Does the Innovative City Pilot Policy Improve Urban Resilience? Evidence from China

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

"the innovative city pilot policy" is a very important factor to the urban resilience, and the manusciript did a good research on the relationship between them. There are some suggestions:

1. It is suggested that the authors can explain how the "Control Variables" were chosen and the more detailed impact pathways.

2. “Notes: ***, **, and * represent the significance levels at 1%, 5%, and 10%, respectively” in Table 3， there are no "**or *" . It is recommended that the authors check the data and the result. So as other tables.

3. How to explain the long trend of the effects of the policy，it is a research problem. And, it is suggested that the authors can supplement the theoretical explanation of the "innovative city pilot policy" .

4. "4.5.3 Heterogeneity of city size":  it is very confusing that the "large cities have already largely exhausted their potential to benefit from the ICPP". The city can be regarded as an organic network, and the large cities can be supported by better structural and technological effects. It is suggested that the authors supplement meaningful heterogeneity analysis.

5. The English writing in the manuscript can be revised.

 

Comments on the Quality of English Language

Moderate editing of English language required.

Author Response

Reply1

Comments 1: It is suggested that the authors explain how the "Control Variables" were chosen and the more detailed impact pathways.

Response 1: Thanks for adding something valuable to our manuscript. Controlling variables is an important component of the article's assessment of the influence of the ICPP on urban resilience. The theoretical justifications for the manner in which the article's control variables impact urban resilience need a more thorough approach. We provide the following proof for each of the article's four control variables: Environmental rules compel highly polluting businesses to change, boost the development of new technologies, thereby supporting industrial structure improvements and increasing industrial resilience. A city's IP protection level has a multiplicative effect on business resiliency, innovation conversion rate, and R&D excitement. Population density has a more immediate influence on urban resilience. While dense populations can help new ideas spread, they also increase the risk of urban overpopulation and make it harder for cities to react to and recover from emergencies. Government investment in research, technology, and education boosts human capital, fosters the generation and dissemination of knowledge and technology, and elevates the city's intellectual capital. This leads to more specialized solutions in the event of a catastrophe and more recovery industries after a disaster. The exact changes can be found in the second line of the story on page 6.

Comments 2: "Notes: ***, **, and * represent the significance levels at 1%, 5%, and 10%, respectively." In Table 3， there are no "**or *." It is recommended that the authors check the data and the results. So as other tables.

Response 2: We appreciate your thorough analysis of our outcomes. No private information is sourced for any of the article's facts. We created assessment metrics and ran regression tests after thoroughly cleaning and filtering the data. In light of this, we have made all process codes and data variables available for evaluation. The additional materials in this version now include screenshots of our Stata regression model procedures, which we used to back up our results.

Comments 3: How to explain the long-term trend of the effects of the policy is a research problem. And, it is suggested that the authors can supplement the theoretical explanation of the "innovative city pilot policy".

Response 3: We appreciate you bringing to our attention the fact that the analysis should have accounted for the policy's trend in long-term impacts. To further prove our study conclusion—that creative municipal policies may boost urban resilience—it is crucial to know whether the policy's benefits can have a long-term trend. The article's difference-in-differences model served as the basis for our parallel trend assumption. There was no discernible impact of extraneous variables on urban resilience before the policy was implemented, as there was no discernible trend in urban resilience between the control and experimental groups. Consistently significant regression coefficients for four years after policy implementation demonstrate the innovative city strategy's direct effect on urban resilience. This also suggests a pattern to the policy's long-term impacts. We have included further information on the long-term trend of the innovative city pilot policy's influence on urban resilience in the paper, particularly in the section on page 11 that discusses parallel trend assumptions based on the test findings. Additionally, on page 7, in the second paragraph, we clarified the data to include the period from 2004 to 2020, which is more advantageous for analyzing potential long-term patterns of the program.

Comments 4: "4.5.3 Heterogeneity of city size":  it is very confusing that the "large cities have already largely exhausted their potential to benefit from the ICPP". The city can be regarded as an organic network, and the large cities can be supported by better structural and technological effects. It is suggested that the authors supplement meaningful heterogeneity analysis. Thank you for pointing this out.

