Climate Change Exposure and Cash Holdings

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The topic of the study is relevant and falls into the mainstream of priority areas of research in this field of knowledge. The article corresponds to the scope of the journal and the issues. The research uses methods that allow authors to get answers to the questions posed. The presented article contains original scientific results linking the companies' cash holdings with climate change both from the point of view of leveling uncertainty and from the point of view of using opportunities. An interesting methodology has been built into the work. The text of the study is characterized by a fairly high level of originality (above 70 percent) based on the results of verification in the anti-plagiarism system. Despite the noted advantages and positive characteristics, the work can be improved by answering some controversial questions. It seems advisable to adjust the structure of the article. Currently, the section "introduction" provides both relevance and literature analysis, methodology, data, and results. It is recommended to move part of the text to the appropriate sections, removing redundant information from the introduction. This is especially true for those results that, according to the authors, provide an increase in scientific knowledge in theory and methodology. It is recommended to correct the title of Section 2.1. Climate Change on Cash Holdings. More attention needs to be paid not only to the data on the three constituent aspects of the study but also to how well they correspond to each other. For example, data on companies from 34 countries are used for Firm-Level Climate Change Exposure Measures. And financial data are obtained from Compustat North America. Similarly, the question is about weather data, since data from sensors in the USA (10,000 weather stations across the US.) is used. It is recommended to pay attention once again to the justification of the correctness of using all the research data (explanations are partially given, but they do not answer all the questions). In conclusion, it is recommended to strengthen the part on the practical application of the research results.

Author Response

Comment #1

The topic of the study is relevant and falls into the mainstream of priority areas of research in this field of knowledge. The article corresponds to the scope of the journal and the issues. The research uses methods that allow authors to get answers to the questions posed. The presented article contains original scientific results linking the companies' cash holdings with climate change both from the point of view of leveling uncertainty and from the point of view of using opportunities. An interesting methodology has been built into the work.

1.The text of the study is characterized by a fairly high level of originality based on the results of verification in the anti-plagiarism system.

Response: We have thoroughly revisited all sections of the content. In doing so, we have reviewed and refined the phrasing, structure, and citations where necessary to eliminate any potential issues.

2. Despite the noted advantages and positive characteristics, the work can be improved by answering some controversial questions. It seems advisable to adjust the structure of the article. Currently, the section "introduction" provides both relevance and literature analysis, methodology, data, and results. It is recommended to move part of the text to the appropriate sections, removing redundant information from the introduction. This is especially true for those results that, according to the authors, provide an increase in scientific knowledge in theory and methodology. It is recommended to correct the title of Section 2.1. Climate Change on Cash Holdings.

Response: We have revised the structure of the article, focusing particularly on the introduction and data sections. The introduction has been restructured to improve clarity and eliminate redundant information. Additionally, we have expanded the discussion in the data and sample construction section to better demonstrate how the datasets correspond to one another. Furthermore, the title of Section 2.1 has been updated to "The Effect of Climate Change on Cash Holdings" to more accurately reflect its content.

3.More attention needs to be paid not only to the data on the three constituent aspects of the study but also to how well they correspond to each other. For example, data on companies from 34 countries are used for Firm-Level Climate Change Exposure Measures. And financial data are obtained from Compustat North America. Similarly, the question is about weather data, since data from sensors in the USA (10,000 weather stations across the US.) is used. It is recommended to pay attention once again to the justification of the correctness of using all the research data (explanations are partially given, but they do not answer all the questions). In conclusion, it is recommended to strengthen the part on the practical application of the research results.

Response:

We have included detailed explanations in the revised manuscript to justify the integration of firm-level climate change exposure measures, financial data, and weather data. For the firm-level climate change exposure dataset, we now provide a thorough discussion on its coverage, and time-trend distribution, emphasizing the rationale behind our sample selection. Regarding the station-level weather data, we have added further details explaining the methodology used to convert this information and link it with firm-level quarterly data effectively.

To ensure greater clarity, we have added a new subsection in the data section that comprehensively addresses the construction of the sample and demonstrates how each dataset aligns with the study’s research objectives. Additionally, we have provided Table 1, which provides an overview of the sample distribution by industry and time period. These additions strengthen the justification of our data integration and offer a clear representation of the dataset used in the study.

