Dynamic Capital Structure Adjustment: An Integrated Analysis of Firm-Specific and Macroeconomic Factors in Korean Firms

Round 1

Reviewer 1 Report

Based on the provided document, the article titled "Dynamic Capital Structure Adjustment: An Integrated Analysis of Firm-Specific and Macroeconomic Factors in Korean Firms" presents a comprehensive analysis of the impact of internal and external factors on the capital structure of Korean companies in the period from 1995 to 2021. The authors use a panel model and the Generalized Method of Moments (GMM) to solve the endogeneity problem resulting from the inclusion of lagged variables.

Strengths:

Scope of Study: The article covers an extensive period from 1995 to 2021, which includes various economic conditions such as stability and financial crises, allowing for an in-depth analysis of the adaptability of firms' capital structures.

Methodology: The use of a dynamic panel model and GMM estimation is appropriate for addressing endogeneity issues and allows for a more nuanced understanding of the dynamics of capital structure.

Integration of Factors: The authors successfully integrate firm-specific factors with macroeconomic ones, providing a holistic view of the determinants of capital structure.

Empirical Findings: The article provides intriguing findings, such as the varying influence of internal and external factors depending on economic conditions and a faster adaptation of capital structure during stable periods.

Weaknesses:

Market Specificity: While the study focuses on Korean firms, it could better highlight how the results can be generalized (or not) to other markets and economies.

Methodological Discussion: The article could discuss in more detail the limitations of the methodology used, including potential challenges associated with the choice of instruments in GMM estimation.

Comparative Analysis: A more detailed comparison with existing studies concerning other countries would be beneficial to better understand the uniqueness or similarities in the dynamics of capital structure of Korean firms.

Unclear Aspects:

Data Interpretation: Some of the interpretations of the data could be clarified, particularly how certain variables' influence on capital structure may differ across industries within the Korean market.

Assumptions and Sensitivity: The assumptions underlying the GMM methodology and the sensitivity of the results to these assumptions could be more explicitly addressed.

The literature review in the article provides a comprehensive overview of the main theories of capital structure, such as the trade-off theory and the pecking order theory. However, it could benefit from a more critical analysis of the existing literature, perhaps by highlighting gaps or inconsistencies in previous studies that the current research addresses. Additionally, the article could engage more with recent literature to position its findings within the context of the latest developments in the field.

Recommendations:

Expansion of Context: The authors could consider adding comparisons with firms from other countries to strengthen the universality of their conclusions.

Detailed Methodological Analysis: A more thorough examination and discussion of the limitations and assumptions of the GMM method, including the choice and sensitivity of instruments, would be valuable.

Future Research: The article could inspire further research focusing on the impact of new economic crises on firms' capital structures, especially in the context of the changing global financial landscape.

Author Response

Weaknesses:

Market Specificity: While the study focuses on Korean firms, it could better highlight how the results can be generalized (or not) to other markets and economies.

Our Answer: Thank you for your insightful comment on the generalizability of our study's findings beyond the context of Korean firms. We appreciate the opportunity to clarify and expand upon the broader applicability of our research. Our study meticulously examines the capital structure determinants of Korean firms across different economic periods, integrating both firm-specific and macroeconomic factors. While the immediate context is Korea—a country that has experienced a remarkable transition from an emerging to a developed economy—our research methodology and findings offer valuable insights that can be extrapolated to other markets, particularly those with similar economic and developmental trajectories.

Firstly, our use of a dynamic panel data model and the Generalized Method of Moments (GMM) estimation addresses the endogeneity issue, a common challenge in capital structure research. This methodological approach is applicable and valuable across various market contexts, providing a robust framework for analyzing the impact of both internal and external factors on firm capital structure. Secondly, the dual period analysis—comparing stable and crisis conditions—underscores the dynamic nature of capital structure management in response to changing economic environments. This aspect of our study is particularly relevant to economies undergoing rapid development or facing significant economic shifts, offering insights into how firms adapt their financial strategies in response to macroeconomic fluctuations. Furthermore, while our specific findings pertain to Korean firms, the underlying principles regarding the interplay between profitability, market conditions, and capital structure adjustments have broader implications. For instance, the increased significance of macroeconomic factors during crisis periods and the agility of firms in adjusting their capital structures in stable conditions may resonate with firms in other developed or rapidly developing economies.

In conclusion, while our study is grounded in the context of Korean firms, the methodologies employed and the insights gained regarding the determinants of capital structure and their dynamics across different economic periods have broader applicability. They contribute to the global discourse on corporate finance and capital structure management, offering valuable perspectives for researchers, policymakers, and practitioners in other markets and economies facing similar financial and economic challenges. We believe that highlighting these points in our study not only enriches the understanding of capital structure dynamics in Korea but also invites further research and discussion on the applicability of these findings in a global context.

Methodological Discussion: The article could discuss in more detail the limitations of the methodology used, including potential challenges associated with the choice of instruments in GMM estimation.

