Has China’s Pilot Policy of Water Ecological Civilization City Construction Reduced Water Pollution Intensity?
Round 1
Reviewer 1 Report
The topic on policy evaluation of Water Ecological Civilization City Construction is intresting. I have some comments and suggestions below, which I hope are useful to the authors.
1.Reseatch Hypotheses.
The Primary Hypothesis part mainly describe the content of the policy document, while lacking basic theoretical elaboration. ALso, the mechanism Hypothesis part is the same problem, lacking theoretical elaboration rather than just interpretation of policy documents.
2. Results.
For the parallel trend hypothesis test, only the results of Eq. 2 are displayed, the results of Eq. 3 is missing. The figure of Eq.(3) needs to be supplemented. In addition，usually 95% confidence interval should be used to test whether the model satisfy the parallel trend hypothesis, while authors use 90% confidence interval instead. Authors need to elaborate it.
For the spatial relevance part, auther try to use SLM model to control the spatial spillover. The result of table 4 shows that the the coefficient of spatial lag item rho is 1.0126. It seems odd that the rho value is greater than 1. Generally, the absolute value of rho should be less than 1. Greater than 1 means that the spatial spillover effect of surrounding areas is greater than its own. Authors needs to elaborate it. In addition, the indirect effect is not explained correctly. The indirect effect is the impact of the policies in the surrounding areas on the own region, rather than the impact of the policies in the own region on the surrounding areas. (see, for instance, Sun et al. 2017)
For the mechansim test, the author uses DID method to test the impact of policies on the five channels. However, this cannot prove that the mechanism is effective. The author needs to further prove that the five channels can affect the water pollution density respectively. As emphasized above, there is a lack of theoretical analysis of mechanisms.
For the regional heterogeneity part, the authors try to use sample grouping to compare the coefficient differences of different groups to illustrate regional heterogeneity. However, the coefficient difference of the two groups of regressions cannot be directly compared. The authors should consider Chow test or Seemingly Unrelated Regression method to test the significance of coefficient difference.
3.Language.
The English expression of this paper is ok, but there are also some grammatical errors that make it difficult to read. For example, “and the impacts are smaller for cities with high economic development level are less than those cities with low economic development level” (page 15, line 565-567)
Reference:
Sun, X., Chen, F. and Geoffrey, H. (2017) Spatial Perspective
on Regional Growth in China: Evidence from an Extended Neoclassic Growth Model, Emerging Markets Finance and Trade, 53(9), 2063-2081.
Elhorst, P. (2012) Spatial econometrics: from cross-sectional data to spatial panels, Springer.
Chung, S and Hewings, G. (2015) Competitive and Complementary Relationship between Regional Economies: A Study of the Great Lake States, Spatial Economic Analysis, 10, 205-229
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The selected topic of this paper has certain novelty and practical significance. The language expression is generally fine，although there are some small mistakes.It can be published in this journal after minor revision. Some suggestions are as follows：
1、The introduction can be further deepened.
2、The figures in the paper are not standard enough, authors should modify them according to the requirements of the journal.
3、The conclusion is too complicated and needs further refinement.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The quality of the manuscript has been greatly improved. The last comment and suggestion is as follows.
From the figure (2) and (3), It looks that the 95% confidence interval does not contain zero value, which means it doesn't meet the requirements of parallel trend hyposesis unfortunately. The authors perhaps should use the control group and treatment group constructed by PSM method to conduct parallel trend test.
Author Response
We sincerely appreciate all valuable comments and suggestions, which helped us to improve the quality of the manuscript.
We are sorry that we made a mistake in Figure 2 and Figure 3 in the first revised version. Figure 2 and Figure 3 in the first revised version are made based on the fact that the units of some variables are RMB, but we actually converted the units of some variables into dollars. Figure 2 and Figure 3 in this revised version are 95% confidence interval figures.
Reviewer 2 Report
The Manuscript can be published in this journal after careful examination by all authors.
Author Response
We sincerely appreciate all valuable comments and suggestions, which helped us to improve the quality of the manuscript.
We have checked and revised the language in the paper.
