The Impact of Financial Development and Green Finance on Regional Energy Intensity: New Evidence from 30 Chinese Provinces
Round 1
Reviewer 1 Report
The paper is the result of more extensive work. I recommend publishing after a major revision.
In the paper, I miss the connection between "words formulated by hypotheses H1-H3" and mathematical models (1-4). The statistical formulations of the hypotheses that cover H1-H4 should be rigorously stated.
State the statistical formulation of H1 to H4 in the theoretical part. Complete the mathematical expression of H1 - H4, and test statistics, and their critical domains. Cite their derivation from statistical publications or derive it yourself in the appendix.
The explanation of the variables in section 3.2 follows only after the models in section 3.1 where these variables are used. This isn't very pleasant for the reader. Necessary to rearrange as part of the revision.
In the article, there is an insufficient citation of articles from the field of statistics, especially on lag models.
The explanation of "neighborhood" and the construction of the matrix W entering the lag model is missing.
I'm missing the formula for the LM test, etc.
The theoretical part for spatial regression modeling is missing, or it is written very superficially. The authors have slipped in the presentation of the results when they do not present the article's readers with sufficient material on the lag issue.
This weak point of the article needs to be strengthened.
----------------------------------------------
List of proofreadings:
Line 219
The reference is only to models (2-4), but not to model (1).
Line 220, etc.
Model (1), (2), (3), (4) Do not use bold font. Bold font only for matrix W.
Line 228
What are the assumptions on the error term?
Line 275 [Do not use bold font in formulas (5) and (6). I recommend rewording text like this:]
The first way to normalize indicators is given by equation (5). The second variant of normalization is possible according to equation (6). In further calculations, we will use formula (5).
Line 233
for the {hypothesis} H3
Line 236
Model (4). Do not use bold font. What is Phi? What are the assumptions on the error term?
Line 275 [Do not use bold font in formulas (5) and (6). I recommend rewording text like this:]
The first way to normalize indicators is given by equation (5). The second variant of normalization is possible according to equation (6). In further calculations, we will use formula (5).
Line 320
The variance seems to me to be too much for HC and too little for FT and SDE. Please check the correctness of the characteristics.
eq. (7) Do not use bold font.
eq. (8) Do not use bold font.
eq. (9) Do not use bold font.
eq. (10) Do not use bold font for W_{ij}. Bold only the matrix W.
eq. (11) Do not use bold font. Add symbol. after the last equation.
Line 348
W … known spatial weight matrix, where W_ij=1, if the boundary between i-th and j-th region exists, W_ij=0 if there is no boundary
Line 329
There is a lack of theory for LAG models. Here the authors move on to the presentation of the data.
Add citations to models:
Spatial Lag Model (SAR), Spatial Error Model (SEM), Spatially-lagged
X Model (SLX), Spatial Durbin Model (SDM), Spatial Durbin Error Model (SDEM), and Kelejian-Prucha Model (SAC).
For example, you can work with publications:
Anselin, L. (2001). Spatial Econometrics, 14th chapter, A Companion to Theoretical Econometrics, Copyright by Blackwell Publishing Ltd, 310-330.
Ansein, L., Rey, S. (1991). Properties of Tests for Spatial Dependence in Linear Regression. Geographical analysis 23, 112–131.
Moran, P. A. P. (1950). Notes on Continuous Stochastic Phenomena. Biometrika 37 (1), 17–23. DOI:10.2307/2332142. JSTOR 2332142. http://www-hsc.usc.edu/~mereditf/Spatial%20Statistics%206.pdf
Add the LM test for spatial error autocorrelation (Anselin, 2001).
Line 327 y_i, y_theta, i, 1,...,30, not bold font
Line 328 R, 1,...,30 not bold font.
Line 358 Explain the p-value calculation. Is it given by the chi-square with one degree of freedom? Add citation.
Line 378 Add citation of tests (LR, LM, Robust LM, Hausman Wald) and their basic ideas to the theoretical part of your paper.
Line 398, 478, 518, 534, 557 and 579
How do the results sound about hypotheses H1, H2, H3, and H4?
State the statistical formulation of H1 to H4 in the theoretical part. Complete the test statistics and their critical domains
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper applied the spatial econometric model, and the panel threshold model to investigate the mechanisms of financial development and green finance on regional energy intensity. I think some conclusions of this paper are interesting. But the research can be improved in the following aspects.
1. In line 30 page 1, the Chinese government proposed that China would strive to peak carbon emissions by 2030 and carbon neutrality by 2060, not 2021.
2. In paragraph 1 page 2, you authors should better cite some references for the views on China's financial market.
3. In line 198 page 4, the opinion of the increase in regional energy intensity should better cite some references. The same opinion appeared in page 13 again, but I can not understand this conclusion based on the results in Table 4.
4. In section 3.1.1, the number of models should be "model 1-model 3", and in section 3.1.2,"The following model 6" should be model 4.
5. The Chinese map in figure 4 should be kept intact.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This is a well written article that can contribute to the existing knowledge. I have three bigger issues and several smaller.
First, The article would benefit from more through literature review. For now literature is used to support hypothesis development and methodology in general, but we do not learn about what has been written in the area from a broader perspective.
Second, because of this, there is no discussion with the existing scholarship at the end of the article that would examine the results in light of existing words.
Third, as a result of this, the article and its conclusions are focused on the issue at hand, but we do not learn what all of this means from a broader perspective. The dilemma between growth and pollution has been examined broadly, but what is the contribution of this paper to this discussion?
I am wondering if this is correct: „China's government, as a significant energy user, has suggested achieving "carbon neutrality" and "carbon peaking" by 2021 [3] (lines 29-30). Those two concepts are totally different and while the latter could be true for 2021, it is not true for the former.
Line 41: Is this correct? “As the financial industry has grown“ – because the text does not talk about financial sector. Should it be “Industry” only? The introduction is not very clear about when it talks about “financial industry” as a financial sector and when it talks about industry as such.
The introduction presents research design only in one sentence. More needs to be added, plus the introduction will need to be more clear about which three financial variables the paper studies – they introduction has a broad start, it mentions a lot of things, but it is not very clear which of these things are general introduction (a background to the problem) and which are connected to the research design of the paper.
H1 and H2 include word “dramatically” – I do not think it is necessary here because the result are supposed to say if the increase/decrease was “dramatic”, or to use better word “sharp”, “considerable”, etc. (to avoid “significant” not to create confusion with the statistical results. H3 is not really worded as a H.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Thank you for responding to my comments. The edits improved the paper sufficiently.
I found a few minor typos:
(4) ln m, ln(m)
(6) do not write "W12", ... , "Wij", "1/abs(y_i-y_j)" , "1", "i=j" in bold
(9) splitting a long formula is necessary
(11) do not write the equation in bold
(13) in the numerator of the fraction "I-E(I)"
(16), (17) vector e write in bold, vector y write in bold
(16), (17) do not write equation in bold
Formulas (13), (14), (17) : The fact that a random variable follows, for example, a normal distribution, is usually written using an arrow with n->infinity above it.
Use "N->infinity" instead of "i" here, because you are working with a random sample of length N.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Dear Authors,
thank you for putting so much energy into revising your paper. I would just like to suggest one very minor issue, that I would like to leave up to you if you wish to revise: the section "7. Possible Research Contributions and Shortcomings" lists contribution of your research to our knowledge, but does not put it into the context of the previous studies. So, we do not know what was known before your results added a new layer of knowledge. I think that this section would greatly benefit from discussing your results with existing literature (as I have suggested already in my previous review, but I guess my wording was not clear - sorry about that).
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