Our Answer: The Generalized Method of Moments (GMM) serves as a cornerstone in the econometric analysis of dynamic panel data, celebrated for its versatility in addressing endogeneity issues through instrumental variable techniques. Despite its widespread application across economic research, the methodology is not devoid of challenges, particularly in the selection and application of instruments. The following discussion delves into these limitations, offering a comprehensive examination of the challenges inherent in GMM estimation.

One of the primary challenges associated with GMM estimation lies in the selection and validation of instruments used to address endogeneity concerns. While we employed rigorous procedures to identify suitable instruments and conducted diagnostic tests to assess their validity, it is important to acknowledge that instrument choice is not without its limitations. Firstly, the effectiveness of instruments in addressing endogeneity relies on the validity of the underlying assumptions, such as instrument relevance and exogeneity. Despite our efforts to justify instrument validity through utilizing diagnostic tests, such as the Hansen J test or the Sargan-Hansen test, there may still exist unobserved lurking variable or measurement errors that compromise the instrumental variable strategy. Secondly, the potential presence of weak instruments poses a significant limitation to GMM estimation, leading to biased parameter estimates and inflated standard errors. Despite our efforts to mitigate this issue by employing diagnostic tests (sing measures like the Hansen J-statistic) and sensitivity analyses, the possibility of weak instrument bias cannot be entirely ruled out, particularly in datasets with limited variation in the instruments.

Instrument Validity and Selection: A paramount concern in GMM estimation is the rigorous selection of instruments. The validity of these instruments hinges on their ability to fulfill two critical conditions: relevance, where the instruments must exhibit a strong correlation with the potentially endogenous regressors, and exogeneity, where they must remain uncorrelated with the error term. The process of instrument selection, therefore, becomes a delicate balancing act. Misidentification or the use of invalid instruments can precipitate biased and inconsistent estimations, undermining the integrity of the research findings. The reliance on often arbitrary choices or weak instruments exacerbates this risk, making the validation of instrument appropriateness a pivotal aspect of the GMM methodology.

Over-identification and Weak Identification: The curse of over-identification arises when the number of instruments surpasses the number of endogenous variables. While theoretically advantageous for validating instrument integrity through over-identifying restrictions tests, in practice, it can engender weak identification issues. These issues manifest when the instruments, despite their abundance, fail to adequately inform on the endogenous variables, diluting the precision and reliability of parameter estimates. The inclusion of weak instruments, characterized by a tenuous correlation with the endogenous variables, further aggravates this predicament, potentially leading to misleading inference.

Testing for Over-identifying: Restrictions: The Sargan-Hansen tests for overidentifying restrictions are instrumental in assessing the exogeneity of the instruments. However, their effectiveness is contingent upon sample size and the strength of the instruments. In smaller samples, these tests suffer from low power, challenging the researcher's ability to conclusively ascertain the instruments' validity. This limitation underscores the necessity for caution in interpreting test results, particularly in studies constrained by sample size.

Finite Sample Bias: The asymptotic properties of GMM estimators, while robust in theory, do not immunize them against the biases present in finite samples. Small sample sizes can distort GMM estimators, leading to biased and inconsistent outcomes. This phenomenon necessitates a critical examination of the estimator's performance in finite samples, highlighting a significant limitation in studies with restricted data availability.

Complexity and Computational Demand: The implementation of GMM, especially within the framework of system GMM, is marked by computational intensity and methodological complexity. The estimator's sensitivity to specification choices demands meticulousness in model construction and an adherence to robustness checks. This complexity not only poses challenges in estimation but also in the interpretation and replication of results, emphasizing the importance of transparency and diligence in the application of GMM.

Ensuring the reliability and validity of findings in GMM-based research necessitates careful attention to instrument selection and validation, addressing over- and under-identification issues, and mitigating biases due to finite sample sizes. To mitigate concerns related to these aspects, we implemented a comprehensive testing strategy. This strategy included conducting the Sargan test or Hansen test for overidentifying restrictions, which helps verify that our instruments do not exhibit correlation with the error term, thereby ensuring their validity. Recognizing the critical role of sensitivity analysis in econometric research, we further conducted robustness checks. These checks were designed to evaluate how sensitive our results are to variations in model specifications and the choice of instruments. We varied the lag length and utilized alternative estimation methods, including the two-step GMM, to ensure the robustness of our findings. The consistency of our results across different tests bolsters our confidence in the conclusions drawn from our research. In response to your request, we have also added a paragraph at the end of our estimation methodology section. This paragraph, highlighted in red, outlines the limitations inherent to our methodology and details the robustness checks we performed to address these limitations, further solidifying the integrity of our research outcomes.

Comparative Analysis: A more detailed comparison with existing studies concerning other countries would be beneficial to better understand the uniqueness or similarities in the dynamics of capital structure of Korean firms.

