Do Urban Innovation Policies Reduce Carbon Emission? Empirical Evidence from Chinese Cities with DID
Round 1
Reviewer 1 Report
Dear editor,
Thank you for inviting me to evaluate the article titled “Do urban innovation policies reduce carbon emission? Empirical evidence from Chinese cities with DID”. The paper provides a very comprehensive description of the pilot policy impact on carbon emissions and the specific policy influencing factors. Recommendations for promoting urban green and low-carbon transformation in China can help us to better promote sustainable social development. It is worth publishing and I believe that readers could draw meaningful insight from this paper.
There are a few places that the authors could improve upon.
1. In section 3.2, the four hypotheses are not perfectly juxtaposed. H2, H3 and H4 are more likely the details of H1, it would be better to reformat the structure.
2. In Figure 1, the comparison of the change trends of carbon emissions before and after 2010 is explained. I also notice that the trend curve of non-pilot cities has a clear decline. I would like to see some explanation for this change.
3. In line 517-527, the main findings show that the pilot policy can help reduce the carbon emission. However, I would like to see some specific data comparisons, to make the findings more obvious.
At the same time, some minor points in tables/figures should be noticed.
1. In table 1, the national innovation city in China 2022 is incorrect that there are two “Dezhou”, please check the table again to make sure it has no errors.
2. In Figure 1 legend, “Non-pilot cities” and “Pilot Cities”, the format needs to be uniform (The first letter needs to be the same case).
Author Response
Comment 1. In section 3.2, the four hypotheses are not perfectly juxtaposed. H2, H3 and H4 are more likely the details of H1, it would be better to reformat the structure.
Response: We are grateful for the comment from the reviewer. We have refined H2, H3 and H4 condensation into a new H2, as a detailed addition to further exploration of the H1 influence pathway.
“H2：The pilot policy of national innovation city can reduce CO2 emissions by enhancing green technology and innovation，promoting the transformation and upgrading of industrial structures and increasing government spending on technological innovation.”
Comment 2. In Figure 1, the comparison of the change trends of carbon emissions before and after 2010 is explained. I also notice that the trend curve of non-pilot cities has a clear decline. I would like to see some explanation for this change.
Response: We are grateful for the comment from the reviewer. In Section 5.1, we supplemented the relevant explanations for the obvious decline in per capita carbon emissions of non-pilot cities in Figure 2 in the later period.
“However, it is worth noting that since 2016, the per capita carbon emissions of non-pilot cities have shown a clear downward trend, while the growth rate of pilot cities has also shown a more obvious slowdown. This paper speculates that the main reason for this phenomenon may be that China has made important arrangements for low-carbon emission reduction at this time point. In fact, 2016 is of great significance for China's low-carbon development. In this year, China took the lead in signing the "Paris Agreement" and actively promoted its implementation, demonstrating to the world its determination and will to be more proactive in addressing climate change. This iconic move also means that China will take more active and bold actions to achieve the "carbon peaking and carbon neutrality" goals.”
Comment 3. In line 517-527, the main findings show that the pilot policy can help reduce the carbon emission. However, I would like to see some specific data comparisons, to make the findings more obvious.
Response: We are grateful for the comment from the reviewer, and we have added specific data comparisons in Section 5.2.
“Regression (2) shows that under the influence of the national innovative city pilot policy, the average per capita carbon emissions of the pilot cities are reduced by about 8.38%.”
Comment 4. In table 1, the national innovation city in China 2022 is incorrect that there are two “Dezhou”, please check the table again to make sure it has no errors.
Response: We are grateful for the comment from the reviewer. We have revised and checked the content of Table 1 to ensure that it is correct.
Comment 5. In Figure 1 legend, “Non-pilot cities” and “Pilot Cities”, the format needs to be uniform (The first letter needs to be the same case).
Response: We are grateful for the comment from the reviewer. We have unified the capitalization of the first letter in the legend of Figure 2.
Reviewer 2 Report
I read the whole manuscript, it is interesting but before acceptance the manuscript revision is needed.
1. Please improved the English language of the manuscript.
2. Please improve the figure quality from Fig. 2-Fig. 5. They are not understanding.
3. In the conclusion section please write the main object of the research work.
4. Please cited some more papers regarding the CO detection in the internal areas.
Author Response
Comment 1. Please improved the English language of the manuscript.
Response: Thank you so much for your kind advice, we have conducted a careful proofreading of the manuscript and the language has been improved.
Comment 2. Please improve the figure quality from Fig. 2-Fig. 5. They are not understanding.
Response: We are grateful for the comment from the reviewer. We have made appropriate adjustments to Figure 2–Figure 5. Please see the revised figures in the manuscript with track-changes.
