The Impact of Coordinated Two-Way FDI Development on Carbon Emissions in Belt and Road Countries: An Empirical Analysis Based on the STIRPAT Model and GMM Estimation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is a clearly written and original treatment of an interesting topic, the effects of two-way FDI on environmental performance in countries directly affected by China’s Belt and Road Initiative (BRI).  I believe that it suffers from two weaknesses, one broad and one having to do with presentation.

My broad concern is that the paper finds a beneficial impact of combined IFDI and OFDI on CO2 emissions, but it doesn’t include a measure that could balance or even outweigh this impact:  the effect of ODI in relocating emissions to even less developed countries.  The authors admit that “Most route-adherent states utilize IFDI introducing low-carbon sectors while employing OFDI relocating high-carbon operations.”  So how can we reach a judgment on the impact of the BRI on sustainable development without including data on this relocation of pollution, what the authors term the “pollution paradise hypothesis”?

My presentation concerns are with Section 4, “Empirical Research”.  The discussion of “a bidirectional FDI coupling degree calculation formula”, and then the use of that formula, is not at all clear; nor is it clear where this comes from (other than from “physics”), nor why it is chosen.  There seems to be no explanation for how the values of α, β, and γ are chosen (other than that one other paper chose them), nor any discussion of robustness tests.  The measures IOFDI, T, DIO, and CIO are not clearly defined nor differentiated from each other.

And regarding the results, all the tables seem to show positive and significant coefficients for the impact of “a stronger interaction between the two-way FDI” (I think) and the two CO2 measures.  Of course the discussion reports a negative impact.  What am I missing?

Smaller points:

The authors do not seem to be aware of the fact that the Heritage Foundation index of country-level freedom is quite controversial, as it is clearly based on freedom FOR BUSINESSES rather than any other type.  For example, as I understand it, a country’s freedom index is lowered if it has high tax rates, high levels of government spending, deficits, and debt, strong labor rights, strong financial regulation, and – most ironically for this paper – strong environmental regulations.  (See, for example, https://www.networkideas.org/news-analysis/2005/06/index-of-economic-freedom/.)  The authors acknowledge this to some degree but with loaded language such as “reasonable tax policies”.

Finally, a few problems with tables:

