Foreign Direct Investment and Carbon Emission Efficiency: The Role of Direct and Indirect Channels
Round 1
Reviewer 1 Report
The paper is well-written.
1. Sentences at line 210, 121, 103, 105 necessite adding references for them.Check carefully the text.
2. The FDI affected environment and sustainability negatively according to the findings and this can be understood from the abstract but in a very hard way.It should be clearly stated in the abstract. The abstract, if read by readers, might be understood as FDI did not harm the environment which is not what you mean. I can understand that this is also what you meant, we are at the same line. But I am worried that the abstract if misleading for young researchers especially.
3. In section 2,subheadings concentrate on FDI. This is ok since it is central. But, production is also central and that literature is missing. Add a sub heading under heading 2 as you did: production and emissions. Discuss linear and nonlinear empirics that provided novel findings with this regard. Also, I suggest addition of
The nonlinear relationship of environmental degradation and income for the 1870-2011 period in selected developed countries: the dynamic panel-STAR approach
Markov-switching vector autoregressive neural networks and sensitivity analysis of environment, economic growth and petrol prices
4. I suggest figures for results obtained for CEE at Table 2. This will increase visual representation of the findings here.
5. Include datasources column to Table 1. It is missing.
6. At Table 2, the last column heading (year) is in Chineese. Revise.
7. Lines between 326 and afterwards has wrong spacing.
8. Unit root test results are missing. They should be added as table. If variables are not stationary, then the whole econometric results should be recalculated. Correct afterwards after adding unit root tests (1st and 2nd generation). Conduct homogeneity test for the panel. Conduct cross-sectional dependence test for panel. Revise the empirical application necessarily by adding tables for these tests. They are necessary. These tests cannot be avoided in empirical research. This is the most major problem in the paper.
9. Table 5 will be revised after the critique 8 above.
10. Tobit regression is a binary dependent model is it not? Tobit results cannot be compared with fixed effects regression results. These are different models and are not nested models. Remove the robustness Table.
11. Findings and conclusion clearly show the positive effects of FDI increases on environmental hazardous emission increases. Conclusion and abstract should emphasize this more. (As I noted before. This time for conclusion).
12. Regarding comments 8-11: These comments should be considered together. The paper has a major problem. No unit root tests (first and second gen. tests including Pesaran's CADF and Maddala-Wu test), no crosssectional dependence tests, no homogeneity tests are conducted. As a result of this, the estimators used are not prune to these mistakes. The estimations should be done with a cross-sectional dependence robust estimator that is also consistent in terms of heterogeneity. I assumed that these tests will reveal that, which shuld be expected for regional analysis such as this paper. If not, this is another concern. How can two regions of China can be independent? How can the data such as this be homogeneous? Therefore, conduct these tests carefully and interpret the results carefully. Present the results. According to the test results, the empirical section will be changed. As a result, authors should also revise the abstract and conclusion and all discussions and comments related to the econometric results of this paper.
Author Response
Reviewer 1
The paper is well-written.
Point 1: Sentences at line 210, 121, 103, 105 necessite adding references for them.Check carefully the text.
Response 1: Thank you very much for the careful review of the experts. We add new references 6,8 at line 103, 105, respectively. The content at line 121 is from Chu et al.(2022) (at line 116). The content at line 210 is from Malik et al. (2020) (at line 209).
Point 2: The FDI affected environment and sustainability negatively according to the findings and this can be understood from the abstract but in a very hard way. It should be clearly stated in the abstract. The abstract, if read by readers, might be understood as FDI did not harm the environment which is not what you mean. I can understand that this is also what you meant, we are at the same line. But I am worried that the abstract if misleading for young researchers especially.
Response 2: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, the paper rewrites the abstract, and makes it clearly for readers.
Point 3: In section 2, subheadings concentrate on FDI. This is ok since it is central. But, production is also central and that literature is missing. Add a sub heading under heading 2 as you did: production and emissions. Discuss linear and nonlinear empirics that provided novel findings with this regard. Also, I suggest addition of
The nonlinear relationship of environmental degradation and income for the 1870-2011 period in selected developed countries: the dynamic panel-STAR approach
Markov-switching vector autoregressive neural networks and sensitivity analysis of environment, economic growth and petrol prices
Response 3: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, the paper adds a sub heading under heading 2: production and emissions, discussing linear and nonlinear empirics that provided novel findings with this regard. We are very grateful to the expert for providing high-quality articles related to this paper, and cite these articles to improve the meaning and writing.
