The Impact of Air Pollution Risk on the Sustainability of Crop Insurance Losses

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Suggestions for Improvement

Thank you for reviewing this paper. The manuscript is well-constructed.

The title accurately reflects the study’s focus but would be clearer and more precise if it mentioned "air pollution" explicitly.

The abstract is generally informative and well-structured . Briefly mentioning why the study focuses only on PM2.5 may improve the abstract.

The introduction is good. It provides a clear and relatively strong context for the study. Adding references of similar studies that investigate the effects of other pollutants (e.g., water pollutans, ozone or nitrogen oxides) on agricultural risks would improve the comprehensiveness of this section.

The methodology is fine, but the data need to be improved. The use of GAM and XGB has been applied appropriately, and the inclusion of Bai and Perron breakpoint analysis adds valuable insight into temporal shifts in these dynamics. Nonetheless, data up to 2018 (now is September 2024) limits the real-time applicability of the findings. Incorporating more recent data, if available, would certainly enhance the study's rigor. Alternatively, the authors may consider integrating qualitative methods, such as Focus Group Discussions (FGDs). This approach is important to support the results by gathering real-time insights from stakeholders. This empirical data could make the study more rigorous. This would help contextualize the findings and offer more reliable perspectives on recent trends.

The results and intepretation presented in the manuscript are fine. The two cases have been utilised properly, thus make this study stronger. The discussion, however, can be improved by linking the findings to practical implications, particularly around the identified breakpoints. The explanation of the identified breakpoints in 2011 and 2015 could be enhanced. Providing the context for why these years stood out—whether due to environmental shifts, regulatory changes, or significant climatic events—would strengthen the link between the data, the results and their practical impacts.

The suggestion provided, e.g., to integrate air pollution metrics into weather-indexed crop insurance is forward-thinking, but it could be expanded upon. The authors can discuss more practical steps for how air quality data could be operationalized in insurance models. This could involve some collaborative initiatives or potential actions. Furthermore, future research could benefit from looking at how other pollutants affect agricultural risk. This would expand the knowledge and its implications especially to the insurance companies as the focal firms. Including qualitative approaches, such as FGDs, not only overcome the lack of data but could also provide a more grounded understanding of recent trends and developments. This can add the depth and real-world applicability to the recommendations.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

• Clarification of Key Concepts: While the paper provides a general overview of the relationship between PM2.5 and crop insurance losses, it lacks a detailed explanation of how different crops (e.g., corn, soybean) are specifically affected by air pollution. To strengthen the argument, I recommend expanding on the crop-specific impacts of PM2.5 and discussing how these differences influence insurance claims. This will enhance the paper’s depth and provide clearer insights for readers unfamiliar with the nuances of crop insurance.

• Interpretation of Results: The statistical results are well-presented, but their practical implications could be better articulated. For example, the structural breakpoints identified (e.g., in 2011 and 2015) suggest a temporal shift in the relationship between PM2.5 and insurance losses, but the paper does not provide sufficient interpretation of what these shifts signify. It would be beneficial to link these breakpoints to potential environmental, policy, or market changes that may have influenced this relationship. Clearer explanation here would help readers better understand the significance of these findings.

• Discussion on Model Limitations: Although the paper acknowledges that the R-squared values for the insurance loss ratio model are relatively low, there is insufficient discussion on the limitations of both the GAM and XGB models. For instance, relying solely on PM2.5 as a measure of air pollution may overlook other contributing factors (e.g., ozone levels, nitrogen dioxide), which could affect the robustness of the models. I recommend addressing these potential limitations and discussing how incorporating additional variables could improve model performance and generalizability.

• Policy and Practical Implications: The paper would benefit from a stronger emphasis on the broader implications of the findings, particularly in terms of policy and industry practice. Given that crop insurance is a critical tool for agricultural risk management, it is important to discuss how insurance companies can integrate air pollution metrics into their pricing models. Additionally, recommendations for policymakers on addressing the risks posed by air pollution to the agricultural sector would increase the relevance and impact of the paper.

• Literature Review Updates: The paper draws on a solid body of literature, but it would be enhanced by incorporating more recent studies from the last three years, particularly those that focus on the intersection of air pollution, agricultural productivity, and insurance risk. This would not only make the literature review more current but also provide a broader context for the paper’s findings.

Comments on the Quality of English Language

Extensive editing of English language required.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This study quantifies the impact of air quality on crop insurance claims from an actuarial perspec- 13 tive and evaluates the implications for the industry. This topic is interesting but not well organized. 

1. The introduction section does not well writen. I suggest authors should reorganized this section, especially giving more contributions in this section. Also, please state the structure of this paper in the last of this section.

2. The theoretical analysis is missing. A qualified paper at least needs theoretical analysis or literature review, but this article is missing in this aspect. 

3. Many of the figures are not clear, it is suggested that the author modify them to make them clearer.

4. This article uses panel data, and I recommend that the authors use a dual fixed-effect model, including empirical analysis and model construction.

5. The empirical analysis of this paper is relatively simple, suggesting that the author add economic explanations of key results.

6. Some tables are three-wire tables, some are not. I suggest that the authors modify the format of the full text.

Comments on the Quality of English Language

Moderate editing of English language required.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for revising the manuscript and incorporating my recommendations. The manuscript revision is very satisfactory. The additional data have considerably improved the quality of the manuscript. I fully support this article being accepted for publication in MDPI Sustainability. Congratulations.

Reviewer 2 Report

Comments and Suggestions for Authors

The author has addressed my review comments very well.

Reviewer 3 Report

Comments and Suggestions for Authors

I agree with this revsion and it now can be accetped for publication.

Comments on the Quality of English Language

Minor editing of English language required.