Our Answer: To provide a detailed comparison with existing studies concerning other countries and better understand the uniqueness or similarities in the dynamics of capital structure of Korean firms, we can draw insights from various international studies on capital structure dynamics. Here's a summary and comparison based on the findings from different countries:

Transition Economies in Central and Eastern Europe:

Findings: Firms generally increased their leverage during the transition process, with profitability and age being robust determinants of capital structure targets. Banking system development has enabled firms to approach their leverage targets, though information asymmetries remain significant, leading to a preference for internal finance over bank debt (For example, von Hagen, J., & Zhou, J. (2005). The dynamic adjustment towards target capital structures of firms in transition economies. International Finance).

Developing Countries:

Findings: Capital structure choices in developing countries are influenced by the same variables as in developed countries, but persistent differences suggest that country-specific factors play a significant role (For example, Booth, L., Aivazian, V., Demirguc-Kunt, A., & Maksimovic, V. (2001). Capital structures in developing countries. Journal of Finance, 56, 87-130)..

Asia Pacific Region:

Findings: The capital structure decision of firms in Thailand, Malaysia, Singapore, and Australia is influenced by both the environment and firm-specific factors, with the 1997 financial crisis having a significant impact (For example, Deesomsak, R., Paudyal, K., & Pescetto, G. (2004). The determinants of capital structure: Evidence from the Asia Pacific region. Journal of Multinational Financial Management, 14, 387-405).

Global Perspective:

Findings: International determinants of capital structure show firm size, tangibility, industry leverage, profits, and inflation as reliable factors, with institutional quality affecting leverage adjustment speed. High-quality institutions lead to faster leverage adjustments, highlighting the importance of the legal and financial environment (For example, Öztekin, Ö. (2013). Capital structure decisions around the world: Which factors are reliably important? Journal of Financial and Quantitative Analysis, 50, 301-323).

Comparison: Compared to existing capital structure research across various countries, our study zeroes in on the determinants within the Korean context, offering a broader and more in-depth analysis through both firm-specific and macroeconomic lenses. By examining an extended timeline from 1995 to 2021, we innovatively segment this period into 'normal' and 'crisis' phases to scrutinize the distinct behaviors of capital structure across these times. While our primary focus diverges from an international comparative analysis, we acknowledge its relevance through a footnote reference, thereby situating our findings within a broader scholarly dialogue without dedicating extensive space to direct comparison. This approach enriches the discourse by providing a comprehensive view of capital structure dynamics specific to Korea, marked by a significant period analysis and a nuanced classification that enhances understanding of financial strategies during varied economic states.

Newly Added Footnote 2: The scholarly field is rich with analyses on the factors influencing capital structure in diverse national contexts. Although our research encompasses a broad spectrum of these determinants, as outlined in prior studies, our aim diverges from cross-country comparisons. We specifically explore the determinants of capital structure within South Korea from 1995 to 2021. Our innovative approach includes dividing this timeframe into 'normal' and 'crisis' periods, enabling us to investigate the unique capital structure dynamics in each phase.

Unclear Aspects:

Data Interpretation: Some of the interpretations of the data could be clarified, particularly how certain variables' influence on capital structure may differ across industries within the Korean market.

Our Answer: In our study, we indirectly analyzed how various factors influence capital structure across different industries in the Korean market. This was achieved through the segmentation of our dataset to reflect industry-specific dynamics, allowing for a nuanced understanding of capital structure determinants. Our findings reveal that certain variables, such as profitability and leverage, exhibit varying degrees of impact across industries, underscoring the heterogeneous nature of capital structure decisions in response to both firm-specific and macroeconomic conditions. For instance, industries with higher asset tangibility showed a propensity for increased leverage, aligning with theoretical expectations regarding collateral value and debt capacity. However, to enhance clarity on the influence of variables across industries, the paper could benefit from a more detailed discussion on the following aspects, based on the empirical results section:

Industry-specific Influences: The empirical analysis identified several variables with statistically significant impacts on capital structure. For example, the analysis found variables like Return on Assets (ROA), Current Ratio (CR), Price to Book Ratio (PBR), and Ratio of Fixed Assets (FixR) to be significant. A nuanced discussion on how these factors play out across different industries within the Korean market would elucidate the differential impact on capital structure. This could involve comparing industries with high R&D intensity against those more focused on capital expenditure or operational efficiencies.

Macroeconomic Variables Across Industries: The paper highlights the significant impact of macroeconomic variables such as GDP rate, inflation rate, credit premium, and term premium on capital structure. An industry-specific breakdown of these impacts could provide insights into which sectors are more sensitive to economic cycles or external financial conditions. This is particularly relevant for industries that are either highly cyclical or have varying levels of exposure to international markets.

Adjustment Speed Variability by Industry: The analysis of adjustment speeds towards target leverage indicates that firms correct their capital structure deviations at varying rates. Expanding on this by industry could clarify if certain sectors adjust more rapidly due to industry norms, regulatory environments, or market pressures. This would offer a deeper understanding of strategic financial management practices across different segments of the Korean economy.

