Research on the Impact of Sustainable Urbanization on Urban Rural Income Disparity in China
Round 1
Reviewer 1 Report
Following should be addressed:
1) Most importantly, out of all comments, the similarity is too high. It is 41%. In an SCI indexed journal and paper, this is not acceptable. When checked, the first 2 references that are shown similar total to 10% and belong to authors' other papers. It could be thought that it is possible to have similarity to own papers, however, I should repeat my first concern. For and SCI indexed prestigious journal, this makes the paper unacceptable. Similarity should be reduced below 20%, or preferably, below 15%.
2) Descriptive statistics table is missing. Min, max, std dev., skew., kurtosis, JB test should be reported. Otherwise, the rest of the analysis become absolete.
3) The threshold estimates are too high. What is the percentage of observations above threshold and below threshold? The value for threshold is too high for a series after logarithms. Since descriptives for the variables USED in the models are not present, we can just guess but anyways, threshold is high. It should be checked.
4) Bootstrap replications are tested with "0 300 300." 300 is the default in the code in STATA for examples in the help file. Try higher replication values such as 10 000 for reliability.
5) Another concern: Bootstrap replications are tested with "0 300 300." This means, for first threshold no replications. The code is not used correctly. This makes Table 7 questionable. Change replications. First should also be high, not zero.
6) aLast paragraph of the conclusion sound as if the paper is not finished. Why did the authors not estimate models for which they developed research concerns? Further, regarding inverted U shape which was analyzed, no conclusion can be made.
7) Inverted U is not achieved. This also led to comment 6 above. The authors chose not to report a direct result with this respect. They wrote: "This paper focuses on whether there is an inverted U-shaped relationship between 404 the sustainable development of urbanization and the urban-rural income gap. However, 405 the selection of control variables in this empirical analysis model is based on the existing 406 research results. "
The sentence starting with "However" should discuss why inverted U is not achieved or is it achieved in this study. What is the relation of sentence 2 to sentence 1? The use of control variables caused the empirical findings below? Not very convincing since such models are not estimated by including and excluding control variables one by one.
The parameter estimates are
??????
4.351**
??????^2
-3.182***
This is clearly not an inverted U since the peak is reached immediately. Again, since we don't know the range of data, we cannot further comment on this if this turning point is plausable or not.
Ok, the panel regression did not work for inverted U, and isn't it the reason for the authors to shift to threshold panel regression in the first place? Why is this model not emphasized and conclusion is left with uncertain, uncompleted way?
7) In the conclusion, the last sentence is:
This paper does not analyze the influencing factors of the urban-rural 407 income gap. In the next stage, we will continue to study the relationship between the sus- 408 tainable development of urbanization and the income gap between urban and rural areas 409 in combination with the research on the factors affecting the income gap between urban 410 and rural areas.
Authors wrote that they did not investigate determinants of, or in another words, the effects of factors on urban - rural income gap. However, by looking at the title of the paper, we as readers think that we will read a paper on investigation of determinants of rural urban income gap.
8) I am sorry to say this but the wording / sentencing should be improved in the overall paper especially avoid direct statements such as X causes Y negatively. These should be tested empirically.
Most and foremost,
The 41% is a very high similarity. The authors should start by correcting this and also make corrections to the paper. My decision is major revision.
Examples:
Author Response
Dear reviewer, thank you very much for your valuable comments and suggestions. According to your suggestions, we have made a substantial revision of the paper from several aspects, such as results and discussion, conclusions and suggestions, so as to get your approval. And we marked the modified content in yellow. If there is any content that needs to be improved, please contact us. Thank you very much.
Point 1: Most importantly, out of all comments, the similarity is too high. It is 41%. In an SCI indexed journal and paper, this is not acceptable. When checked, the first 2 references that are shown similar total to 10% and belong to authors' other papers. It could be thought that it is possible to have similarity to own papers, however, I should repeat my first concern. For and SCI indexed prestigious journal, this makes the paper unacceptable. Similarity should be reduced below 20%, or preferably, below 15%.
Response 1: Thank you very much for your suggestion. According to your suggestion, we have dealt with the problem of high similarity in the paper. We have reduced the similarity of the paper to less than 15%.
Point 2: Descriptive statistics table is missing. Min, max, std dev., skew., kurtosis, JB test should be reported. Otherwise, the rest of the analysis become absolete.
Response 2: Thank you very much for your suggestion. According to your suggestion, we added a descriptive statistical table and JB test. Please see Lines 228-234.
Point 3: The threshold estimates are too high. What is the percentage of observations above threshold and below threshold? The value for threshold is too high for a series after logarithms. Since descriptives for the variables USED in the models are not present, we can just guess but anyways, threshold is high. It should be checked.