Reviewer 2 Report

Comments and Suggestions for Authors

The work is very interesting and original. The effects of climate change on business management, particularly on cash holding, are very important to estimate. However, the paper has in my opinion two weaknesses that may lower the quality level of the paper. The econometric model is not explained in depth. Why was that type of regression used? How were multicollinearity and hetereskedasticity of the data taken into account? What is the level of robustness of the estimates? are questions that the paper does not give a clear and comprehensive answer to.
What is missing, in my opinion, is a section discussing the results and its implications. The transition from results to conclusions is too quick and reductive.
A further major improvement on the paper is therefore recommended.

Author Response

Comment #2

The work is very interesting and original. The effects of climate change on business management, particularly on cash holding, are very important to estimate. However, the paper has in my opinion two weaknesses that may lower the quality level of the paper.

1. The econometric model is not explained in depth. Why was that type of regression used? How were multicollinearity and hetereskedasticity of the data taken into account? What is the level of robustness of the estimates?

Response: We have revised the discussion to clarify the econometric model, as detailed in Section 4.1. The analysis employs a panel data regression to examine the relationship between climate change exposure and firms’ cash holdings, capturing both cross-sectional and temporal variations. Fixed effects at the two-digit SIC level account for unobservable industry-specific factors, while year-fixed effects address time-specific influences such as macroeconomic or regulatory changes, ensuring unbiased estimates.

The model evaluates both immediate (k=0) and delayed (k=1) effects by analyzing cash holdings in the current and subsequent quarters. To ensure robust hypothesis testing, standard errors are clustered at both firm and year levels, following Petersen (2009). This dual clustering accounts for cross-sectional and temporal dependencies, providing reliable estimates and a comprehensive view of the impact of climate change exposure on firms' liquidity management.

To ensure the robustness of our estimates, we performed several additional analyses. Specifically, we tested our findings using alternative measures of climate change exposure, applied weather-based instrumental variables (IV), and utilized the Paris Agreement as a validation framework. Additionally, we excluded observations from economic crisis periods to confirm that the results are not influenced by extreme events. These robustness checks demonstrate that our findings are consistent and reliable across various measures, instruments, and sample adjustments. For further details, please refer to Sections 4.3–4.7 of the manuscript.

2. Are questions that the paper does not give a clear and comprehensive answer to. What is missing, in my opinion, is a section discussing the results and its implications. The transition from results to conclusions is too quick and reductive. A further major improvement on the paper is therefore recommended.

Response: We have revised and expanded Sections 4.1 and 4.2 to provide a detailed discussion of our baseline findings and their implications. Specifically, we have included explanations for the observed results, such as why firms increase cash holdings in response to opportunities but do not do so for risks. This enhanced discussion bridges the gap between the results and conclusions, offering a more comprehensive interpretation of the findings. Additionally, we have elaborated on the practical significance and broader implications of our results, ensuring a clearer and more logical transition from the results to the conclusions.

Reviewer 3 Report

Comments and Suggestions for Authors

This paper addresses the important and timely topic of how climate change exposure, both as an opportunity and as a source of uncertainty, affects firms' cash holdings. The study employs a novel textual analysis to quantify climate change exposure using earnings conference call transcripts and applies rigorous econometric methods. While the paper has the potential to make a valuable contribution, it is currently limited by several critical issues that need to be addressed.

1. The contribution to the existing literature needs better articulation. For instance, how does this study go beyond prior research on corporate cash holdings or climate-related disclosures? There is insufficient engagement with the literature on precautionary motives for cash holdings, environmental performance, and climate change risks. Incorporating a more thorough comparison with existing studies would strengthen the paper’s positioning.

2. How are the terms used in the textual analysis (e.g., climate-related keywords) mapped to specific opportunities or uncertainties? Are there risks of misclassification or noise in the data? The paper does not address the robustness of the textual analysis. For example, could alternative methods (e.g., dictionary refinement or supervised machine learning) yield consistent results? The time coverage of the conference call dataset and its representativeness for different industries and firm sizes are not discussed.

3. While the paper uses an IV approach to address endogeneity, the choice and validity of instruments are problematic: Weather variables are highly correlated with regional climate conditions but may lack direct relevance to firm-level cash-holding decisions. This undermines their ability to satisfy the exclusion restriction. The paper does not adequately explore alternative instruments, such as policy shocks or industry-specific climate change initiatives, which could provide stronger exogeneity.

4. Importantly, the paper identifies a positive relationship between climate opportunities and cash holdings but does not convincingly explain why this occurs. Key questions remain unanswered: Are firms stockpiling cash to capitalize on future climate-related investments, such as renewable energy projects? Alternatively, does this reflect inefficiencies in cash allocation driven by managerial conservatism under uncertainty?