Crisis Period Analysis by Industry: The study delves into capital structure dynamics during crisis and non-crisis periods. An industry-level analysis of these periods could highlight specific challenges or resilience factors inherent to certain sectors. For instance, industries such as technology and healthcare might exhibit different leverage dynamics in crisis periods compared to more traditional sectors like manufacturing or utilities.

To conduct the industry analysis as described, however, the study must reorganize the original dataset by industry classification and undertake empirical analysis for each distinct industry within Korea. This approach may encounter challenges, such as small sample sizes in certain industries, which could introduce bias. Addressing these issues within the constraints of time and resources available is not feasible. Furthermore, the primary aim of our research is to explore capital structure variations between 'normal' and 'crisis' periods, not industry-specific behaviors. Consequently, this topic is suggested for future research, emphasizing the potential insights from comparing capital structure dynamics across different industries during stable and turbulent times.

New Addition in Conclusion: Furthermore, exploring variations in capital structure dynamics across diverse industries during periods of stability and turbulence could provide another intriguing avenue for research.  

Assumptions and Sensitivity: The assumptions underlying the GMM methodology and the sensitivity of the results to these assumptions could be more explicitly addressed.

Our Answer: We recognize the importance of clearly articulating the assumptions underlying our econometric approach and the potential impact of these assumptions on our results.

The GMM methodology is predicated on several key assumptions, including the instrument relevance and validity, which are crucial for ensuring the consistency and efficiency of the estimators. In our study, we carefully selected instruments based on their theoretical relevance to the model and empirical evidence suggesting their exogeneity. To address potential concerns regarding these assumptions, we conducted a series of tests, including the Sargan test (or Hansen test of overidentifying restrictions) to evaluate the validity of our instruments, ensuring they are not correlated with the error term.

Recognizing the potential for heteroscedasticity in our data, we employed robust GMM estimation using the using STATA GMM estimation command xtabond2 and specifying the robust option to calculate robust standard errors which helps in addressing heteroscedasticity. It is important to acknowledge that no estimator is entirely immune to heteroscedasticity.

Furthermore, we acknowledge the importance of sensitivity analysis in econometric research. To this end, we performed robustness checks to assess the sensitivity of our results to different model specifications and instrument sets. This included varying the lag length and employing alternative estimators, such as the two-step GMM, to verify the robustness of our findings. Our results remained consistent across these tests, reinforcing our confidence in the reliability and validity of our conclusions.

In response to your request, we have added a paragraph at the end of our estimation methodology section. This paragraph, highlighted in red, outlines the limitations inherent to our methodology and details the robustness checks we performed to address these limitations, further solidifying the integrity of our research outcomes.

The literature review in the article provides a comprehensive overview of the main theories of capital structure, such as the trade-off theory and the pecking order theory. However, it could benefit from a more critical analysis of the existing literature, perhaps by highlighting gaps or inconsistencies in previous studies that the current research addresses. Additionally, the article could engage more with recent literature to position its findings within the context of the latest developments in the field.

Our Answer: The trade-off and pecking order theories provide static frameworks for understanding capital structure, positing that firms aim for an optimal leverage ratio (trade-off theory) or prioritize financing sources based on a hierarchy to minimize costs (pecking order theory). Dynamic capital structure theory, however, views capital structure as an evolving process, where firms continuously adjust their leverage in response to changing internal and external conditions to reach or maintain their optimal structure. This perspective acknowledges the temporal dimension of financing decisions, incorporating the insights from both static theories to explain how firms adaptively manage their debt and equity mix over time, aiming to balance the benefits and costs of debt dynamically rather than achieving a once-and-for-all optimal structure. We are of the opinion that the literature review has already sufficiently covered this point. Nonetheless, in light of suggestions to incorporate more recent studies, we have included an additional three references in the literature review section and additional one in the empirical analysis section. These new entries and their concise summaries are highlighted in red for easy identification.

Four newly added references:

1. Faulkender, M., & Petersen, M. A. 2014. Investment and capital constraints: Repatriations under the American Jobs Creation Act. Review of Financial Studies, 27(8), 2406-2449.
2. Lemmon, E. D., & Zender, J. (2016). Asymmetric information, debt capacity, and capital structure. Journal of Financial Economics, 120(2), 251-274.
3. Memon, P. A., Mid-Rus R., & Ghazali, Z. B., 2021, Adjustment speed towards target capital structure and its determinants, Economic Research, 34(1), 1966-1984.
4. Shikimi M. 2020. Bank loan supply shocks and leverage adjustment. Economic Modelling 87: 447–460.

Recommendations:

Expansion of Context: The authors could consider adding comparisons with firms from other countries to strengthen the universality of their conclusions.

