The Future of ESG in Multinationals: How Digital Twin Technologies Enable Strategic Value Creation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper explores how Digital Twin (DT) technologies serve as strategic enablers of Environmental, Social, and Governance (ESG) performance in multinational corporations. Grounded in socio-technical systems theory and stakeholder theory, the study employs a multi-method design—combining in-depth case studies of Siemens, Unilever, Tesla, and BP, capital market event analysis, and machine learning–assisted regression—to demonstrate that DT adoption yields tangible ESG improvements, such as lower emissions, enhanced safety, better supplier compliance, and faster reporting. The authors further show that DT-related ESG announcements trigger statistically significant positive abnormal stock returns and that the DT–ESG link is moderated by governance structures and ESG maturity. The paper compellingly argues that DTs transcend their technical function by fostering transparency, auditability, and stakeholder trust, thereby driving both internal transformation and external legitimacy. While the study makes a timely contribution to the emerging literature on ESG digitalization, several methodological, theoretical, and practical issues require attention before the manuscript can be considered suitable for publication. Detailed comments follow.

 

1. Model 3 suffers from multicollinearity, which invalidates the results and prevents verification of the core moderating effect. The construction of the ESG Maturity Index (ESGMI) includes the dependent variable, ESG Score, potentially leading to overfitting and rendering the estimated coefficients statistically meaningless. Although the authors acknowledge this issue, they fail to address it—e.g., by excluding ESG Score or replacing it with a non-overlapping indicator such as “clarity of ESG strategy”—thus rendering the central hypothesis that “ESG maturity moderates the DT–ESG relationship” untestable.

 

2. The paper assumes by default that “digital transformation (DT) adoption improves ESG performance,” yet it does not test for reverse causality: firms with stronger ESG performance and greater resources may be better positioned to adopt DT technologies first. Quantitative evidence shows that DT adopters have a significantly higher average ESG score (0.79) than non-adopters (0.72), making it impossible to rule out the possibility that “high ESG performance is a prerequisite for DT adoption.”

 

3. In Model 1, firm size (log of total assets) exhibits a significant negative association with ESG performance, contradicting established theory—which posits that larger firms possess more resources, greater capacity for ESG investment, and thus typically achieve better ESG outcomes. The authors neither investigate the reasons behind this counterintuitive finding nor provide a coherent explanation, leaving a critical gap in the logical chain.

 

4. The paper recommends that “policymakers should incentivize DT adoption” and “develop open ESG data systems,” but it fails to specify concrete details—such as the types of incentives, technical standards for the data infrastructure, or priority industries for support—thereby offering limited actionable guidance for policy design.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Please, see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study examines the strategic role of digital twin (DT) technology in improving the environmental, social, and governance (ESG) performance of multinational corporations. Drawing on sociotechnical systems theory and stakeholder theory, it presents meaningful case studies demonstrating that DT adoption leads to quantifiable ESG improvements, including reduced emissions, improved safety, and shorter reporting cycles. However, the following revisions are necessary to ensure the paper's completeness.

1. The abstract clearly summarizes the study’s purpose, methodology, and main findings; however, it fails to mention the significant methodological limitations identified in the quantitative analysis, which may lead to an overestimation of the study’s strengths.

2. In Model 3, the ESG Maturity Index (ESGMI) was constructed by including the dependent variable, the ESG Score, within the index. As a result, multicollinearity was observed in Model 3, and an R² value close to 1.00 indicated overfitting, rendering the coefficients analytically meaningless.

3. The event study employed the market model to calculate expected returns, using the MSCI World Index as a proxy for market returns. However, while the daily stock return data source (Yahoo Finance) was specified, detailed information on the market indices used for sector-level benchmarking was insufficient.

4. In Model 2, the interaction term between DT adoption and the ESG Committee (DT × ESG Committee) was statistically insignificant (p = 0.202). Although the authors claim a potential synergy effect based on mean score differences, this interpretation is grounded in pattern observation rather than statistical evidence. It should be clearly acknowledged that the moderating effect was not statistically supported, and such an interpretation is based on descriptive rather than inferential statistics.

5. In the baseline regression of Model 1, firm size (Log Assets) showed a significant negative effect on ESG performance (β₂ = −0.0119, p = 0.013). This finding contradicts the common expectation that larger firms possess greater ESG-related resources and capabilities. While this result is discussed in detail in the Discussion section, its statistical significance should also be explicitly highlighted in the Results section.

6. Tesla’s DT–ESG announcement generated a statistically insignificant market reaction (+0.42%). The author attributes this weak response to the ambiguous ESG framing of the announcement (as discussed in the Discussion section); however, more contextual details—such as the specificity of the announcement content—should be clearly presented in the Results section to substantiate this interpretation.

Author Response

Please, see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have carefully addressed all the previous comments with thorough revisions. The manuscript has been significantly improved and is now suitable for publication. No further comments.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Dear Reviewer,

Thank you very much for your careful reading of the manuscript and for the constructive comments. I am grateful for the opportunity to improve the paper. All suggested changes have been implemented, and the revised manuscript includes Track Changes to highlight modifications.

Thank you again, and please, see manuscript for consulting the changes.

Reviewer 2 Report

Comments and Suggestions for Authors

Overall, the previous revision requirements have been well reflected, but some additional revisions are needed.

 1. The specific hypotheses (H1, H2, and H3, which are later illustrated in Figure 4) are not explicitly stated within the text of the Materials and Methods. A strong Materials and Methods section should conclude by clearly listing the formalized hypotheses to be tested.

2. The synthesis of these two theories (DTs as mediators between internal capabilities and external accountability) is crucial. While verbally explained, the proposed theoretical model's positioning of DTs should be briefly introduced visually or elaborated upon earlier to ground the conceptual argument more effectively.

3. The manuscript correctly identifies that the construction of the ESG Maturity Index (ESGMI) resulted in structural multicollinearity because it included the dependent variable (ESG Score). This led to the exclusion of Model 3 from inferential testing. This part needs some explanation.

4. The finding that firm size (Log Assets) exhibits a statistically significant negative association with ESG performance is a major and unexpected empirical result. Although briefly mentioned in the Results section, this interpretation should be minimized here and instead flagged for further discussion in Section 4.

5. The policy recommendations are strong, focusing on incentives and standardized data systems (CSRD, ISSB alignment). These should be explicitly tied back to the successful case examples (e.g., Siemens' CSRD-aligned dashboard and BP's methane detection) to reinforce the practical basis of the recommendations.

Author Response

Dear Reviewer,

Thank you very much for your careful reading of the manuscript and for the constructive comments. I am grateful for the opportunity to improve the paper. All suggested changes have been implemented, and the revised manuscript includes Track Changes to highlight modifications.

Thank you again, and please, see manuscript for consulting the changes.