Response 3: Thank you very much for your suggestion. According to your suggestion, we added descriptive statistics. Please see Lines 228-229. We revised the bs value to 10000 and recalculated the threshold regression model. See Lines 293-294 for the calculation results. The final threshold estimate and 95% confidence interval are shown in Table 10.
Point 4: Bootstrap replications are tested with "0 300 300." 300 is the default in the code in STATA for examples in the help file. Try higher replication values such as 10 000 for reliability.
Response 4: Thank you very much for your suggestion. According to your suggestion, we revised the bs value to 10000 and recalculated the threshold regression model. Please see Lines 293-294.
Point 5: Another concern: Bootstrap replications are tested with "0 300 300." This means, for first threshold no replications. The code is not used correctly. This makes Table 7 questionable. Change replications. First should also be high, not zero.
Response 5: Thank you very much for your suggestion. It is indeed our negligence in our work to mistakenly write "300 300 300" as "0 300 300". Based on your suggestion, we set the bs value to "10000 10000 10000". Please see Lines 293-294.
Point 6: Last paragraph of the conclusion sound as if the paper is not finished. Why did the authors not estimate models for which they developed research concerns? Further, regarding inverted U shape which was analyzed, no conclusion can be made.
Response 6: Thank you very much for your suggestion. The last paragraph in Part 6 (Conclusions and Suggestions) is the research content of the next stage of the team, added according to the suggestions of the previous review experts. We’re sorry because we made a mistake, which made you misunderstand. In the first paragraph of Part 6 (Conclusions and Suggestions), this paper summarizes the research conclusions of this paper. Please see Lines 313-320. In addition, please see lines 250-284 for relevant conclusions of the nonlinear panel regression model. See lines 293-310 for relevant conclusions of the threshold regression model.
Point 7: Inverted U is not achieved. This also led to comment 6 above. The authors chose not to report a direct result with this respect. They wrote: "This paper focuses on whether there is an inverted U-shaped relationship between the sustainable development of urbanization and the urban-rural income gap. However, the selection of control variables in this empirical analysis model is based on the existing research results. " The sentence starting with "However" should discuss why inverted U is not achieved or is it achieved in this study. What is the relation of sentence 2 to sentence 1? The use of control variables caused the empirical findings below? Not very convincing since such models are not estimated by including and excluding control variables one by one. The parameter estimates are4.351**-3.182***. This is clearly not an inverted U since the peak is reached immediately. Again, since we don't know the range of data, we cannot further comment on this if this turning point is plausable or not. Ok, the panel regression did not work for inverted U, and isn't it the reason for the authors to shift to threshold panel regression in the first place? Why is this model not emphasized and conclusion is left with uncertain, uncompleted way?
Response 7: Thank you very much for your suggestion and advice. The last paragraph in Part 6 (Conclusions and Suggestions) is the research content of the next stage of the team, added according to the suggestions of the previous review experts. We’re sorry because we made a mistake, which made you misunderstand. In the first paragraph of Part 6 (Conclusions and Suggestions), this paper summarizes the research conclusions of this paper. Please see Lines 313-320. In addition, please see lines 250-284 for relevant conclusions of the nonlinear panel regression model. See lines 293-310 for relevant conclusions of the threshold regression model.
The main research ideas of this paper are as follows: firstly, through the analysis of the impact path, it is proposed that there may be an inverted U-curve relationship between the sustainable development of urbanization in China and the income gap between urban and rural residents. Secondly, through the nonlinear panel regression model, it is verified that there is an inverted U-curve relationship between the sustainable development of urbanization and the income gap between urban and rural residents in China. The results show that the quadratic coefficient is negative, and there is an inverted U-curve relationship through the significance test. Please see Lines 250-284. However, the results of the nonlinear panel regression model alone are not very convincing. Therefore, this paper constructs a threshold regression model. The results show that the sustainable development of urbanization passes the single threshold test. When SDLU is less than 0.209, the impact coefficient is 6.685, and when SDLU is greater than 0.209, the impact coefficient is -2.577. All passed the significance level test of 1%. Please see Lines 293-310.
Point 8: In the conclusion, the last sentence is: This paper does not analyze the influencing factors of the urban-rural 407 income gap. In the next stage, we will continue to study the relationship between the sus- 408 tainable development of urbanization and the income gap between urban and rural areas 409 in combination with the research on the factors affecting the income gap between urban 410 and rural areas. Authors wrote that they did not investigate determinants of, or in another words, the effects of factors on urban - rural income gap. However, by looking at the title of the paper, we as readers think that we will read a paper on investigation of determinants of rural urban income gap.