Our Answer: To enhance our understanding of the nuances or commonalities in the capital structure dynamics of Korean firms in comparison with those in other countries, we engage with a variety of international studies on the dynamics of capital structure. Unlike prior research that spans multiple nations, our investigation specifically targets the factors influencing capital structure within the South Korean context. This approach allows for a comprehensive and detailed exploration through the prism of both firm-level and macroeconomic variables. Our analysis, covering the period from 1995 to 2021, innovatively divides this timeframe into 'normal' and 'crisis' periods, enabling us to examine the distinctive capital structure behaviors in these distinct phases. Although our study primarily concentrates on the internal determinants of capital structure in Korea, rather than an international comparative analysis, we recognize the importance of situating our work within the global academic discourse. This acknowledgment is made through a footnote that references the relevant international studies, thereby integrating our findings into the wider body of scholarly work without allocating substantial text to direct comparisons.

Newly Added Footnote 2: The scholarly field is rich with analyses on the factors influencing capital structure in diverse national contexts. Although our research encompasses a broad spectrum of these determinants, as outlined in prior studies, our aim diverges from cross-country comparisons. We specifically explore the determinants of capital structure within South Korea from 1995 to 2021. Our innovative approach includes dividing this timeframe into 'normal' and 'crisis' periods, enabling us to investigate the unique capital structure dynamics in each phase.

Detailed Methodological Analysis: A more thorough examination and discussion of the limitations and assumptions of the GMM method, including the choice and sensitivity of instruments, would be valuable.

Our Answer: Your feedback accentuates the imperative of maintaining methodological transparency and conducting a critical examination of econometric methodologies—a stance that we emphatically endorse. In the preparation of our manuscript, we were diligent in elucidating our methodological framework with the intent to provide a thorough and intelligible presentation. This effort was characterized by a judicious selection of instruments, informed by theoretical rationales and empirical validations, to fortify the robustness of our Generalized Method of Moments (GMM) estimations.

We elaborated on the foundational assumptions integral to our approach, including the validity of instruments and the non-existence of serial correlation in the error terms, which are pivotal for ensuring the reliability of GMM-derived outcomes. Furthermore, we undertook comprehensive sensitivity analyses, aimed at evaluating the resilience of our results to alterations in model specifications and the composition of instrument sets. The outcomes of these analyses, detailed in the results section of our manuscript, corroborate the stability and dependability of our conclusions across a spectrum of model specifications. As previously mentioned, we have incorporated an additional paragraph at the conclusion of our estimation methodology section. Highlighted in red, this paragraph delineates the limitations associated with our methodology and elucidates the robustness checks conducted to mitigate these limitations, thereby enhancing the credibility of our research findings.

Given the depth of methodological detail and the extent of robustness evaluations included in our manuscript, we are of the opinion that the discourse strikes an optimal balance between comprehensiveness and succinctness. This balance not only ensures the manuscript's accessibility but also facilitates its readability, thereby advancing the scholarly discourse in econometric research methodologies.

Future Research: The article could inspire further research focusing on the impact of new economic crises on firms' capital structures, especially in the context of the changing global financial landscape.

Our Answer: We concur with the assessment that our study lays the groundwork for subsequent investigations into how emerging economic crises influence the capital structures of firms, particularly within the evolving global financial context. Additionally, a comparative analysis of the capital structure dynamics between Korean firms and those operating in different economies may yield deeper understanding of worldwide financial methodologies. Moreover, examining the differential impacts of capital structure dynamics across various industries during periods of economic stability and volatility presents a compelling research direction. These suggestions have been articulated in the conclusion section as potential areas for future research, with the newly added content distinctly marked in red.

Reviewer 2 Report

The manuscript is a short, well-designed empirical paper on the capital structure of Korean non-financial companies in 1995–2021 based on a dynamic panel model. Unlike the previous research on the topic, the authors include both microeconomic and macroeconomic variables in their model and subdivide the examined period into "normal" and "crisis" years, which seems to be the first study on capital structure employing such an approach. Given the overall soundness of the research, the following improvements are recommended for the paper to be published.

Firstly, one could mention a third "theory" for capital structure, Hyman Minsky's financial instability hypothesis. While not intended to explain the optimum capital structure of a company per se, it nonetheless does cover the motives for speculative and Ponzi finance, which are linked to macroeconomic factors. Minsky claims that during the "good times", companies tend to borrow more, becoming less risk-averse.

Secondly, the choice of microeconomic and microeconomic variables is not explained in sufficient detail in the literature review section (as expected) but in the model section. The easiest solution would be to add a link to this explanation in the literature review.

Thirdly, the default spread is defined as a difference between AAA-rated and BBB-rated corporate bonds. The rating scale of which company was selected for this purpose? A simple reference to S&P would be enough.

Fourthly, the system (Blundell-Bond) GMM estimator should be explained in the estimation methodology section since it is later used in the paper. Currently, only the difference (Arellano-Bond) estimator is covered in any detail. Also, a reference to StataCorp's website is hardly sufficient to explain the IPS panel unit root test.

Finally, the paper has no discussion section except a brief link to Tang (2010) to compare results. A deeper comparison with/to the literature review is required to justify the significance of the obtained results and the value added of the paper.

One possible misprint in the text is line 332 "lack of insufficient data". Maybe a "lack of sufficient data" is meant?