5. Regarding the robustness check. It would be useful to test the sensitivity of results to alternative climate change exposure measures (e.g., ESG scores, and third-party climate risk indices). Suggest conducting placebo tests, such as using unrelated keywords in the textual analysis, could help validate the exposure measures.

6. The results section often describes findings without critically interpreting them. For instance, what do the magnitudes of the coefficients imply for firms’ financial strategies or climate adaptation efforts? suggest to move the discussion later in another section.

7. The discussion of policy implications is underdeveloped. The paper should address: How policymakers can use these findings to incentivize firms to allocate cash holdings toward sustainability initiatives rather than hoarding. Similarly, the literature review is too descriptive, summarizing studies without critically engaging with them. For example, the discussion of precautionary cash holdings does not address how this concept interacts with climate risks or opportunities.

Comments on the Quality of English Language

The paper is verbose, especially in the methodology and robustness sections, leading to unnecessary repetition. Streamlining these sections would improve readability.

Author Response

Comment #3

This paper addresses the important and timely topic of how climate change exposure, both as an opportunity and as a source of uncertainty, affects firms' cash holdings. The study employs a novel textual analysis to quantify climate change exposure using earnings conference call transcripts and applies rigorous econometric methods. While the paper has the potential to make a valuable contribution, it is currently limited by several critical issues that need to be addressed.

1.The contribution to the existing literature needs better articulation. For instance, how does this study go beyond prior research on corporate cash holdings or climate-related disclosures? There is insufficient engagement with the literature on precautionary motives for cash holdings, environmental performance, and climate change risks. Incorporating a more thorough comparison with existing studies would strengthen the paper’s positioning.

Response: We have followed the suggestion and updated the contribution section of the manuscript to provide a more thorough engagement with the literature, ensuring a stronger positioning of our study.

2. How are the terms used in the textual analysis (e.g., climate-related keywords) mapped to specific opportunities or uncertainties? Are there risks of misclassification or noise in the data? The paper does not address the robustness of the textual analysis. For example, could alternative methods (e.g., dictionary refinement or supervised machine learning) yield consistent results? The time coverage of the conference call dataset and its representativeness for different industries and firm sizes are not discussed.

Response: The climate change exposure data used in this study was developed by Sautner et al. (2023). In their research, they conducted extensive validation tests to ensure the robustness of the measure. Specifically, they validated the variables at the bigram level by employing different sets of bigrams. Furthermore, they tested the Climate Change Exposure metric across multiple dimensions, including industry variation, time-series variation, and its correlation with firms’ carbon emissions. Additionally, they performed variance decomposition to further highlight the reliability and robustness of their climate change exposure measure. Their analysis provides detailed insights into the measure's applicability across industries and temporal trends.

To strengthen the robustness of our analysis, we present Table 1, which illustrates the sample distribution of our dataset by industry and year. This table provides a clear representation of the dataset's characteristics, including its industry coverage and year distribution, offering additional context and transparency for the analysis.

3. While the paper uses an IV approach to address endogeneity, the choice and validity of instruments are problematic: Weather variables are highly correlated with regional climate conditions but may lack direct relevance to firm-level cash-holding decisions. This undermines their ability to satisfy the exclusion restriction. The paper does not adequately explore alternative instruments, such as policy shocks or industry-specific climate change initiatives, which could provide stronger exogeneity.

Response: To address this concern, we conducted additional robustness tests, which are discussed in detail in Sections 4.6 and 4.7. First, we incorporated the Paris Agreement as a policy shock to further validate our findings. This approach allows us to examine whether a major global climate agreement, which is exogenous to individual firms, has consistent effects on their cash-holding behavior, reinforcing the robustness of our results.

Additionally, we conducted subsample analyses by excluding observations from periods of significant economic disruption, such as the Global Financial Crisis (GFC) and the COVID-19 pandemic. These periods are characterized by heightened economic uncertainty and potential anomalies in firm behavior that could distort our results. By isolating our analysis from these periods, we ensure that the observed relationships between climate change exposure and cash holdings are not driven by external, short-term shocks or macroeconomic fluctuations.

Together, these robustness checks demonstrate that our baseline results are consistent, reliable, and reflective of the long-term relationship between climate change exposure and firm-level cash-holding decisions, independent of temporary external disruptions or uncertainties. These efforts address the reviewer’s concerns and strengthen the validity of our findings.

4. Importantly, the paper identifies a positive relationship between climate opportunities and cash holdings but does not convincingly explain why this occurs. Key questions remain unanswered: Are firms stockpiling cash to capitalize on future climate-related investments, such as renewable energy projects? Alternatively, does this reflect inefficiencies in cash allocation driven by managerial conservatism under uncertainty?

