Does Global Value Chain Embedment Contribute to Environmental Pollution in Emerging Economies?

Round 1

Reviewer 1 Report

Congratulations! This is a good paper.

Author Response

Thank you for your high evaluation, and we will continue to work hard in the future.

Reviewer 2 Report

Dear Author(s):

I had the opportunity to review your paper entitled "Does global value chain embedding contribute to environmental pollution in emerging economies?" submitted to the Journal of Sustainability. In a quantitative study of 11 emerging economies from 1998 to 2019, I interpret your findings as support for two hypotheses, although both findings should be rephrased as a single inverted u-shaped relationship (see below). Overall, the study is interesting and your methods mostly fine. However, you need to improve your writing considerably to avoid misinterpretation of your findings. Hopefully, my remarks will help you to improve the paper’s presentation of core findings. Best of luck!

1.     First and foremost, the level of scientific rigor is low. There are too many grammar-related mistakes distracting from your findings. Moreover, there is a severe degree of inconsistency in style. Sometimes you use empty spaces, sometimes you don’t. Empty spaces should never appear BEFORE full stops or commas but ALWAYS after. The bad English makes it very hard to follow your strain of thoughts. I caught myself quite often guessing what you might mean due to a lack in precision of scientific terms. It is therefore most important for you to apply a professional proofreading and copy-editing service. A few general recommendations: Use shorter sentences (e.g., line 37: new sentence after “the world. Among”); use active instead of passive voice wherever possible (e.g., line 43: delete unnecessary fillers such as “it can be seen that”; line 399: “would be provided”); line 300: reduce header to “Hypotheses”; line 358: change “explanatory” (=x) to “explained (or dependent)” variable (=y); line 434: show full term before first use of its abbreviation (FGLS); line 455: “a significantLY negative correlation”; line 467: “aN inhibitory”; etc.  

2.     In line 43: do you mean “represented” instead of “representative”? If not, are these 11 countries truly representative of “average” emerging countries? Did you test for it or stick to an established definition of “representative emerging economies”? I checked footnote 2 and Korea does not seem to be an emerging economy. I suppose this is South Korea, correct? To my knowledge, South Korea is a developed economy with the highest per capita of patent applications in the world due to Samsung, LG, Huyndai, etc. Please show convincingly how you end up with these 11 economies. Typically, a panel analysis of just 11 cases over 22 years is a very small sample and potentially too small to partial out highly correlated effects. You may argue that your tested slopes are significantly different from zero. However, statistical significance should not be the only criteria. Effect sizes determine how important the mechanisms under study are. Do you have any global fit criteria of your DV, e.g., R-square within/between?         

3.     Your hypotheses are indefinite. When does “initially” end (H1)? What do you mean by deepening (H2)? Obviously, you are testing a single inverted u-shaped relationship, so you just need to postulate the negatively quadratic term. The linear term only shows whether there is a turning-point within the observed data or just decreasing marginal effects that might be more appropriately modeled via a negative inverse function (-1/x). Please test alternative fit functions and stick to common guidelines in testing u-shaped or moderated u-shaped relationships following Haans et al. (2016). It helps if you depict the non-linear relationship in a graph. Is there a maximum level of pollution within the observed data of GVC or not?        

4.     In testing for endogeneity using the instrumental approach, what instruments did you use? It seems you use only past values as instruments of future values of GVC. However, an important assumption is not met: The independence of the instrument with the residual/error term! Moreover, what is your exclusion criteria in the first stage of the estimation? The exclusion criterion needs to hold theoretical considerations apart from its statistical properties. How did you instrument the quadratic term? Did you use past values of the quadratic term? Please be aware of “forbidden regression” practices when dealing with non-linearity (see Haans et al. 2016).

5.     In general, your paper calls for a major revision. All the very best!     

References

Haans, R. F., Pieters, C., & He, Z. L. (2016). Thinking about U: Theorizing and testing U‐and inverted U‐shaped relationships in strategy research. Strategic management journal, 37(7), 1177-1195.

Author Response

Thank you for your detailed and constructive comments, especially for providing a rather critical reference. Here are our corresponding revisions.