Response 4: We are willing to draw on your viewpoints and continue reflecting on the article's shortcomings. Large cities, due to industrial agglomeration and well-developed infrastructure, are better able to leverage scale and agglomeration effects, thus accelerating the enhancement of urban resilience. However, certain large cities, characterized by excessive scale and high density, face potential risks to their urban resilience, which may impact the effectiveness of the ICPP. Given the article's findings that the impact of ICPP on urban resilience in large cities is not significant, we provide a more reasonable theoretical explanation for the heterogeneity of urban resilience in these cities.

On one hand, urban ecosystems may be negatively impacted by over-expansion of city limits and over-extraction of urban material resources for urbanization projects. Concurrently, the ICPP's push for industrial parks and other forms of industrial agglomeration might reduce urban ecological resources' natural metabolism and circulation, making big cities less resilient to and less able to adapt to natural catastrophes. On the other hand, large cities, on the other hand, with their high population density and infrastructure, are more prone to foster urban concerns like traffic congestion, heat island effects, and air pollution. People moving into innovative cities at an unprecedented rate could make them less resilient to natural catastrophes and more vulnerable to exposure dangers. The third paragraph on page 19 of the paper and the relevant conclusion in the second paragraph on page 20 show the changes we made.

Comments 5: The English writing in the manuscript can be revised.

Response 5: The sections of the article that address all your suggested revisions have also been specifically modified, hoping that our enhancement could meet betteer academic standards.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Congratulations for your research, very interesting.

Only two questions, regarding regressions r-squared (r2): even knowing that for the methodology you employed r2 is not necessarily critical, it is, nevertheless, a relevant indicator concerning the goodness-of-fit of your models.

Why you just show R2 twice?

Why you never refer to R2 adjusted to the number of variables used?

Many thanks 

Comments on the Quality of English Language

English fair enough

Author Response

Reply2

Comments 1: Only two questions, regarding regressions r-squared (r2): even knowing that for the methodology you employed r2 is not necessarily critical, it is, nevertheless, a relevant indicator concerning the goodness-of-fit of your models.

Why you just show R2 twice?

Why you never refer to R2 adjusted to the number of variables used?

Response 1: We really appreciate your priceless recommendations for strengthening our work. Presenting R-squared will help to show the improvement in the academic rigor of the paper as it is a useful measure of the quality of fit of a linear model.

When we begin carrying out empirical experiments, we assessed the R-squared and modified R-squared for all models to make sure our model fits very well, and the results are all within a respectable range. Because our primary focus was on the article's difference-in-differences model's core explanatory variable coefficients, we consistently endeavored to make sure the natural experiment results on urban resilience from the innovative cities pilot policy were scientifically valid by running a battery of significance tests and robustness checks, therefore neglecting the accuracy of the model's goodness of fit.

But as you pointed out, it is impossible to overlook the part R-squared plays in evaluating model fit. Consequently, we have actively incorporated the display of modified R-squared in the main linear regression model tables in our study. The F-test, R-squared, and modified R-squared values for each linear regression model are shown in the following Table for the purpose of providing more transparent proof of model fit. All of the models' F-tests have appropriate significance levels, suggesting meaningful linear associations, aside from the test failure in Table 7 attributable to an inadequate number of city groups in the Northeast region. All corrected R-squared values are within a tolerable range, indicating no overfitting interference. The article's description of model fit is lacking, and we hope our enhancement may fix that.

Table 3							Table 4
	(1)		(2)		(3)			(1)	(2)	(3)
	RESL		RESL		RESL			Weighted	On Support	Weight_Reg
F	906.7***		935.8***		759.1***			293.3***	759.1***	216.7***
R^2	0.788		0.787		0.788			0.785	0.788	0.785
R-squared	0.787		0.786		0.787			0.783	0.787	0.784
Table7								Table 9
	(1)	(2)		(3)		(4)		(1)	(2)	(3)
	East	Midst		West		Northeast		Small	Large	Medium
F	257.7***	157.2***		482.5***		-		95.7***	316.0***	203.6***
R^2	0.766	0.817		0.798		0.806		0.805	0.787	0.795
R-squared	0.762	0.811		0.795		0.790		0.799	0.784	0.792
	Table 8
	(1)							(2)
	High level							Low level
F	193.9***							443.0***
R^2	0.810							0.783
R-squared	0.806							0.781

Notes: ***, **, and * represent the significance levels at 1%, 5%, and 10%, respectively.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

Please state what are small, medium and big cities based on the number of inhabitants or other characteristics

The indicators are clearly stated (which is important as those are a choice that somebody else might do differently). Please explain why those indicators are chosen.