Point 4: I suggest figures for results obtained for CEE at Table 2. This will increase visual representation of the findings here.
Response 4: Thank you very much for the careful review of the experts. In the new manuscript, Table 2 is Table 3 because we add a Table in the introduction according to the reviewer’s suggestions. We use both figure (Fig.1) and table (Table 3) to increase visual representation of the findings.
Point 5: Include data sources column to Table 1. It is missing.
Response 5: Thank you very much for the careful review of the experts. In the new manuscript, Table 1 is Table 2 because we add a Table in the introduction according to the reviewer’s suggestions. We add data sources column to Table 2 (Regional carbon emission efficiency index system).
Point 6: At Table 2, the last column heading (year) is in Chinese. Revise.
Response 6: Thank you very much for the careful review of the experts. We revise it.
Point 7: Lines between 326 and afterwards has wrong spacing.
Response 7: Thank you very much for the careful review of the experts. We adjust spacing between 326 and afterwards.
Point 8: Unit root test results are missing. They should be added as table. If variables are not stationary, then the whole econometric results should be recalculated. Correct afterwards after adding unit root tests (1st and 2nd generation). Conduct homogeneity test for the panel. Conduct cross-sectional dependence test for panel. Revise the empirical application necessarily by adding tables for these tests. They are necessary. These tests cannot be avoided in empirical research. This is the most major problem in the paper.
Response 8: Thank you very much for the careful review of the experts. At first, we conduct both the first and second generation-based panel data unit root testing, and the results showed that each variable series is stationary. In the process of modification, we read a large number of literature. According to Ahmad et al. [2020], the first generation-based panel data unit root testing procedures fail to take into account the cross-sectional dependence, while the second generation tests involve the cross-sectional dependency. Taking into consideration the results in Table 6, we present Maddala-Wu test results. We conduct both homogeneity test and cross-sectional dependence test for the panel, and revise the empirical application necessarily by adding tables for these tests.
Point 9: Table 5 will be revised after the critique 8 above.
Response 9: Thank you very much for the careful review of the experts. In the new manuscript, Table 5 is Table 8 because we add 3 tables in preceding sections according to the reviewer’s suggestions. Table 5 is revised after the critique 8 above.
Point 10: Tobit regression is a binary dependent model is it not? Tobit results cannot be compared with fixed effects regression results. These are different models and are not nested models. Remove the robustness Table.
Response 10: Thank you very much for the careful review of the experts. We remove the old robustness Table. Since a reviewer thinks that robustness analysis should be carried out to authenticate the results, we use the sample size changing method to conduct a robustness test using the same method.
Point 11: Findings and conclusion clearly show the positive effects of FDI increases on environmental hazardous emission increases. Conclusion and abstract should emphasize this more. (As I noted before. This time for conclusion).
Response 11: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, we rewrite the abstract and the conclusion, and show the results clearly.
Point 12: Regarding comments 8-11: These comments should be considered together. The paper has a major problem. No unit root tests (first and second gen. tests including Pesaran's CADF and Maddala-Wu test), no crosssectional dependence tests, no homogeneity tests are conducted. As a result of this, the estimators used are not prune to these mistakes. The estimations should be done with a cross-sectional dependence robust estimator that is also consistent in terms of heterogeneity. I assumed that these tests will reveal that, which shuld be expected for regional analysis such as this paper. If not, this is another concern. How can two regions of China can be independent? How can the data such as this be homogeneous? Therefore, conduct these tests carefully and interpret the results carefully. Present the results. According to the test results, the empirical section will be changed. As a result, authors should also revise the abstract and conclusion and all discussions and comments related to the econometric results of this paper.
Response 12: Thank you very much for the careful review of the experts. At first, we conduct both the first and second generation-based panel data unit root testing, and the results showed that each variable series is stationary. In the process of modification, we read a large number of literatures. According to Ahmad et al. [2020], the first generation-based panel data unit root testing procedures fail to take into account the cross-sectional dependence, while the second generation tests involve the cross-sectional dependency. Taking into consideration the results in Table 6, we present Maddala-Wu test results. The results of unit root tests, cross sectional dependence tests, homogeneity tests are presented in different tables. According to the test results, the abstract，the empirical section，the conclusion and all discussions and comments related to the econometric results of this paper are changed.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The article titled “FDI, technological innovation and carbon emission efficiency” is very interesting, very well written, and informative. However, the author (s) must incorporate the following issues to improve the article.