Comment 3. In the conclusion section please write the main object of the research work.
Response: We are grateful for the comment from the reviewer. We have appropriately supplemented the research objectives of this paper in the conclusion section.
“In recent years, innovations in science and technologies have been increasingly important to achieve the goals of "carbon peaking and carbon neutrality". This study takes China's national strategic program in the field of scientific and technological innovation—the national innovation city pilot scheme as an example, to explore and verify the effect of this pilot policy on reducing carbon emissions in pilot cities and its underpinning mechanisms . This paper provides solid proof that scientific and technological innovation empowers the process of carbon peaking carbon neutrality, which is of great significance for the transformation of Chinese cities to a greener and more sustainable future.”
Comment 4. Please cited some more papers regarding the CO detection in the internal areas.
Response: We are grateful for the comment from the reviewer. We have added the relevant references in section 2.2 of the article.
Reviewer 3 Report
This article attempts to examine whether innovative city pilot help to improve the city’s carbon emissions and through which mechanisms does the pilot affect the carbon emissions of Chinese cities. The authors use the Difference in Difference (DID) method to investigate the impact of innovative city pilot policy on the carbon emissions of pilot cities and the underlying mechanism. They find that this pilot policy of national innovation cities, with innovation-driven development as the core strategy for urban economic and social development, plays an important supporting role in promoting the green and low-carbon transformation of the cities in China.
The paper is interesting and quite well-written. However, I have several concerns as follows.
1. The sample included only cities and it is hard to say that the policy is reduced the CO2 emission nationally. This is because dirty industries/factories may be forced to reallocate to rural areas and even other cities that are not in the pilot. In other words, the policy makes some cities cleaner and some other cities/areas dirtier.
2. Point (1) can be easily seen in Figure 1. Before the pilot, in general, the pilot cities emitted more CO2 than non-pilot cities, but after the pilot, there is a sharp increase in CO2 emission in the non-pilot cities. And this significantly weakened the conclusions, which derive from the DiD regression. Even the PSM approach could not deal with this issue since the PSM approach takes into account only observed characteristics (quite a limited number of variables in this paper). I encourage the authors to use external instrument variables rather than PSM.
3. Figure 1 also shows that before the pilot, on average, pilot cities emitted more CO2 than non-pilot cities. This situation may be the reason the cities are selected to be a pilot city. Therefore, it is not really a quasi-natural experiment. Which cities picked as pilot cities are endogenous and thus it may cause bias in the estimations. As point 2, I encourage the authors to use external instrument variables rather than PSM.
4. The authors want to examine the mechanism. It shows that pilot cities use more green technology, spending more on Science and Technology and ICT. However, as a reader, I am not sure whether these variables really work in the case of China or not. I wonder why the authors do not show that these variables really affect the CO2 emissions in Table 3.
Author Response
Comment 1 The sample included only cities and it is hard to say that the policy is reduced the CO2 emission nationally. This is because dirty industries/factories may be forced to reallocate to rural areas and even other cities that are not in the pilot. In other words, the policy makes some cities cleaner and some other cities/areas dirtier.
Response: We are grateful for the comment from the reviewer, and we have made the following explanations and responses to your questions.
We explain your comments as follows. Firstly, in China, cities are administrative divisions that include not only the urbanized areas but also its surrounding rural areas. Therefore, the cities’ statistics does not only include the emission data in the urban areas but also the rural areas within the cities’ administrative boundaries. And as a result, weather a factory moves from the urban region to the rural regions or not does not change the statistics within the cities’ administrative boundaries. Only if a factory moves across the borders of different cities, no matter to the rural or urban regions, the statistics of a city will be changed.
In addition, in the long run, we believe that the pilot policy is also beneficial for reducing the carbon emission level of non-pilot cities. On the one hand, it can be seen from Figure 2 that although the growth rate of per capita carbon emissions in non-pilot cities increased in the first few years after the implementation of the pilot policy, there has been a clear downward trend since 2015. On the other hand, the ultimate goal of the pilot policy is to build an innovative country, not just limited to the pilot cities themselves. Therefore, with the continuous expansion of the pilot list of the national innovation city pilot, it is believed that the country's overall carbon emission level will also be greatly improved.
Comment 2 Point (1) can be easily seen in Figure 1. Before the pilot, in general, the pilot cities emitted more CO2 than non-pilot cities, but after the pilot, there is a sharp increase in CO2 emission in the non-pilot cities. And this significantly weakened the conclusions, which derive from the DiD regression. Even the PSM approach could not deal with this issue since the PSM approach takes into account only observed characteristics (quite a limited number of variables in this paper). I encourage the authors to use external instrument variables rather than PSM.