Author Response

Firstly, one could mention a third "theory" for capital structure, Hyman Minsky's financial instability hypothesis. While not intended to explain the optimum capital structure of a company per se, it nonetheless does cover the motives for speculative and Ponzi finance, which are linked to macroeconomic factors. Minsky claims that during the "good times", companies tend to borrow more, becoming less risk-averse.

Our Answer: Hyman Minsky's financial instability hypothesis posits that financial markets are inherently unstable due to the cyclical nature of investment and financing decisions. Specifically, during periods of economic prosperity, firms become increasingly optimistic, leading to higher levels of borrowing. Minsky categorizes this borrowing behavior into three regimes: hedge finance, speculative finance, and Ponzi finance, with speculative and Ponzi finance being indicative of increased risk-taking and potential vulnerability to financial instability.

While traditional capital structure theories, such as the Modigliani-Miller theorem, trade-off theory, and pecking order theory, primarily focus on the optimization of a firm's debt and equity mix under varying conditions of market imperfections, taxes, bankruptcy costs, and information asymmetry, Minsky's hypothesis introduces a macroeconomic perspective to the discussion. It suggests that the broader economic environment and the cycle of financial conditions significantly influence firms' financing decisions beyond the micro-level considerations of capital cost minimization and financial structure optimization. Minsky's distinction between speculative and Ponzi finance is particularly relevant to understanding how firms' capital structures evolve in response to macroeconomic signals. During "good times," firms may engage in speculative finance, where they borrow with the expectation that cash flows will cover interest payments but not the principal, or even Ponzi finance, where cash flows do not cover either, in anticipation of asset appreciation. This behavior aligns with the notion that firms become less risk-averse in favorable economic conditions, potentially leading to over-leveraging and increased vulnerability to subsequent economic downturns. Incorporating Minsky's financial instability hypothesis into the analysis of capital structure dynamics offers a comprehensive view that encompasses both individual firm-level strategies and the influence of macroeconomic trends on these strategies. Empirically, this perspective could be explored through the examination of leverage ratios, financing behaviors, and default rates across different economic cycles, thereby providing insights into the practical manifestations of speculative and Ponzi finance phenomena in varying industry contexts and macroeconomic conditions.

While Minsky's financial instability hypothesis does not directly propose an optimal capital structure, its implications for understanding the cyclicality of borrowing behaviors and the potential for financial instability are valuable. By integrating Minsky's macroeconomic lens with traditional capital structure theories, scholars and practitioners can gain a deeper understanding of the complex interplay between firm-level decisions and the broader economic environment in shaping capital structure dynamics. We acknowledge the significance of Minsky's hypothesis in enriching the capital structure discourse. It bridges the gap between micro-level capital structure decisions and macroeconomic conditions, offering a holistic framework for analyzing financial strategies in different economic phases. Integrating Minsky's insights could enhance the predictive power of capital structure models, especially in understanding the conditions under which firms might adopt more aggressive financing strategies that increase their exposure to financial instability. Thank you for your insightful comment.

Secondly, the choice of microeconomic and microeconomic variables is not explained in sufficient detail in the literature review section (as expected) but in the model section. The easiest solution would be to add a link to this explanation in the literature review.

Our Answer: In the model delineated in the subsequent section, we have incorporated all variables identified during the literature review, which encompasses both firm-specific and macroeconomic factors. It is crucial to emphasize that these variables are acknowledged determinants of capital structure and have been widely employed in a plethora of previous studies focusing on capital structure. For further elucidation, a footnote has been appended accordingly.

Newly added footnote 1: In the forthcoming section, our proposed model integrates all variables outlined in the literature review, including both firm-specific and macroeconomic factors. It is noteworthy to mention that these variables, identified as key determinants of capital structure, have been extensively applied in a broad array of previous capital structure research.

Thirdly, the default spread is defined as a difference between AAA-rated and BBB-rated corporate bonds. The rating scale of which company was selected for this purpose? A simple reference to S&P would be enough.

Our Answer: Our dataset was sourced from FnGuide, a premium data provider based in Korea. The default spread, defined as the yield differential between AAA and BBB rated bonds, is pre-calculated by FnGuide. This measure is pertinent to companies listed on the Korea Stock Exchange (KSE), where over 700 companies are actively listed and traded. It is generally observed that the average credit rating among these companies is BBB, with leading Korean corporations, including Samsung, potentially achieving AAA ratings.

Fourthly, the system (Blundell-Bond) GMM estimator should be explained in the estimation methodology section since it is later used in the paper. Currently, only the difference (Arellano-Bond) estimator is covered in any detail. Also, a reference to StataCorp's website is hardly sufficient to explain the IPS panel unit root test.