1. I could hardly understand the marginal contribution of this paper, as the authors review the literature in many aspect, which do not serve one research question. The authors must explicitly narrate the contribution/novelty of the study in one concrete paragraph in introduction section.
2. The second main issue is the base or theoretical foundation of the study. On which particular theoretical or conceptual framework the topic/study is investigated. Please explain how equations 2, 3 & 4 are constructed?
3. Please revised the sentence in abstract which is written as
“Based on the research, we offer advices and suggestions on 16 expanding FDI inflow, increasing technological innovation, improving carbon emission efficiency 17 and reducing carbon emissions”.
4. Please provide a table of recent literature relevant to topic in chronological order.
5. The results are meaningful but robustness or sensitivity analysis should be carried out to authenticate the results.
6. Discussion of main results with contextualization i.e. consistency or contradiction with prior studies/events is missing. The author (s) needs to provide contextualization to the main findings.
7. The policy implications should be drawn from obtained results and should be precisely linked with your study findings.
Comments for author File: 
Comments.pdf
Author Response
Reviewer 2
The article titled “FDI, technological innovation and carbon emission efficiency” is very interesting, very well written, and informative. However, the author (s) must incorporate the following issues to improve the article.
Point 1: I could hardly understand the marginal contribution of this paper, as the authors review the literature in many aspect, which do not serve one research question. The authors must explicitly narrate the contribution/novelty of the study in one concrete paragraph in introduction section.
Response 1: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, we explicitly narrate the contribution/novelty of the study in one concrete paragraph in introduction section. The modified part has been expressed in red font.
Point 2: The second main issue is the base or theoretical foundation of the study. On which particular theoretical or conceptual framework the topic/study is investigated. Please explain how equations 2, 3 & 4 are constructed?
Response 2: Thank you very much for the careful review of the experts. The base of this study is the effect of foreign capital flow on environment in the host country. Foreign enterprises have more green technologies and environmental management systems than local enterprises. Local enterprises can absorb and obtain advanced green technologies and management experience from foreign enterprises through imitation effects and reverse engineering effects, so as to improve carbon emission efficiency. In order to meet foreign high environmental standards, foreign enterprises usually pay more attention to skills training and environmental awareness training for employees. With the flow of labor among enterprises, some employees of foreign enterprises may flow to local enterprises, bringing green technology and management experience to local enterprises, thus improving carbon emission efficiency.
Since a reviewer thinks that the econometric model should include the linear effects (X) and the interactions of X with the independent variable of interest, we rewrite the econometric models. Equations (2), (3) and (4) are equations (2), (3) now, and we explain how equations (2), (3) are constructed. The modified part has been expressed in red font.
Point 3: Please revised the sentence in abstract which is written as “Based on the research, we offer advices and suggestions on 16 expanding FDI inflow, increasing technological innovation, improving carbon emission efficiency 17 and reducing carbon emissions”.
Response 3: Thank you very much for the careful review of the experts. We revise the sentence in abstract.
Point 4: Please provide a table of recent literature relevant to topic in chronological order.
Response 4: Thank you very much for the careful review of the experts. The recent literature relevant to topic in chronological order is given in Table 1. There are four topics listed in Table 1 and each of four topics has some relevant literature presented in chronological order.
Point 5: The results are meaningful but robustness or sensitivity analysis should be carried out to authenticate the results.
Response 5: Thank you very much for the careful review of the experts. Robustness analysis is carried out to authenticate the results in Section 4. The sample size changing method is adopted. Since the financial crisis occurred in 2008 and had a vital impact on global economy and investment, the sample data may be inconsistent before and after 2008. We use the sample data from 2009 to 2019 to conduct robustness analysis and the results are shown in Table 10.
Point 6: Discussion of main results with contextualization i.e. consistency or contradiction with prior studies/events is missing. The author (s) needs to provide contextualization to the main findings.
Response 6: Thank you very much for the careful review of the experts. We add 4.6 Discussion in order to provide contextualization to the main findings. The similarities and differences between this study and prior studies are investigated and presented in 4.6.
Point 7: The policy implications should be drawn from obtained results and should be precisely linked with your study findings.
Response 7: Thank you very much for the careful review of the experts. The paper removes the original policy recommendations and rewrites the policy recommendations based on the conclusions of the empirical analysis to ensure that the conclusions correspond to the policies. The modified section has been expressed in red font.
Author Response File: Author Response.pdf
Reviewer 3 Report
Please refer to the attached report. I strongly encourage the authors to incorporate each of the comments outlined.