Response: We are grateful for the comment from the reviewer. However, we need to keep the comparison between the control group and treatment group reliable and solid, and therefore PSM is necessary before conducting the analysis of DID. On top of that, we chose the variables in PSM following the most relevant theories and frequently citied literature, this method has been examined and reviewed in researches on urban carbon emissions many times and we believe it should be feasible in this research. In addition, we have added several sentences to explain the change in carbon emission reduction of non-pilot cities.
Comment 3 Figure 1 also shows that before the pilot, on average, pilot cities emitted more CO2 than non-pilot cities. This situation may be the reason the cities are selected to be a pilot city. Therefore, it is not really a quasi-natural experiment. Which cities picked as pilot cities are endogenous and thus it may cause bias in the estimations. As point 2, I encourage the authors to use external instrument variables rather than PSM.
Response: We are grateful for the comment from the reviewer. In fact, the quasi-natural experiment does not strictly require the complete exclusion of the influence of human factors. For example, some scholars put forward, "More recently, the use of the term ‘natural’ has been understood more broadly as an event which did not involve the deliberate manipulation of exposure for research purposes (for example a policy change), even if human agency was involved." (Frank de Vocht et al., 2021) In addition, the national innovative city pilot policy has also been defined in many other papers as a quasi-natural experiment and has been extensively studied on this basis. Referring to the existing viewpoints and practices, we believe that the pilot policy meets the standard of "quasi-natural experiment". And we have used multiple test mechanisms to make sure there is no significant differences between our treatment and control group, therefore the comparison is solid. We have also conducted multiple rounds of robustness test in our section 5 to ensure the results from our research are solid. Please see section 5.3 and 5.4.
Comment 4 The authors want to examine the mechanism. It shows that pilot cities use more green technology, spending more on Science and Technology and ICT. However, as a reader, I am not sure whether these variables really work in the case of China or not. I wonder why the authors do not show that these variables really affect the CO2 emissions in Table 3.
Response: We are grateful for the comment from the reviewer. In 3.2.2 of the articles, we have preliminarily deduced how the selected three variables affect the carbon emissions of the pilot cities. Moreover, in the process of derivation, we cited the papers and opinions of many Chinese scholars, so we believe that the selection of these three variables is in line with China's national conditions.
Reviewer 4 Report
Dear Autros
The article's subject matter is not new, but it is quite an interesting presentation of the discussed topic.
The article needs improvement:
- - Figures 1, 3, 4, and 5 are minimal and show nothing - the size of the figures should be corrected
- - In my opinion, the article lacks a discussion of the results obtained. Presenting only charts and tables without discussing and summarizing them is insufficient. Therefore, I believe that the article should be supplemented with a discussion of the results of the analyzes carried out.
Author Response
Comment 1 Figures 1, 3, 4, and 5 are minimal and show nothing - the size of the figures should be corrected
Response: We are grateful for the comment from the reviewer, and we have corrected the size of Figures 1,3,4 and 5.
Comment 2 In my opinion, the article lacks a discussion of the results obtained. Presenting only charts and tables without discussing and summarizing them is insufficient. Therefore, I believe that the article should be supplemented with a discussion of the results of the analyzes carried out.
Response: Thank you so much for your comments. Now section 5 revised as “empirical results and discussions” and we have added discussions to each statistical result accordingly. We have also added more discussion in our conclusion and implication section. Please see our revised manuscript with track-changes.
Round 2
Reviewer 3 Report
Thank you for your comments. I am not really satisfied with your responses, especially regarding the mechanisms. I still wonder why you do not estimate ones since you have data on your hand but use the conclusions from other papers.
Regarding the cities boundary, I understand your points, but what I mean is not the rural areas around the cities, but townships and other rural areas.
Author Response
We really appreciate your prompt comments. We would like to make more clarifications on our revisions. Thank you.
- We kept the PSD-DID method because it has been frequently practiced in studies on pilot schemes. We have already published one ESI top 1% most cited paper with the method studying the low-carbon pilot scheme in China. We think it is more appropriate to use PSM-DID rather than instrumental variables to evaluate the effectiveness of the pilot policy. In addition, all the other three reviewers agreed with the method.
- As to the mechanisms, we are not using the conclusions from other papers. In effect, when we conduct our research, we cannot just presume mechanisms without any proof. The previous research serves as theoretical basis upon which most valuable and possible mechanisms are to be tested in our research.
- As to the concept of city/township/rural areas, I believe it is irrelevant to our research. Even if a factory moves from one city to another (no matter to the rural areas or townships), its carbon emission data will still be included in the city statistics where it locates.