Our Answer: In response to your feedback, we recognize the necessity of briefly explaining the distinction between the difference (Arellano-Bond) and system (Blundell-Bond) GMM estimators in our estimation methodology section. The system Generalized Method of Moments (GMM) estimator, developed by Blundell and Bond, is an extension of the original difference GMM estimator proposed by Arellano and Bond. The key innovation of the system GMM is its ability to combine equations in differences with equations in levels into a system, thus improving efficiency and addressing some of the limitations inherent in the difference GMM estimator. The difference GMM estimator is designed to eliminate unobserved panel-specific effects by differencing the data, using lagged levels of the variables as instruments for the differenced equations. However, this approach can lead to weak instruments, especially when the variables exhibit persistence over time, leading to biased and inefficient estimates. The system GMM estimator enhances this by adding an additional set of equations that use the lagged differences of the variables as instruments for the equation in levels. This approach helps in mitigating the potential biases and inefficiencies, especially in the presence of persistent variables. By incorporating more information, the system GMM provides more efficient and reliable estimates under certain conditions. A paragraph recently incorporated into our estimation methodology section, intended to address this point, has been distinctly marked in red for ease of identification.

In light of the feedback received on our discussion of the IPS panel unit root test, we acknowledge and value the suggestion to extend our exposition beyond merely directing readers to StataCorp's website for further information. Our initial goal was to achieve a judicious balance between brevity and thoroughness. Recognizing the merit of furnishing a succinct yet comprehensive delineation of the test, including its applicability and significance to our analytical framework, we concur that enhancing the depth of our explanation would augment the manuscript's academic rigor. We maintain confidence in the integrity and methodological solidity of our study. Furthermore, in an effort to heed your counsel, we have revised our approach by omitting the reference to Stata and instead, directly citing the seminal work by Im, K.S., Pesaran, M.H., and Shin, Y. (2003) titled 'Testing for Unit Roots in Heterogeneous Panels' from the Journal of Econometrics, 115, pp. 53-74. This citation not only serves to substantiate our discussion with authoritative grounding but also enriches the scholarly foundation of our paper. The said reference has been duly incorporated into our list of citations.

Thank you once again for your insightful feedback, which contributes to the scholarly discourse and the ongoing refinement of research methodologies.

Finally, the paper has no discussion section except a brief link to Tang (2010) to compare results. A deeper comparison with/to the literature review is required to justify the significance of the obtained results and the value added of the paper.

Our Answer: The initial reference to Tang (2010) was incorrect and should have been accurately cited as Cook and Tang (2010). We appreciate the correction of this oversight. Following this amendment to Cook and Tang (2010) within our literature review and in the model, we have thoroughly detailed their investigation into the dynamics of capital structure adjustment processes, including a comparison of their methodology with ours. Despite the methodological differences, our findings echo those of Cook and Tang (2010), who found that the speed of adjustment in capital structures is faster during periods of economic prosperity and slower during less favorable economic conditions, based on their analysis of U.S. data. Our analysis, utilizing Korean data, reveals a similar trend. To further elucidate this point, we have incorporated a new paragraph highlighting this parallel in our findings in the sectionof 5. Empirical Results.

Newly added paragraph: Cook and Tang (2010), through their analysis of U.S. data, demonstrate that the speed of adjustment is quicker during prosperous periods compared to slower adjustments in less favorable times. Correspondingly, our model, which utilizes data from Korea, exhibits a similar pattern of behavior. Additionally, Shikimi (2020) provides evidence from Japan indicating that firms tend to adjust more rapidly during normal periods as opposed to periods characterized by credit constraints.

One possible misprint in the text is line 332 "lack of insufficient data". Maybe a "lack of sufficient data" is meant?

Our Answer: Indeed, the correct phrasing should be "a lack of sufficient data." Thank you for addressing this typographical error.

Reviewer 3 Report

Article Review: Dynamic Capital Structure Adjustment: An Integrated Analysis 2 of Firm-Specific and Macroeconomic Factors in Korean Firms.

The topic of the article is one that has been widely studied and researched. It is always of academic and business interest, especially when it brings new and different insigts, allowing us to contribute to the evolution and understanding of the topic. My congratulations to the authors for their choice.

To help improve the work, I would like to make the following suggestions:

- In the abstract, in addition to presenting the time frame of the study, the authors should inform the readers of the number and type of companies that are part of the sample, providing them with relevant information, as the composition of the sample is relevant to the interpretation and understanding of the results. The sample includes non-financial companies on the Korean stock exchange.

In the introduction, the authors present the aim of the work as well as describing how it differs from existing work. However, I think that they should restate the previous information (company number and type), as I think that this point is unclear throughout the paper. I also think that the authors should present the structure of the development of the work.

I consider the literature review to be a key point in a scientific paper. It allows the studies, methods, results and conclusions of previous research to be presented, guiding and justifying the following work.

I think the authors should present more up-to-date results and evidence, which would make the discussion of the results more relevant and up-to-date.

The authors only present a paper from 2019, i.e. from the last five years. I believe that they do not present a current and relevant review, even in justifying the choice of variables, whether internal or macroeconomic, which are widely used in the literature and with evidence.

The authors should present an approximate ratio of at least 30% of references from the last 5 years, correctly 50%. This gives the work relevance and timeliness.