Response 8: Thank you very much for your suggestion. We are very sorry, but the focus of this paper is the impact of sustainable urbanization on the urban-rural income gap. The last paragraph in Part 6 (Conclusions and Suggestions) is the research content of the next stage of the team, added according to the suggestions of the previous review experts. We’re sorry because we made a mistake, which made you misunderstand. What we want to express is that when this paper studies the relationship between the sustainable development of urbanization and the urban-rural income gap, the choice of control variables is determined based on the existing research results without empirical analysis. In the next stage, we will further study other variables that affect the urban-rural income gap.
Point 9: I am sorry to say this but the wording / sentencing should be improved in the overall paper especially avoid direct statements such as X causes Y negatively. These should be tested empirically. Most and foremost, the 41% is a very high similarity. The authors should start by correcting this and also make corrections to the paper. My decision is major revision.
Response 9: Thank you very much for your suggestion. According to your suggestion, we have optimized the words/sentences in the paper and reduced the similarity of the paper to less than 15%.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The article presents a very interesting research, with a very current topic and of great projection, both nationally -in China- and international
Therefore, in order to it would be important to explain or analyze the following issues;
In the research, certain affirmations are made, at the beginning of the article that it would be important to specify, since it is affirmed that “China is one of the fastest developing countries in the world today. It has made remarkable achievements in social stability, economic growth, population development, and ecological protection”, although it is stated that there is a “polarization between the rich and the poor, a widening regional development gap, and unbalanced urban-rural development”, from this perspective, what do they refer to then with development (and, more especially, when mentioning social stability). With the above, rather than development, we would be talking about China being one of the countries with the highest growth, but not development.
In the same way, in the research mention is made, in a continuous way, of the existing relationship between the labor force, and, salary increase, and, in a less profuse way, training is indicated; for this reason, it would be interesting to include in the article some reference to the inclusion of whether there are -and, if this is not the case, whether they should be specified, with a view to the future-, educational or training policies, as a primary element of that social stability indicated in the article.
On the other hand, it would be important to dedicate a paragraph to specify what is meant in your research with “sustainable development level of urbanization” (what are the factors that determine it).
Likewise, although it is intuited that the figures are self-made, it would be important to point this out in the document. And, it would be positive to include the meaning of the variables of the mathematical formulas used in the study.
As a complement to the aforementioned studies of the Nobel Prize in Economics Simon Kuznets, it is recommended to bear in mind the works of Warren Thompson, Kingsley Davis or Frank Wallace Notestein, in relation to the Demographic Transition Model, or, Wilbur Zelinsky and his Model of Transition of Mobility.
Finally, to improve the understanding of the study, it would be recommendable to expand the analysis or comment on the tables; for example, in figure 9. It is stated that “This result is consistent with the fixed effect regression model. It further verifies that the relationship between the sustainable development of urbanization and the urban-rural income gap is an inverted Ushape”, therefore, it would be convenient to specify what this conclusion implies, what implications it has in terms of development (economic, social and environmental).
Author Response
Dear reviewer, thank you very much for your valuable comments and suggestions. According to your suggestions, we have made a substantial revision of the paper from several aspects, such as impact mechanism analysis, results and discussion, conclusions and suggestions, so as to get your approval. And we marked the modified content in yellow. If there is any content that needs to be improved, please contact us. Thank you very much.
Point 1: In the same way, in the research mention is made, in a continuous way, of the existing relationship between the labor force, and, salary increase, and, in a less profuse way, training is indicated; for this reason, it would be interesting to include in the article some reference to the inclusion of whether there are -and, if this is not the case, whether they should be specified, with a view to the future-, educational or training policies, as a primary element of that social stability indicated in the article.
Response 1: Thank you for your suggestion. According to your suggestions, we have added suggestions on reducing the urban-rural income gap through strengthening vocational training in the sixth part of the article (conclusions and suggestions). Please see Lines 350-359.
Point 2: On the other hand, it would be important to dedicate a paragraph to specify what is meant in your research with “sustainable development level of urbanization” (what are the factors that determine it).
Response 2: Thank you for your suggestion. According to your suggestion, we have made a specific explanation of the significance of the research on the "sustainable development level of urbanization". Please see Lines 111-115. We also have explained the factors that affect the sustainable development level of urbanization. Please see Lines 117-120.
Point 3: As a complement to the aforementioned studies of the Nobel Prize in Economics Simon Kuznets, it is recommended to bear in mind the works of Warren Thompson, Kingsley Davis or Frank Wallace Notestein, in relation to the Demographic Transition Model, or, Wilbur Zelinsky and his Model of Transition of Mobility.