Line 53 "ideological" - please explain or is it an error?

Author Response

Reply3

Comments 1: Please state what are small, medium and big cities based on the number of inhabitants or other characteristics

Response 1: Large cities with populations above 5 million, medium-sized cities with populations between 3 and 5 million, and small cities with populations below 3 million are the three types of cities based on their permanent resident population that the article refers to in its 2014 State Council notice on adjusting the standards for classifying urban sizes. Referring to the notice's five-category, seven-level system, this article combines small cities, medium-sized cities, and second-tier large cities into the category of small cities, classifies first-tier large cities as medium-sized cities, and merges super-large cities and megacities with populations exceeding 5 million into the category of large cities. The varied features of the impacts of creative urban policy pilots on urban resilience across various city sizes may be more succinctly and clearly shown using the three types of cities. Apart from that, there are 761, 1,393, and 1,769 city samples, respectively. The article's city sample data follows a more uniform distribution pattern, and the more similar distribution of sample sizes across groups helps to eliminate estimate errors caused by large disparities in sample sizes.

Comments 2: The indicators are clearly stated (which is important as those are a choice that somebody else might do differently). Please explain why those indicators are chosen.

Response 2: Your questions and ideas are much appreciated. We have included the following clarifications and adjustments in response to your feedback.

Initially, to thoroughly evaluate the influence of innovative municipal policies on urban resilience systems, the development of our urban resilience indicator system draws upon the methodologies outlined in the article, establishing a multidimensional resilience framework for urban systems within the context of urban resilience in China. Additionally, drawing on the existing literature, we identified scientifically valid variables from four domains, economy, society, ecology, and infrastructure, to develop a sub-dimensional indicator system. We gave more detail on the related results in the section "The Index Construction of Urban Resilience in China" on page 6 and page 7.

In addition, our urban resilience indicators for multidimensional sub-resilience show a decreasing trend from eastern to western areas of China, which aligns with previous research on urban resilience in China. In order to have more scientific evidence， we rated the four central regions of China based on their average five-year resilience ranks on four sub-dimensions of resilience and relevant analysis to tell in the last paragraph on page 7 and the first paragraph on page 8.

Third, to mitigate some of the subjectivity involved in variable selection, we used the entropy weight approach to weigh the indicators during indicator creation; this method determines the degree of redundancy of each element.

Comments 3: Line 53 "ideological" - please explain or is it an error?

Response 3: We really appreciate receiving your insightful revision ideas. The essay failed to adequately convey the creative aspects because of a phrasing mistake on our part. An innovative city achieves its urban growth model through innovation with the help of endogenous variables like people, resources, culture, technology, and systems. Such a city has high-end radiation and is a leader in its area. “Ideological innovation “comprises all inventive aspects other than the aforementioned systemic factors. The ICPP aims to improve enterprises' ability to innovate independently, reform innovation systems and mechanisms, create an innovation atmosphere in the region, foster innovation within cities, and establish a regional technology innovation system that integrates industry, academia, and research. Thank you very much for your suggestion. We have made a more reasonable revision in the second paragraph on page 2 of the original text.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Authors have responded to my comments and the revised manuscript can be accepted now.

 

Comments on the Quality of English Language

Moderate editing of English language required

Author Response

Dear reviewer,   We truly apologize for not contacting you immediately after receiving the revision comments. Yesterday morning, we provided the editor with some questions regarding the revisions. Can we confirm that this minor revision specifically means improving the English writing level of the article and further polishing it? We are concerned that there might be a misunderstanding regarding the modifications, so we would like to confirm with you further.

  We are very grateful for the your revision comments, and we will continue to improve our article to meet the submission standards.