1. We have enlisted the help of MDPI's professional language editors to make a overall check, especially in the parts you have listed.
2. We expanded the sample countries, for the definition of emerging economies, we used the BRIC countries and the Next-11 countries (presented by Goldman Sachs in 2006), which is a generally accepted research range. It is worth noting that not all emerging economies are developing countries, so South Korea is in this category. In addition, our regression model used FGLS, typically it does not report R-squre values.
3. According to the reference you recommended, we approximately calculated the turning poiot at which the direction chages. At the same time, we also used the methods mentioned in this reference to do robustness tests, including adding cubic terms and excluding extreme values.
4. It is a common practice to use one stage lag of the GVC index as an instrumental variable in the study of related problems, and its value is recognized in some of the literature I cited. In practice, the quadric term in the model also used the lag term, it is our fault that we did not explain it.

Thanks again for your hard work and we hope the revisions can be approved.

Reviewer 3 Report

The proposed article studies an important question  –  how global value chain embedding contributes to environmental pollution in emerging economies.

Interesting research has been done. The topic is also important for the readers of the journal. To continue reviewing, a revision should be done. I see the following issues:

1.    Language should be checked.

2.     Methods   should be described in a separate Section in the beginning.

3.    Authors should particularize research Hypothesis.

4.    Before the “Conclusion” separate paragraph “Discussion” should be added.

Resume. In my opinion, the article can be published after revision.

 

Author Response

Thanks for your work and valuable suggestions, we have made the following revisions:

1. We have sought the help of MDPI's professional language editor and made a comprehensive revision of the article.
2. We described the research methods used in the abstract section at the beginning.
3. We gave more descriptions of the research hypothesis and estimated the approximate turning point in the empirical study, which mentioned in hypothesis.
4. In the conclusion section, we added a discussion before the conclusion report.

Thanks again for your work and we hope the above revisions can be approved.

Round 2

Reviewer 2 Report

Dear Author(s):

I revied your paper for the second time. In part, you seem to have done a decent revision. However, you seem not to know how to write a response letter and respond to every question point by point, starting by repeating my questions and justifying your choices to a greater deal in the response letter by citing all relevant sources, not simply stating that “its value is recognized in some of the literature I cited.” Which literatures? Some of my open questions you seem to ignore. I do not believe you disrespect my valuable time on purpose. I guess you are not familiar with a rigorous review process. I had to check “the literature [you] cited” when referring to lagged values of the independent variables as an endogeneity treatment. In line 529, you cited Wang et al. [27]. I screened the entire document but could not find any information relating to “instrument” nor “lagged” values. The empirical study is not even using comparable methods. I honestly checked only one reference and spotted this error. I believe you are not fully aware what constitutes a truly exogenous instrument, are you? What about an exclusion criterion? Without a valid (!) exclusion criterion, the empirical results are worse than solutions without any instruments. Please consult literature on “weak vs. strong instruments” and on “exclusion criterion” when conducting regression analysis with instruments.

Please avoid unnecessary long sentences such as “indicating that the robustness of our conclusion was not changed by potential endogeneity” in line 542. You could shorten this phrase without losing any information “indicating endogeneity-robust findings.” The robustness cannot be changed. Your findings are either robust or not! It seems your new method section did not undergo a professional proofread. Please check all references for any mistakes and do not test my patience here. I expect your best effort. Good luck!

Author Response

Dear reviewer,

    Thanks for your hard work, and I’m sorry for wasting your time with my unnecessary mistakes. The following is my reply.

 

    (1) On the use of instrumental variable, I misquoted the reference, in fact, Zhong et al.(2021) used this approach in their paper “Does the participation in global value chains promote interregional carbon emissions transferring via trade? Evidence from 39 major economies.” That’s my 44^th reference, and I also agree that the lag term of GVC embedding index satisfies the selection principle, namely relevance and exogeneity.

    I’m very sorry for my lack of rigor.

    (2) I re-proofread the article, cut down some long sentences. With regard to your question in line 542, I think the language editors probably think that going from robust to not robust is also a change, I have modified it in the article.

 

    Thank you again for your work, and I hope my reply can be accepted.