Response 3: Thank you for your suggestion. According to your suggestion, we have studied the demographic transition model of Warren Thompson, Kingsley Davis, Frank Wallace Nordstein, and the mobility transition model of Wilbur Zerinsky. Relevant research results are cited. Please see Lines 109-120, and 420-421.
Point 4: Finally, to improve the understanding of the study, it would be recommendable to expand the analysis or comment on the tables; for example, in figure 9. It is stated that “This result is consistent with the fixed effect regression model. It further verifies that the relationship between the sustainable development of urbanization and the urban-rural income gap is an inverted Ushape”, therefore, it would be convenient to specify what this conclusion implies, what implications it has in terms of development (economic, social and environmental).
Response 4: Thank you for your suggestion. According to your suggestion, we have expanded the analysis or comments on the form. Please see Lines 301-309.
Round 2
Reviewer 1 Report
In the revised version, I noted that the paper is advanced and authors integrated the critiques effectively which were directed to the paper in the last round.
I also thank the authors for their detailed response file and integrating the comments effectively to their paper.
Regarding these critiques, I see that the majority of corrections are made well with one exception. This is noted in the first comment below. Also, I added my concerns in the following comments. These are minor issues and can be corrected in very short time. My decision is positive for this paper, however please attend these notes below.
1) Turning point issue: In the previous round, I directed my concerns for the turning point estimation in the paper. Regarding the parameter estimates that the authors obtained for SUL and SUL square, I noted that they were close, leading to suspicion of the turning point especially due to not reporting the descriptive statistics table in the paper. I now see that authors reported descriptive statistics table. I might have not fully explain what I meant in the last round regarding this critique. It still needs attention for correction. I also see that authors pointed at lines of paper for the corrections and I read them. However, the concern is not integrated to those lines. This must be the reason why the discussion of the turning point is not included.
The parameter estimates for SUL and SUL square are 4.351 and -3.182. They are significant. Signs are in the expected direction for existence of an inverted U shape. However they are too close. If the relation is written as,
Y = 4.351 X - 3.182 X^2
by taking partial derivative of Y with respect to X,
dY/dX = 4.351 - 2 * 3.182 X
by equating to zero,
4.351 - 2 * 3.182 X = 0
by leaving X on the left hand side, the turning point is (maximum point of inverted U shape):
X = 0.6836895.
For given min. and max. values of SUL, it is visible that the turning point is in the range of min and max values, which is very good. I want the authors to comment on the turning point, also by checking the data, at which regions or dates this value of SUL is passed and relate it with discussion to China's urban rural disparity in the context of sustainability.
2) Added tables: Due to my request in the previous round, authors included descriptive statistics but in two tables. No need for the second table since JB test and its probability could be written as a single column in the first table to save space.
Other than this, there are two typos in the JB table. First header should be JB, the second is the probability of JB. For the second, authors wrote Chi-square. But it is not. It is a p value of the JB test statistic which follows Chi-square distribution with degrees of freedom of 2 for large samples. Revise these headers. I suggest adding a column to the previous table, write JB [p.] as heading and report accordingly. The descriptives table will look much better.
3) Annotations issues: Check annotations. Throughout the paper, sustainable urban development level variable is written with two different annotations for instance. Starting from the abstract, it was written as "SU", in line 212, 183, 15, 25, 101, 102, 103, 104... in models, it is SUL. So, it is sometimes SUL sometimes SU. Start is made with SU and maintain it in the rest of the paper. Please revise and use one notation everywhere.
4) Recent directions in the literature issue.
Recent contributions with other nonlinear approaches such as markov switching, smooth transition regressions, neural networks are not added in the papers literature section. The following 4 papers I believe will improve the papers literature section with various respects that I will note on top of each:
a. The nonlinear relationship of environmental degradation and income with dynamic panel-STAR approach, that includes smooth versions of transitions instead of sudden changes such as the threshold approach which was also used in the reviewed paper:
https://www.sciencedirect.com/science/article/pii/S2212567116302052
b. the paper below, focuses on Markov-switching and neural networks approaches to environment - economic development relation and implications of petrol as a environmentally hazardous fossil fuel in terms of CO2 releases. The paper critisizes that if nonlinear methods are used to calculate turning points such as the kuznets curve, results would be biased results and this has important effects on sustainability:
https://link.springer.com/article/10.1007/s11356-018-3062-3
c. A recent research that showed the moderating role of urbanization which is in line with the reviewed paper:
https://link.springer.com/article/10.1007/s11356-020-09870-2
d. For turning point and inequality trends with a similar aspect to revived paper, this paper also focuses on China
https://www.sciencedirect.com/science/article/pii/S0016718516302445?casa_token=hPfzwY4zXkwAAAAA:C4MINmc2t-Sj5ss_gzP8HgJiE3oARC2Vx-bDA_0A32gd_zIBDFEkVQMCnfiekG3S5s9kZmhPojo
5) I thank for revising the paper for similarity reduction in the last revision after my critique which was to reduce it below 15%. They noted that it is reduced to 15%. However, I see that it is 20% in this version. I believe 20% is ok however I suggest a final revision with this respect as I asked before to take it to and if possible below 15% in the previous round. However, I did not check sources, if majority is from certain sources or not, I did not check it. I leave this to the authors and make revisions and adding references if it is necessary. I believe that no single source with 4% or more is a policy.