Comments for author File: Comments.pdf
Author Response
Reviewer 3
Point 1: I urge the authors to account for the relevant macroeconomic forces of the sample period as follows:
â‹„ Clearly describe the relevant forces, with references to the relevant literature (which I outline in this report).
â‹„ For any macroeconomic variable Xt , the authors can augment the econometric models (introduced in sub-section 3.3) to include:
– Their linear effects (Xt).
– The interactions of X with the independent variable of interest (i.e., fdii,t × Xt). The macroeconomic variables Xt that the authors should include are the following. For each of these variables, the justiffcation for their inclusion (together with the interaction terms) is provided below:
â‹„ Increasing popularity of CSR/ESG investing:
– The volume of assets under management by investment ffrms that follow ESG criteria in their asset selection increased considerably in the past decade (please refer to Dantas, 2021). This creates direct and indirect incentives for the development of technologies that improve the efffciency of energy sources that emit carbon. Thus, the authors should consider the growth in global assets under management by ESG funds as one of the macroeconomic variables (linear control and interaction with fdii,t).
â‹„ Monetary policies conducted by major central banks—in particular, the US Federal Reserve:
– The sample period comprises the Global Financial Crisis (2008–2009), the Eurozone Sovereign Debt Crisis (2010–2013), and the potential increase in the systemic risk of Chinese banks (see https: //vlab.stern.nyu.edu/srisk). As such, central banks around the world put forth various rounds of quantitative easing (Cortes et al., 2022). Quantitative easing policies by central banks of developed economies, in particular the US Fed, are important because they distort equilibrium values of asset prices, tail risks, exchange rates, and FDI (Cortes et al., 2022; Dedola et al., 2020). As shown in Cortes et al. (2022), spillovers from the Fed’s QE policies can be particularly detrimental for emerging market economies (such as China). The authors should consider the size of the Fed’s assets (data found at https://fred.stlouisfed.org/series/WALCL) as one of the Xt
variables in question.
â‹„ Fiscal deffcits:
– The authors mention that “ffscal expenditures” (finei,t) are already included as one of the control variables (description of control variables in subsection 3.3). However, the manuscript is silent about a central reason why a variable proxying for ffscal conditions must be included in the model as a control: deffcits affect the strength of government spending multipliers and the aggregate credit risk of the banking sector (Silva, 2021). Needless to say, all these channels would affect the levels of FDI and capital investments, which in turn affect technological development and the efffcient utilization of energy sources. I encourage the authors to outline these arguments in the passage they introduce variable finei,t (with reference to Silva, 2021). Moreover, finei,t is also a variable to be interacted with fdii,t (fdii,t × finei,t).
â‹„ Global economic policy uncertainty:
– In addition to economic crises, major geopolitical events occurred during the sample period in question (Brexit, military confficts, etc.). Heightened policy uncertainty affects FDI through several channels (Azzimonti, 2018, 2019). Hence, I encourage the authors to consider the global version of the economic policy uncertainty index proposed by Baker et al. (2016) as the ffnal macroeconomic proxy to be included as a control (together with its interaction with fdii,t). The measure can be found at https://www.policyuncertainty.com/.
Editorial comments. I would consider changing the paper title to emphasize that FDI is the “independent variable” and carbon emission efffciency is the “dependent variable” of interest, while technological innovation is just one (indirect) channel through which the independent variable affects the dependent variable. One suggestion is “Foreign Direct Investment and Carbon Emission Efffciency: The Role of Direct and Indirect Channels.” Lastly, I encourage the authors to avoid acronyms in the paper’s title (i.e., call it “Foreign Direct Investment” as opposed to “FDI”). An academic paper needs to unambiguously introduce the terminology in the body of the manuscript, and using the term FDI in the text implicitly assumes that the reader is familiar with it, which may not be the case should this paper reach a broader readership.
References
Amiti, M., M. Dai, R. C. Feenstra, and J. Romalis (2017). How did china’s wto entry benefft us consumers?
NBER Working Paper. DOI: https://doi.org/10.3386/w23487.
Azzimonti, M. (2018). The Politics of FDI Expropriation. International Economic Review 59(2), 479–510. DOI: https://doi.org/10.1111/iere.12277.
Azzimonti, M. (2019). Does Partisan Conffict Deter FDI Inffows to the US? Journal of International Economics 120, 162–178. DOI: https://doi.org/10.1016/j.jinteco.2019.06.001.
