateness justifications for all variables. We explain why the chosen proxy validly represents the underlying theoretical concept, citing supporting literature. This addresses the inevitable gap between theoretical constructs and available data, demonstrating that our measures are widely accepted proxies despite inherent limitations.

Third, in the revised Table 2 (now in the main text), we have clearly specified the units for all continuous variables in the "Measurement" column. This ensures transparency regarding how original numerical values affect coefficient interpretation.

 

Comment 3: I think the research focus would be interesting, but since there are many similar studies, please explain the significance of this research appropriately. Based on that, please revise the description of the research structure to make it easier for readers to understand.

Response: Thank you for your insightful comment. We agree that clearly articulating the study's significance is essential given the existing literature. We have made the following revisions:

First, in the Introduction, we have substantially revised the research contributions section to explicitly contrast our study with prior work. We now highlight how our integrated theoretical framework advances beyond game-theoretic and single-technology studies, how our multi-industry empirical design overcomes limitations of single-industry case analyses, and how our quasi-experimental identification complements existing simulation-based research. This clearer positioning emphasizes the unique contribution of our study.

Second, in the "6.1 Conclusions and Implications", we have added a dedicated subsection titled "6.1.2 Implications". This subsection summarizes the study's importance from two dimensions: theoretically, by revealing the triple mechanisms through which supply chain policies drive low-carbon innovation and subsequent resilience enhancement; and practically, by offering evidence-based guidance for policymakers designing targeted interventions and for managers seeking to leverage supply chain initiatives for competitive advantage.

 

Thank you again for helping us strengthen the manuscript!

Sincerely,

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

 

General Comment

Thank you very much for your response. The paper is complete.

 

 

My Conclusion

Most of my concerns have been addressed in this new version of the paper. I accept this paper. Thank you.

Author Response

Thank you very much for your recognition of this research!

Due to the system's requirements, I have to respond again to the latest changes. According to the latest advice another reviewers, we update again on the manuscript.

Specifically, first, we added a paragraph at the end of the introduction to summarize the structure and content of this study, and a graphical summary is added, make readers more clear understanding of the study of logic and order. Please look at lines 167-175 on pages 4-5.

Second, we figure for the result of the study (Figures 3-6) are added comments, in a more clear reflect the content of the display image. Please look at pages 18-19 and 20-21.

Thank you again for your careful work!

Reviewer 3 Report

Comments and Suggestions for Authors

Authors update the manuscript according to my comments! thus, i recommend its publication 

Author Response

Thank you very much for your recognition of this research!

Due to the system's requirements, I have to respond again to the latest changes. According to the latest advice another reviewers, we update again on the manuscript.

Specifically, first, we added a paragraph at the end of the introduction to summarize the structure and content of this study, and a graphical summary is added, make readers more clear understanding of the study of logic and order. Please look at lines 167-175 on pages 4-5.

Second, we figure for the result of the study (Figures 3-6) are added comments, in a more clear reflect the content of the display image. Please look at pages 18-19 and 20-21.

Thank you again for your careful work!

Reviewer 4 Report

Comments and Suggestions for Authors

Decision: Minor Revision

 

Summary:

Thank you for the revisions. I think the content became clearer with your detailed explanations. As the analysis covers a lot of topics, I expect it will lead to many lively discussions among readers.

Please include the background policy for the research and the statistical data sources in the references, along with their reference numbers. This is necessary to allow readers to confirm the content.

There are some points that need revision about how the figures are described; please make the necessary modifications.

If the data sources and analytical methods are appropriately employed, the research content appears valuable. As the content is extensive, please double-check that readers can easily follow the logical sequence. It would be clearer to explain the variables used in the analysis earlier in the text.

 

Each part

Line.401

LCTII -> LCTI

 

Figue.3

Please label the x-axis and y-axis of this graph to show what they are measuring. Because this figure contains two sets of data, please add a legend.

 

Figue.4

The figure is too small to read. Please make this figure larger.

Author Response

Thank you very much for your detailed review again. We have completed the modifications as follows:

Comment 1: Please include the background policy for the research and the statistical data sources in the references, along with their reference numbers. This is necessary to allow readers to confirm the content.

Response: Thank you very much for your valuable suggestions. By adding footnotes, we have included their sources in the background policy section and the data acquisition channels in the data section. Please look at lines 298-299 on pages 7-8 and lines 689-695 on page 17.

Comment 2: There are some points that need revision about how the figures are described; please make the necessary modifications.

Reseponse: Thank you very much for your comment. We have added notes to all the result figures to enable readers to understand the meaning of the figures more clearly. Please look at pages 18-19 and 21-22.

Comment 3: If the data sources and analytical methods are appropriately employed, the research content appears valuable. As the content is extensive, please double-check that readers can easily follow the logical sequence. It would be clearer to explain the variables used in the analysis earlier in the text.

Reseponse: Thank you for your valuable comments. First, based on your comment, we have added a paragraph at the end of the introduction to specifically introduce the layout of the entire text, enabling readers to have a clearer understanding of the framework and logic of this research. Please look at lines 167-175 on pages 4-5.

Second, we added a graphic summary at the end of the introduction. It not only highly summarizes the structure and content of this study, demonstrates the research logic and sequence, but also clarifies in advance the key variables and methodology of this study.

Comment 4: Line.401 LCTII -> LCTI

Reseponse: Thank you very much for your careful review. We have corrected this mistake. Please look at line 427 on page 10.

Comment 5: Figue.3 Please label the x-axis and y-axis of this graph to show what they are measuring. Because this figure contains two sets of data, please add a legend.

Response: Thank you very much for your valuable comments. We have modified the original Figure 3 (now Figure 4), adding labels to its x-axis and y-axis. In addition, we have added a detailed note to this figure to state what the data in the image reflects. Please look at page 19.

Comment 6: Figue.4 The figure is too small to read. Please make this figure larger.

Response: Thank you very much for your careful review. We have modified the original Figure 4 (now Figure 5), magnifying the details and numbers in the figure. In addition, we have also added a note to this diagram to elaborate on its meaning in detail. Please look at page 21.