The methodology, I consider to be possible, but it is a correct methodology, the authors validate assumptions and test the robustness of the model, and could also mention some more tests such as validation of muticolinarity.

The data is presented and discussed, and the authors can draw some parallels with previous evidence from other researchers. I think this would add value to the paper.

In the conclusion, the authors describe the most obvious findings, but they should present the limitations and future directions.

The references should be updated, as mentioned in the review, trying to increase the ratio of current work (last 5 years).

Author Response

- In the abstract, in addition to presenting the time frame of the study, the authors should inform the readers of the number and type of companies that are part of the sample, providing them with relevant information, as the composition of the sample is relevant to the interpretation and understanding of the results. The sample includes non-financial companies on the Korean stock exchange.

Our Answer: Per your request, the initial sentence in the abstract, "This study explores the capital structure determinants of Korean firms over an extensive period (1995-2021)," has been revised to: "This research investigates the factors influencing the capital structure of 271 non-financial firms listed on the Korean Stock Exchange (KSE) over a broad period from 1995 to 2021."

In the introduction, the authors present the aim of the work as well as describing how it differs from existing work. However, I think that they should restate the previous information (company number and type), as I think that this point is unclear throughout the paper. I also think that the authors should present the structure of the development of the work.

 Our Answer:

1. Per your request, the initial sentence in the Introduction, "By segmenting the analysis this way, we offer a nuanced exploration of how Korean firms' capital structure strategies are tailored to different economic conditions," has been revised to: "Through this segmented analysis, we provide a detailed examination of how 271 non-financial firms listed on the KSE adapt their capital structure strategies to various economic environments.”
2. Responding to your further request, we have also included an overview of our work's structure at the end of the Introduction as follows: The following section will commence with the Literature Review, succeeded by discussions on the Model, Data, and Methodology. This will be followed by the presentation of Empirical Results, and the paper will conclude with the Conclusion section.

I consider the literature review to be a key point in a scientific paper. It allows the studies, methods, results and conclusions of previous research to be presented, guiding and justifying the following work.

I think the authors should present more up-to-date results and evidence, which would make the discussion of the results more relevant and up-to-date. The authors only present a paper from 2019, i.e. from the last five years. I believe that they do not present a current and relevant review, even in justifying the choice of variables, whether internal or macroeconomic, which are widely used in the literature and with evidence. The authors should present an approximate ratio of at least 30% of references from the last 5 years, correctly 50%. This gives the work relevance and timeliness.

Our Answer: Pursuant to your request, we have incorporated four additional references into the manuscript, allocating three to the Literature Review and one to the Empirical Results section (these additions are highlighted in red within the document). Nevertheless, we recognize that our efforts may not fully align with your expectations. We acknowledge the guideline suggesting that approximately 30%, ideally 50%, of references should be from the last five years to ensure the work's relevance and currency. However, we also maintain that the references must be pertinent to our study. As outlined in our paper, our research builds upon the foundational work of Cook & Tang (2010). Given the limited advancements in this area since their publication, we faced challenges in identifying recent studies that directly align with our research focus. This scarcity of directly relevant recent literature underscores the innovative nature of our study.

The methodology, I consider to be possible, but it is a correct methodology, the authors validate assumptions and test the robustness of the model, and could also mention some more tests such as validation of muticolinarity.

Our Answer: We agree that the methodology employed in this study is both possible and correct, as it was carefully designed to validate assumptions and test the robustness of the model. We have indeed put in significant effort to ensure that their methodology is sound. Your recommendation to incorporate tests for multicollinearity is appreciated. While the paper does not explicitly address this aspect, recognizing multicollinearity is essential for ensuring the independence of variables, thereby influencing the reliability of the model's results. However, it's worth noting that the independent variables used in our study closely align with those in our primary reference, Cook & Tang (2010), as well as numerous previous studies. Also, while System GMM can handle near-integrated regressors, severe collinearity can still affect its efficiency and potentially lead to biased estimates, For System GMM, the primary concern is ensuring instrument validity rather than collinearity among regressors [1-3]. Therefore, we are confident that any potential multicollinearity issues have likely been assessed and addressed in previous research.

• Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of econometrics, 87(1), 115-143.
• Bond, S. R. (2002). Dynamic panel data models: a guide to micro data methods and practice. Portuguese economic journal, 1, 141-162.
• Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT press.

The data is presented and discussed, and the authors can draw some parallels with previous evidence from other researchers. I think this would add value to the paper.

Our Answer: The book-value based data is available as a PDF file upload. The market-value based data can be provided upon request.

In the conclusion, the authors describe the most obvious findings, but they should present the limitations and future directions.

Our Answer: In conclusion, this paper endeavors to elucidate the most salient findings while also delineating the study's limitations and suggesting avenues for future research. As requested, any further amendments made to the manuscript are prominently marked in red in the concluding section.

The references should be updated, as mentioned in the review, trying to increase the ratio of current work (last 5 years).

Our Answer: The bibliography has been augmented to include four additional recent publications.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

My comments have been taken into account