Baker, S. R., N. Bloom, and S. J. Davis (2016). Measuring Economic Policy Uncertainty. The Quarterly Journal of Economics 131(4), 1593–1636. https://doi.org/10.1093/qje/qjw024.
Cortes, G. S., G. P. Gao, F. B. Silva, and Z. Song (2022). Unconventional Monetary Policy and Disaster Risk: Evidence from the Subprime and COVID–19 Crises. Journal of International Money and Finance 122, 102543. DOI: https://doi.org/10.1016/j.jimonfin.2021.102543.
Dantas, M. (2021). Are ESG Funds More Transparent? Available at SSRN 3269939. DOI: http: //dx.doi.org/10.2139/ssrn.3269939.
Dedola, L., G. Georgiadis, J. Gräb, and A. Mehl (2020). Does a Big Bazooka Matter? Quantitative Easing Policies and Exchange Rates. Journal of Monetary Economics. https://doi.org/10.1016/ j.jmoneco.2020.03.002.
Silva, F. B. G. (2021). Fiscal Deffcits, Bank Credit Risk, and Loan-Loss Provisions. Journal of Financial and Quantitative Analysis 56(5), 1537–1589. DOI: https://doi.org/10.1017/S0022109020000472.
Response 1: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, we rewrite the econometric models. The econometric model includes the linear effects (Xt) and the interactions of Xt with the independent variable of interest in 3.3 Econometric models.
We clearly describe the relevant forces, with references to the relevant literature. We are very grateful to the expert for providing high-quality articles related to this paper, and cite these articles to improve the meaning and writing.
We use the new model to conduct regression analysis, and we rewrite Section 4 Results. According to the test results, the abstract，the introduction，the conclusion and the discussions related to the econometric results of this paper are changed.
We change the title and the new title is “Foreign Direct Investment and Carbon Emission Efffciency: The Role of Direct and Indirect Channels.” We avoid acronyms in the paper’s title, and use Foreign Direct Investment instead of FDI in this study.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors carefully adressed all the critiques. Thankful for this. I believe that this final version is ready for publish and I am happy to see this final version.
I have two very minor notes which can be adressed during proof checks and for these, no third round is needed.
1. I see that the equations are larger in size now in terms of font. They are large in the text in some places and also in the regressions. Mostly, this minor problem is evident for a formula in a paragraph. An overlook will easily show these.
2. In conclusion, the first paragraph goes into the findings a little fast. Add an overlook to the paper here similar to the abstract. An additional 1 or 2 sentences at the first paragraph will improve the conclusion for the readers who jump reading directly the conclusion in the future. This final note is with this regard, i.e., for the readers who read conclusions to get an idea about the paper.
Author Response
Reviewer 1
Point 1: I see that the equations are larger in size now in terms of font. They are large in the text in some places and also in the regressions. Mostly, this minor problem is evident for a formula in a paragraph. An overlook will easily show these.
Response 1: Thank you very much for the careful review of the experts. We modify the size of equations, which is uniform with the text.
Point 2: In conclusion, the first paragraph goes into the findings a little fast. Add an overlook to the paper here similar to the abstract. An additional 1 or 2 sentences at the first paragraph will improve the conclusion for the readers who jump reading directly the conclusion in the future. This final note is with this regard, i.e., for the readers who read conclusions to get an idea about the paper.
Response 2: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, we add an overlook to the paper similar to the abstract in Section 5, and make it clear for readers. The modified part has been expressed in red font.
Author Response File: Author Response.docx
Reviewer 2 Report
1. I found some grammar and syntax errors. Please improve the overall write up of manuscript.
2. The literature has ignored some recent studies. Please include the recent studies, for instance;
i. https://doi.org/10.1016/j.energy.2021.122703
ii. https://doi.org/10.1371/journal.pone.0230503
Author Response
Reviewer 2
Point 1: I found some grammar and syntax errors. Please improve the overall write up of manuscript.
Response 1: Thank you very much for the careful review of the experts. According to the reviewer’s suggestions, grammar and syntax errors are corrected throughout the paper. The modified part has been expressed in red font.
Point 2: The literature has ignored some recent studies. Please include the recent studies, for instance;
- https://doi.org/10.1016/j.energy.2021.122703
- https://doi.org/10.1371/journal.pone.0230503
Response 2: Thank you very much for the careful review of the experts. We are very grateful to the expert for providing high-quality articles related to this paper, and cite these articles to improve the meaning and writing.
Author Response File: Author Response.docx