Overall result:
I believe that the paper has improved greatly after revisions and I congrad the authors for careful corrections they made. However, issues exist which could be easily corrected in very short time. I believe that authors could do these corrections promptly and the paper will be ready.
My decision is positive for this paper after these corrections.
Author Response
Dear reviewer, thank you very much for your valuable comments and suggestions. According to your suggestions, we have made a substantial revision of the paper from several aspects, such as Literature review, results and discussion, similarity, so as to get your approval. And we marked the modified content in yellow. If there is any content that needs to be improved, please contact us. Thank you very much.
Point 1: For given min. and max. values of SU, it is visible that the turning point is in the range of min and max values, which is very good. I want the authors to comment on the turning point, also by checking the data, at which regions or dates this value of SUL is passed and relate it with discussion to China's urban rural disparity in the context of sustainability.
Response 1: Thank you very much for your suggestion. According to your suggestion, we have analyzed the temporal and spatial differences of the SU values of 31 provinces in China based on the turning point. We have specifically analyzed which regions or times passed the SU value. Please see Lines 259-276.
Point 2: Added tables: Due to my request in the previous round, authors included descriptive statistics but in two tables. No need for the second table since JB test and its probability could be written as a single column in the first table to save space.
Other than this, there are two typos in the JB table. First header should be JB, the second is the probability of JB. For the second, authors wrote Chi-square. But it is not. It is a p value of the JB test statistic which follows Chi-square distribution with degrees of freedom of 2 for large samples. Revise these headers. I suggest adding a column to the previous table, write JB [p.] as heading and report accordingly. The descriptives table will look much better.
Response 2: Thank you very much for your suggestion. According to your suggestion, we have combined the two tables. On the basis of the original descriptive table, JB [p.] is added as a new column. Please see Lines 224-228.
Point 3: Annotations issues: Check annotations. Throughout the paper, sustainable urban development level variable is written with two different annotations for instance. Starting from the abstract, it was written as "SU", in line 212, 183, 15, 25, 101, 102, 103, 104... in models, it is SUL. So, it is sometimes SUL sometimes SU. Start is made with SU and maintain it in the rest of the paper. Please revise and use one notation everywhere.
Response 3: Thank you very much for your suggestion. According to your suggestion, we have unified the two different annotations in the text. We have checked the full text and adopted SU to represent sustainable urbanization.
Point 4: Recent directions in the literature issue. Recent contributions with other nonlinear approaches such as markov switching, smooth transition regressions, neural networks are not added in the papers literature section. The following 4 papers I believe will improve the papers literature section with various respects that I will note on top of each:
Response 4: Thank you very much for your suggestion. According to your suggestion, we have studied the four papers you recommended. These papers are very valuable. These papers have made valuable achievements in studying the nonlinear relationship between variables. However, the theme of these papers is not the urban-rural income gap, so we did not insert these articles into the literature review part of the paper. According to the research methods in these papers, we have supplemented our research directions in the next stage and added relevant references. Please see Lines 369-371, and 461-471.
Point 5: I thank for revising the paper for similarity reduction in the last revision after my critique which was to reduce it below 15%. They noted that it is reduced to 15%. However, I see that it is 20% in this version. I believe 20% is ok however I suggest a final revision with this respect as I asked before to take it to and if possible below 15% in the previous round. However, I did not check sources, if majority is from certain sources or not, I did not check it. I leave this to the authors and make revisions and adding references if it is necessary. I believe that no single source with 4% or more is a policy.
Response 5: We very much agree with your proposal. According to your suggestion, we further reduced the similarity of the paper. The similarity of the paper is monitored by using the Turnitin tool, and the results are shown in the figure below (in the attachment).
Author Response File: Author Response.pdf