• Table 1 lists “East and Southeast Asia” twice;
• Table 2 doesn’t seem to differentiate between IFDI and OFDI; and
• Table 2 claims there are 987 observations for each variable, but the authors have earlier stated that "For instances of missing data, an interpolation method was employed to fill in the gaps."  It would be interesting to know which variables are most affected, and by how much.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript, "The Impact of the Coordinated Development of Two-way FDI along the Belt and Road Countries on Carbon Emissions," reveals current concerns in the field of sustainability and international investment analysis. However, after analyzing it, we considered that some aspects need to be improved as follows: 
1. The title of the article: "The Impact of the Coordinated Development of Two-way FDI along the Belt and Road Countries on Carbon Emissions" is concise and clear, but could be optimized to better highlight the method and specifics of the study. Thus, I believe that explicitly mentioning the STIRPAT model or the use of GMM methods would increase the visibility of the paper among readers interested in applied econometric methodologies. 
2. The abstract (page 1, lines 9–24) succeeds in capturing the essence of the research, presenting the objectives, methods, and major conclusions, however, the wording is too technical and full of complex concepts (e.g., "tripartite mechanisms-scale expansion, technological diffusion, and structural transformation"), which may make it difficult for non-specialist readers to understand. I would recommend introducing a more accessible introductory sentence that highlights the practical importance of the study for environmental and development policies.
3. The list of keywords (page 1, lines 25–26) is well-oriented but insufficient; terms such as "Belt and Road" and "carbon emissions" are included, but some concepts essential to the methodology and context are missing. I would recommend adding "STIRPAT model," "GMM regression," and "empirical analysis," as these reflect the analytical tools and strategies used. In addition, an expression such as "sustainable investment" would better link the study to the field of sustainable development.
4. The introduction (page 2, lines 28–44) describes the general framework of the Belt and Road Initiative in detail and emphasizes aspects of cooperation and sustainable development, but the text is predominantly narrative and does not sufficiently highlight the research problem. The authors should explicitly formulate the "literature gap" and specify the main question that the study answers.
5. The literature review (page 2, lines 54–65) is more descriptive than critical, so I would recommend, in order to add value, including a comparative synthesis highlighting where the results of previous studies contradict each other and what gaps remain. For example, if some research supports the "pollution haven" effect and others confirm the "pollution halo" hypothesis, it should be explicitly stated what context determines the differences.
6. Literature Review:
- The "pollution haven" hypothesis (page 4, lines 136–141) is correctly mentioned, but it is presented only at a theoretical level, so in order to convince the reader, concrete examples from the Belt and Road context would be necessary. For example, a case study on the relocation of the petrochemical industry in Pakistan or Vietnam could strengthen the argument. In addition, the authors could discuss whether this phenomenon has been observed in all countries or only in those with weak regulations.
- The "pollution halo" hypothesis (page 4, lines 152–159) is presented, but it is unclear whether the authors validate it in their own analysis or use it only as a theoretical background. I recommend that the transition from the literature to the authors' own hypotheses be made more explicit. For example, it would be useful to specify that the article will directly test whether investments attract green technologies or, on the contrary, lead to increased emissions.
7. The theoretical section (pages 6–7, lines 239–290) explains the effects of scale, technology, and structure on carbon emissions. Although the explanation is complete, the density of the text makes it difficult for a non-specialist reader to follow. My suggestion would be to introduce a figure or conceptual diagram to visualize the mechanisms and causal relationships between FDI, the economy, and emissions.
8. Hypotheses H1–H3 (page 7, lines 323–328) are clearly presented, but they do not discuss the possibility of contradictory effects. For example, in H2, technology is considered a factor that increases emissions intensity, but the literature also mentions cases where green innovations have reduced emissions. It would be advisable to mention this duality and explain why, in the context of the BRI, one effect or the other is more likely to prevail. In addition, the hypotheses could be formulated in a more "operationalizable" way, explicitly specifying the variables to be tested.
9. Data and sample (page 8, lines 330–338): the authors use a sample of 47 Belt and Road countries, but do not explain in detail why certain countries were included and others excluded. Clarifying the selection criteria (e.g., data completeness, economic relevance, statistical accessibility) would strengthen the credibility of the analysis. In addition, the interpolation of missing data is mentioned, but without describing the associated methodological risks.
10. The definition of variables (page 9, lines 341–364) is well done, but they are presented in the text in a way that can make reading difficult. My recommendation is to introduce a separate summary table listing all variables, their exact definitions, and data sources. For example, for "technological innovation," it would be useful to clearly state that it is measured by the number of articles published, with reference to the database used.
11. The basic results (page 11, lines 434–445) are interesting and confirm the hypothesis regarding the positive impact of FDI coordination on reducing emissions intensity, but the numerical interpretation is brief and does not provide concrete examples. The authors could add a practical illustration, for example, explaining what a 0.03% reduction means in terms of emissions for an average country in the sample. In addition, it would be useful to mention the confidence intervals and the economic (not just statistical) significance of the results.
12. The section on mechanisms (pages 15–16, lines 569–605) shows that technological innovation has a paradoxical effect, increasing the intensity of emissions. This is an important finding, but the explanation provided is rather superficial, with the authors merely suggesting that BRI countries have limited innovation capacity and prioritize economic growth. It would be desirable to provide additional references confirming the rebound effect in other contexts, and it would also be useful to discuss the differences between types of innovation (green technologies vs. expansion technologies).
13. The analysis by region (pages 18–19, lines 663–679) is valuable because it highlights the heterogeneity of the effects, but the numerical presentation in the table could be supplemented by comparative graphs or maps, which would make the results more visible and easier to interpret. For example, a comparative bar chart between South Asia and Central Europe would clearly show the differences in the impact of FDI coordination. In addition, the authors could add an explanation of the structural or institutional factors that explain regional variations.
14. Conclusions and recommendations (pages 19–20, lines 680–686) summarize the main results but do not include a discussion of the limitations of the research. This is essential for credibility; for example, interpolation of missing data, possible collinearity issues between variables, or difficulties in accurately measuring technological innovation are methodological limitations that should be acknowledged.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is much improved in both content and clarity, and I appreciate the authors’ hard work in both aspects.  I have just a few more concerns that I believe can be addressed easily.

First, I remain a bit puzzled regarding the theoretical or logical explanation for why coordination of incoming and outgoing FDI is important for reducing carbon emissions.  As I asked in my previous comments, whatever the impact of incoming FDI, why is that affected by having some kind of “balance” of outgoing FDI?  One possible explanation would relate to the pollution haven story:  given the positive or negative impact of incoming FDI, the “exporting” of pollution-intensive industries has a beneficial (local) impact.  But I am curious as to what the authors or the literature might think.

This is relevant to my suggestion that Table 6 receive more attention and discussion, in particular equation 11, which seems to suggest that while both incoming and outgoing FDI have an impact, the impact of the former is much greater (and perhaps the interaction, though statistically significant, does not add much).  This MAY be related – I cannot tell – to the statement on p. 20 that “the direct pathway contributes 73.6% and the indirect route through industrial upgrading accounts for 26.4%”.  Where do we find those figures?

Regarding the Tables, two points:

• While the sign of the coefficients is corrected in Table 4 (not 3), it appears that the signs of the coefficients in Tables 6, 7, and 11 remain “wrong”.
• Regarding Table 5, please explain what “L.lnCO₂” is and why it is there.

Finally, for H₂, should “greater” be “lesser”?

Thank you and best wishes with the revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript has undergone substantial revisions based on the feedback received during the evaluation process. I believe the current version addresses all major concerns and meets the standards required for publication

Author Response

Thanks for your positive comments.