Response: We have updated the discussion section to address the implications of our baseline results and to explore why firms increase cash holdings in response to climate change opportunities. To investigate this further, we conducted additional analyses examining the relationship between climate change opportunities and corporate investment. Our findings reveal a strong positive relationship between the two, indicating that firms with greater climate change opportunities are more likely to increase their investments in areas such as renewable energy, sustainable technologies, and other climate-related projects. This evidence suggests that the observed increase in cash holdings is primarily a strategic response, enabling firms to capitalize on future growth opportunities arising from climate change, rather than reflecting inefficiencies in cash allocation or managerial conservatism.

 

5. Regarding the robustness check. It would be useful to test the sensitivity of results to alternative climate change exposure measures (e.g., ESG scores, and third-party climate risk indices). Suggest conducting placebo tests, such as using unrelated keywords in the textual analysis, could help validate the exposure measures.

Response: In Section 4.3 of our analysis, we already employ alternative climate change measures to conduct robustness checks. Specifically, we focus on two measures: firm-level environmental risk and ESG ratings. Regarding the validation of the exposure measure, Sautner et al. (2023) conducted extensive validation tests to ensure its robustness. These include tests with different sets of bigrams, assessments of face validity for climate change bigrams, audit studies involving human reading, comparisons with pre-specified keyword approaches, perturbation tests for individual bigrams, and comparisons using initial bigrams only. These comprehensive tests confirm the reliability of their measure.

As we rely on the measure provided by Sautner et al. (2023) through their publicly available database, we were unable to access the raw textual data or replicate their validation processes. However, this measure is well-established in the academic literature. Given its rigorous validation and adoption in existing studies, we consider this measure robust for the purposes of our analysis.

6. The results section often describes findings without critically interpreting them. For instance, what do the magnitudes of the coefficients imply for firms’ financial strategies or climate adaptation efforts?

Response: We have also updated the results section to include a more detailed discussion of the implications of our baseline results. In particular, we now highlight how the findings relate to firms’ financial strategies and the potential for climate-related investments. Additionally, we present further analyses exploring the relationship between climate change opportunities and corporate investment. These analyses reveal a strong positive association, providing evidence that firms with higher climate change opportunities strategically increase their cash holdings to fund future investments in sustainable technologies, renewable energy projects, and other climate-related initiatives.

7. The discussion of policy implications is underdeveloped. The paper should address: How policymakers can use these findings to incentivize firms to allocate cash holdings toward sustainability initiatives rather than hoarding. Similarly, the literature review is too descriptive, summarizing studies without critically engaging with them. For example, the discussion of precautionary cash holdings does not address how this concept interacts with climate risks or opportunities.

Response: We have addressed the reviewer’s comments by expanding the discussion of how our findings contribute to policymaking in the conclusion section. Additionally, we have revised the literature review to include a critical discussion of existing studies and to highlight how our theoretical framework interacts with both climate change opportunities and risks. These revisions ensure that our contributions to the literature and policy implications are clearly articulated and effectively integrated into the manuscript.

 

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

 Thank you for clarifying in the second version, while the revised draft makes notable strides in addressing climate finance literature, several critical issues must be resolved to elevate the quality of this work for publication.

My main concern is still focused on the instrument part, as it is a strong methodology that captures the causal effects.

1. Incorporating the Paris Agreement as an exogenous shock is a reasonable approach to validate the study’s findings. However, the response does not sufficiently detail how this shock aligns with the research question and whether the Paris Agreement introduces heterogeneity in firm responses based on industry, region, or pre-existing exposure to climate risks and opportunities. The authors should clarify whether the Paris Agreement affected firms uniformly or if its impact varied across sectors, particularly those more exposed to climate-related regulations (e.g., energy, transportation, manufacturing).

2. Excluding periods of significant economic disruption, such as the Global Financial Crisis (GFC) and the COVID-19 pandemic, helps isolate the relationship between climate change exposure and cash holdings. However, this exclusion might also eliminate meaningful variation in firm behavior during periods of heightened uncertainty, which are precisely when climate-related risks and opportunities could be most salient.

3. The policy shock approach and subsample exclusions do not eliminate the potential for unobserved confounders that may correlate with both climate change exposure and cash holdings.

4. While the inclusion of lagged cash holdings provides some temporal robustness, the potential for reverse causality remains insufficiently addressed. Firms with high cash reserves may disproportionately engage in climate-related opportunities, reversing the hypothesized causal flow. A difference-in-differences design or matching approach could help disentangle these dynamics more convincingly.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf