Review Reports

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

This manuscript examines the impact of land misallocation on regional economic resilience in China, emphasizing the mediating role of technological progress and spatial spillover effects. The topic aligns well with the scope of Land. Some specific suggestions are as follows:

1. While the measurement of economic resilience follows established literature, the manuscript would benefit from a slightly clearer distinction between “resistance” and “recovery” dimensions of resilience in the conceptual discussion.
2. The construction of the spatial economic weight matrix based on GDP differences and geographic distance is reasonable. However, a short intuitive explanation should be added to clarify why economic similarity is particularly relevant for technological spillovers and resilience interactions among cities.
3. The robustness checks are comprehensive, but their discussion is relatively brief. A short paragraph summarizing why these tests reinforce confidence in the baseline results would enhance readability.
4. Minor grammatical and stylistic issues remain, particularly in longer sentences in Sections 2 and 3. A careful language polish would improve readability and conciseness.

Author Response

Comments 1. The construction of the spatial economic weight matrix based on GDP differences and geographic distance is reasonable. However, a short intuitive explanation should be added to clarify why economic similarity is particularly relevant for technological spillovers and resilience interactions among cities.

Response: We appreciate the reviewer's valuable suggestion. In response, we have supplemented Section 3.2.1 with an explanatory note clarifying that technological diffusion tends to occur more readily between cities of comparable economic development levels, as they typically share similar knowledge absorption capacities and industrial structures. Moreover, cities of analogous economic scale often encounter comparable exposure to external shocks and operate within similar policy environments. This addition enhances the reader's understanding of the underlying logic behind the construction of the spatial weight matrix.

Comments 2. While the measurement of economic resilience follows established literature, the manuscript would benefit from a slightly clearer distinction between “resistance” and “recovery” dimensions of resilience in the conceptual discussion.

Response: We thank the reviewer for this important reminder. In response, we have supplemented Section 3.1.2 with an explanation that clarifies the distinction between the "resistance" and "recovery" dimensions of economic resilience. Furthermore, we note that the model employed in this study, based on Martin (2012), has been widely validated in the existing literature as effectively capturing both dimensions. This addition serves to justify the appropriateness of our chosen measurement approach.

Comments 3. The robustness checks are comprehensive, but their discussion is relatively brief. A short paragraph summarizing why these tests reinforce confidence in the baseline results would enhance readability.

Response: We appreciate the reviewer's suggestion. In response, we have added a concluding paragraph at the end of Section 4.2. This summary explicitly states that the core findings of this study remain robust after systematically controlling for issues such as endogeneity, structural bias, and specification bias in the spatial matrix. This reinforces the credibility and practical relevance of our research results.

Comments 4. T Minor grammatical and stylistic issues remain, particularly in longer sentences in Sections 2 and 3. A careful language polish would improve readability and conciseness.

Response: We thank the reviewer for this comment. In response, we engaged a native English speaker to perform a professional language polish on the entire manuscript, with particular attention to refining the clarity of expression in longer sentences, especially those in Chapters 2 and 3.

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

This paper examines how land misallocation affects regional economic resilience in China using a theoretical model and spatial econometric analysis of 95 cities. The topic is timely and relevant, and the empirical results are generally robust. However, several substantive issues need to be addressed to strengthen clarity, rigor, and interpretation.

Major Comments: The paper should better distinguish between resistance, recovery, and adaptation, and justify how the chosen measure captures these dimensions. The price ratio proxy requires clearer justification and a more explicit discussion of its limitations and what forms of misallocation it does not capture.

Key assumptions (e.g. free factor mobility, short-term technological rigidity) are strong and should be discussed more explicitly in terms of realism and implications. The GDP-distance-based matrix needs stronger justification or additional robustness checks using alternative spatial specifications. Reliance on patent counts as a proxy for technological level should be more critically discussed, including potential biases.

The manuscript should clarify which exogenous shocks underpin the resilience measure and how shock periods are defined. The authors should better delimit which findings are China-specific and which may generalise to other contexts.

Author Response

Comments 1. The paper should better distinguish between resistance, recovery, and adaptation, and justify how the chosen measure captures these dimensions. The price ratio proxy requires clearer justification and a more explicit discussion of its limitations and what forms of misallocation it does not capture.

Response: We thank the reviewer for the important suggestions. Regarding the use of the price ratio as a proxy variable for the independent variable (land misallocation), we have supplemented Section 3.1.1 with a discussion of its rationale, as suggested. We have also included a discussion on the limitations of this variable. Concerning the measurement of the dependent variable (economic resilience), we have added an explanation in Section 3.1.2 that clarifies the distinction between the "resistance" and "recovery" dimensions. Furthermore, we note that the selected measurement model is widely adopted in the existing literature, which justifies its use in this study.

Comments 2. Key assumptions (e.g. free factor mobility, short-term technological rigidity) are strong and should be discussed more explicitly in terms of realism and implications. The GDP-distance-based matrix needs stronger justification or additional robustness checks using alternative spatial specifications. Reliance on patent counts as a proxy for technological level should be more critically discussed, including potential biases.

Response: We appreciate the reviewer's valuable suggestions. In response, we have made the following three key revisions: (1) Regarding the justification of research assumptions: following the reviewer's advice, we have supplemented Section 2.2 with a discussion that clarifies the distinction between the theoretical assumptions employed in this study and real-world applications, while also providing a rationale for these assumptions. (2) Regarding the justification of the spatial matrix specification: as suggested by the reviewer, we have added an explanation in Section 4.2. Specifically, we note that a robustness check was conducted by re-estimating the model using an economic matrix that excludes geographical factors (the results are presented in Appendix B). The findings from this check are consistent with the main conclusions of the paper. (3) Regarding the limitations of using patent counts as a proxy for technological level: we have expanded Section 4.3 to include a detailed discussion of this issue. Furthermore, we have provided additional justification for the rationale behind selecting this proxy variable in the current research context.

Comments 3. The manuscript should clarify which exogenous shocks underpin the resilience measure and how shock periods are defined. The authors should better delimit which findings are China-specific and which may generalise to other contexts.

Response: We appreciate the reviewer's comments. Regarding the definition of shocks: this study utilizes annual panel data and does not define a specific shock period for a single exogenous event (such as a financial crisis). Instead, each year is treated as an observational unit potentially subject to various unexpected shocks. Resilience is captured by measuring the deviation of actual growth from the national trend. This approach, employed in studies such as Martin (2012), is widely used to capture the effects of shocks from multiple periods and sources. Concerning the generalizability of the conclusions: we have clarified in Section 5 the distinctions between findings that are specific to the Chinese context and those with potential broader applicability, thereby enhancing the value of this research for an international audience.

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors Detailed remarks from the review     The authors have prepared an interesting topic that fits well within the scope and focus of the journal.

1. In the introduction, the research objective (main goal) has been precisely defined, but it would be advisable to also specify the detailed objectives. This would certainly strengthen the importance of the research topic. It may also be worth considering whether to expand the introduction with research hypotheses.
2. 4. Empirical Analysis – The authors assumed that the research area consists of 95 Chinese cities. I suggest that these cities be indicated graphically or described in more detail. I do not share the opinion that the fact that these cities account for 70% of agricultural production automatically makes them representative (line 451).
3. The data sources used are described too generally: China Land and Resources Statistical Yearbook and the Wind databases. It would be advisable to at least present the general developmental trends of these variables. I consider that this part of the article should be supplemented.
4. The Conclusion section needs to be expanded. At present, it contains only general recommendations for local and regional economic policy. The authors studied only part of the country – are there studies concerning other regions of China? What recommendations can be made in this regard?

Other remarks:

1. Literature – the review of previous studies by other authors should be supplemented, including a broader review of international literature, especially from regions outside China.
2. I recommend adding an appendix that would, for example, provide characteristics of the variables used.

Author Response

Comments 1. In the introduction, the research objective (main goal) has been precisely defined, but it would be advisable to also specify the detailed objectives. This would certainly strengthen the importance of the research topic. It may also be worth considering whether to expand the introduction with research hypotheses.

Response: We sincerely thank the reviewer for this important suggestion. We fully agree that specifying detailed research objectives and formulating clear hypotheses contribute to a more rigorous research framework. Consequently, we have added a dedicated sentence at the end of the Introduction (Section 1), preceding the statement of marginal contributions, to explicitly outline the three specific objectives of this study: (1) to empirically examine the direct impact of land misallocation on economic resilience; (2) to reveal its spatial spillover effects; and (3) to dissect the core mediating mechanism of "impeding technological progress." Regarding the research hypotheses, since we have formally proposed two hypotheses (Hypothesis 1 and Hypothesis 2) at the end of the theoretical model and analysis section (Section 2.2), we have simply added a summary statement in the Introduction (Section 1) to inform readers that specific hypotheses will be derived from the theoretical model for subsequent empirical testing. This enhancement strengthens the logical coherence of the study.

Comments 2. The authors assumed that the research area consists of 95 Chinese cities. I suggest that these cities be indicated graphically or described in more detail. I do not share the opinion that the fact that these cities account for 70% of agricultural production automatically makes them representative (line 451).

Response: We thank the reviewer for the important suggestion. In accordance with this advice, we have incorporated a discussion on the distribution and representativeness of the sample cities into Appendix A. Due to length considerations, this discussion has been placed in the appendix at the end of the manuscript.

Comments 3. The data sources used are described too generally: China Land and Resources Statistical Yearbook and the Wind databases. It would be advisable to at least present the general developmental trends of these variables. I consider that this part of the article should be supplemented.

Response: We thank the reviewer for the reminder. We have substantially supplemented the description regarding the source of land price data and the sample selection and setup. First, in the data source section, we have specified that the data are derived from the "Land Market Transactions" section of the China Land and Resources Statistical Yearbook. Second, in the variable specification section, we have emphasized that since 2012, the China Land and Resources Statistical Yearbook has distinguished between market-based conveyance prices and non-market allocation (e.g., administrative allocation) data. This study utilizes only the market-based conveyance prices to mitigate interference from non-market factors.

Comments 4. The Conclusion section needs to be expanded. At present, it contains only general recommendations for local and regional economic policy. The authors studied only part of the country – are there studies concerning other regions of China? What recommendations can be made in this regard?

Response: We thank the reviewer for the important reminder. In response, we have expanded Section 5 to include a discussion on the limitations and generalizability of our research findings. We believe this addition will allow a global readership to derive valuable insights from our conclusions that are relevant to their own contexts.

Comments 5. Literature – the review of previous studies by other authors should be supplemented, including a broader review of international literature, especially from regions outside China. I recommend adding an appendix that would, for example, provide characteristics of the variables used.

Response: We thank the reviewer for the important suggestions. In response, we have supplemented the Introduction with relevant international literature. Furthermore, following the recommendation, we have added two appendices: Appendix A discusses the distribution and representativeness of the variable sample, and Appendix B reports the results of the robustness checks using an alternative spatial matrix.

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors Peer-Review Report Journal: Land (MDPI)   Manuscript Title: How Does Land Misallocation Weaken Economic Resilience? Evidence from China   Manuscript ID: land-4097269   This study addresses a critical gap in regional economics by investigating how distortions in land factor allocation (specifically between industrial and commercial uses) impact regional economic resilience in China. While existing literature has extensively covered labour and capital misallocation, the role of land—a factor heavily regulated by the state in China—is often overlooked. The paper effectively combines a theoretical model with empirical evidence from a panel of 95 Chinese cities (2012–2024). The core finding “land misallocation reduces resilience by suppressing technological advancement” is significant. The identification of a "price scissors gap" (low-priced industrial land and high-priced commercial land) as a driver of inefficient firm survival provides a clear mechanism for this impact.   Theoretical Framework and Hypotheses The use of the Olley-Pakes model to weigh regional technological levels against firm market shares is a robust choice. It logically supports Hypothesis 1: that land misallocation impairs regional economic resilience. The authors introduce in section 2.2, the assumption that firm A uses land/labour while firm B uses capital/labour is a useful simplification but may over-abstract the reality where industrial firms (firm A) are often capital-intensive in modern China. This duality should be clarified according to relevant literature.   Methodology and Data The choice of the "Single-variable Shock Econometric Model" for measuring resilience (Equation 7) is well-justified and allows for better cross-sectional comparison than multi-indicator systems. The authors show technical rigor by employing System-GMM (SYS-GMM), the Han-Phillips GMM, and an Instrumental Variables (IV) approach. Using topographic slope as an IV is a natural choice, should be addressed with enough bibliographical evidence. The study uses 95 cities because they specifically disclose price data by land use. While these cities represent 70% of non-agricultural GDP, the authors should briefly discuss if the exclusion of smaller, more rural cities might bias the results toward more developed urban dynamics, and support with deeper evidence the adequacy of the sample size.   Empirical Results and Discussion The finding that land misallocation in one city might positively affect the resilience of similar cities (the "industrial crowding-out" effect) is a relevant finding. It suggests that inefficient land policies in one region can inadvertently benefit neighbours by pushing high-quality firms elsewhere. A more extended discussion and possible additional consequences should be included on the discussion.

Author Response

Comments 1. The use of the Olley-Pakes model to weigh regional technological levels against firm market shares is a robust choice. It logically supports Hypothesis 1: that land misallocation impairs regional economic resilience. The authors introduce in section 2.2, the assumption that firm A uses land/labour while firm B uses capital/labour is a useful simplification but may over-abstract the reality where industrial firms (firm A) are often capital-intensive in modern China. This duality should be clarified according to relevant literature.

Response: We thank the reviewer for the important reminder. In accordance with the suggestion, we have added a note at the end of Section 2.2, clarifying that the assumption represents a theoretical abstraction designed to clearly elucidate the underlying mechanism. We acknowledge the capital-intensive nature of industrial enterprises in reality. We emphasize that the core logic remains valid as long as land misallocation distorts the flow of resources toward more efficient sectors, a premise supported by the subsequent empirical findings.

Comments 2. The choice of the "Single-variable Shock Econometric Model" for measuring resilience (Equation 7) is well-justified and allows for better cross-sectional comparison than multi-indicator systems. The authors show technical rigor by employing System-GMM (SYS-GMM), the Han-Phillips GMM, and an Instrumental Variables (IV) approach. Using topographic slope as an IV is a natural choice, should be addressed with enough bibliographical evidence.  

Response: We thank the reviewer for the suggestion. As advised, we have supplemented the section on addressing endogeneity (Section 4.2.1) by adding references to the existing literature that supports the use of variables such as topographic slope as instrumental variables.

Comments 3. The study uses 95 cities because they specifically disclose price data by land use. While these cities represent 70% of non-agricultural GDP, the authors should briefly discuss if the exclusion of smaller, more rural cities might bias the results toward more developed urban dynamics, and support with deeper evidence the adequacy of the sample size.  

Response: We thank the reviewer for the comments regarding sample representativeness. We have provided a detailed discussion on the distribution and representativeness of the sample in Appendix A. It should be noted that this study focuses on price distortions in industrial and commercial/residential land. Given that China's rural land supply system operates independently and such distortions are typically not pronounced in smaller cities, our sample, while potentially skewed toward larger cities, effectively encompasses the vast majority of China's non-agricultural sectors and medium-to-large cities. Consequently, the sample remains representative for the objectives of this research.

Comments 4. The finding that land misallocation in one city might positively affect the resilience of similar cities (the "industrial crowding-out" effect) is a relevant finding. It suggests that inefficient land policies in one region can inadvertently benefit neighbours by pushing high-quality firms elsewhere. A more extended discussion and possible additional consequences should be included on the discussion.

Response: We thank the reviewer for the highly valuable suggestion. In response, we have expanded the analysis in Section 4.1 to include a more extensive discussion of the "industrial crowding-out effect." This addition explores the potential risks of regional policy competition and long-term resource misallocation that this effect may trigger, and proposes corresponding implications for policy coordination.

Round 2

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The authors have addressed the review comments. I have no more comment. 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study examines how land misallocation affects regional economic resilience in China, integrating a theoretical model with spatial econometric and mediation analyses using panel data from 95 cities. The topic aligns well with the scope of Land. However, the manuscript requires major revisions before it can be considered for publication.

1. The theoretical model is lengthy but not sufficiently integrated.Recommend to provide a concise conceptual diagram linking: land misallocation → factor allocation → technology → resilience.
2. RES values appear extremely large (e.g., -15.82 to 75.66). These magnitudes are unusual and raise concerns about the validity of the metric.
3. Using the ratio of commercial-to-industrial land prices is reasonable but problematic:Industrial land in many cities is not sold publicly—prices may be administratively set, producing biases.
4. The choice of a GDP-distance spatial weight matrix is unconventional.GDP-based weights may create artificial connections between distant cities with similar GDP (e.g., Urumqi and Xiamen).
5. Although the paper attempts GMM and spatial GMM,land misallocation is likely endogenous to resilience through fiscal stress and land policy adjustments. A more convincing identification strategy is needed.

Some minor comments:

1. Somesentences are too long and need simplification for readability.
2. Table 1: “Patient” should be corrected to “Patent”.

Author Response

Comments 1. The theoretical model is lengthy but not sufficiently integrated. Recommend to provide a concise conceptual diagram linking: land misallocation → factor allocation → technology → resilience.

Response: We appreciate the reviewer's suggestion. In response, we have streamlined the theoretical model section. To enhance clarity, we have added a succinct summary at the beginning of Section 2, which outlines the mechanism through which land misallocation impacts economic resilience. Subsequently, Section 2.1 formalizes the relationship between land misallocation and economic resilience using a mathematical model, while Section 2.2 employs mathematical modeling to analyze the intrinsic mechanisms of this impact. Although no additional framework diagram has been introduced, this concise introductory analysis (under 200 words) effectively presents the relationships and mechanisms among the key variables to readers.

Comments 2. Using the ratio of commercial-to-industrial land prices is reasonable but problematic: Industrial land in many cities is not sold publicly—prices may be administratively set, producing biases.

Response: We thank the reviewer for this insightful comment. Accordingly, we have added a detailed explanation regarding this variable choice at the end of Section 3.1.1. Since China's land statistics have separately reported market-based and non-market-based land transactions since 2012, our study exclusively utilizes data from market-based land sales for the empirical analysis. This approach systematically excludes samples with administratively set prices, thereby mitigating potential bias.

Comments 3. RES values appear extremely large (e.g., -15.82 to 75.66). These magnitudes are unusual and raise concerns about the validity of the metric.

Response: We appreciate the reviewer's observation. The sample period encompasses the COVID-19 pandemic, during which economic data for some cities exhibited significant anomalies. All outlier data points have been carefully verified and cross-checked by the authors. Crucially, the removal of these outliers does not alter the core conclusions of our empirical tests. Furthermore, in the revised manuscript, we have applied winsorization to all variable samples in the re-conducted empirical tests to further minimize the influence of extreme values.

Comments 4. The choice of a GDP-distance spatial weight matrix is unconventional. GDP-based weights may create artificial connections between distant cities with similar GDP (e.g., Urumqi and Xiamen).

Response: We fully agree with the reviewer's suggestion and have implemented a significant revision accordingly. While production factors often flow between cities of comparable economic scale, geographical proximity remains a crucial consideration even among similarly sized cities. Therefore, a spatial economic weight matrix should incorporate both economic scale and geographical distance. Following the reviewer's advice and drawing upon established literature, we have modified the weight matrix. The new matrix uses the product of the geographical distance and the economic disparity (in terms of GDP) between two cities as its key variable. All spatial econometric models have been re-estimated using this updated spatial economic weight matrix.

Comments 5. Although the paper attempts GMM and spatial GMM, land misallocation is likely endogenous to resilience through fiscal stress and land policy adjustments. A more convincing identification strategy is needed.

Response: We thank the reviewer for highlighting this critical point. To better address endogeneity concerns in the econometric models, we have supplemented the original System GMM approach with an Instrumental Variables (IV) strategy. Specifically, in Section 4.2.1: (1) We employ the lagged dependent variable as an instrument to construct a System GMM model, which is further integrated with the spatial econometric framework to form a Han-Phillips GMM model, thereby alleviating endogeneity bias. (2) Drawing on established methods in the literature, we use city terrain slope and plot ratio as instrumental variables and apply a Two-Stage Least Squares (2SLS) approach for robustness checks. These instruments are strongly correlated with land misallocation and satisfy the exclusivity condition, a claim supported by passing relevant instrument validity tests. These results collectively reinforce the robustness of our empirical findings.

Comments 6. Some sentences are too long and need simplification for readability. Table 1: “Patient” should be corrected to “Patent”.

Response: We are grateful to the reviewer for noting these details. We have substantially condensed the text throughout the manuscript to improve readability and have corrected the typographical error ("Patient" to "Patent") in Table 1.

Reviewer 2 Report

Comments and Suggestions for Authors

This paper examines how variations in land allocation and land-use policy influence broader economic performance.
The paper claims to integrate land misallocation and economic resilience into a unified analytical framework, but the theoretical arguments are mostly extensions of existing misallocation and TFP literature with limited innovation. The theoretical model is overly mechanistic, lacks clarity, and provides no genuinely new insights beyond prior work.  The empirical design does not convincingly establish causal effects. Endogeneity between resilience and land policy remains unresolved despite superficial GMM attempts. The chosen instruments are inappropriate and insufficiently justified, making the causal claims unreliable. The operationalization of the core constructs (land misallocation; economic resilience), especially resilience, relies on highly simplified assumptions and non-standard transformations. These decisions lack justification and may generate artificially strong results.

The authors use multiple spatial models (SEM, SARM, SDM, Dynamic SDM) without clear justification for model choices or interpretation of cross-model inconsistencies. The spatial weight matrix based on GDP differences is poorly motivated and may bias results. The mediation analysis fails patents do not significantly mediate the relationship, yet the authors continue to assert the mechanism as if it were validated. This is a major flaw: the mechanism is central to the paper’s contribution, yet the evidence contradicts it.

Several datasets (e.g., patent counts, land price ratios) are not fully described, and key details such as city selection criteria, missing values, or data cleaning procedures are absent. Reproducibility is hindered by incomplete methodological transparency. The manuscript is overly long, repetitive, and poorly structured. Many sections include redundant explanations of basic concepts, while important empirical interpretations are brief and unclear. Figures and tables are not adequately integrated into the narrative.

Author Response

Comments 1. The paper claims to integrate land misallocation and economic resilience into a unified analytical framework, but the theoretical arguments are mostly extensions of existing misallocation and TFP literature with limited innovation. The theoretical model is overly mechanistic, lacks clarity, and provides no genuinely new insights beyond prior work.

Response: We thank the reviewer for this comment. Our study indeed builds upon the existing literature concerning the relationship between factor misallocation and TFP. However, prior research has predominantly focused on labor and capital factors, with limited attention given to the misallocation of land factor. Furthermore, while existing studies mainly examine the impact of misallocation on TFP, the more recent concept of economic resilience remains underexplored in this context. Therefore, while our analytical framework draws from established models, the substitution and refinement of the core research variables represent an innovative shift in perspective. We believe the findings offer practical guidance for optimizing land allocation to mitigate economic shocks. To address the reviewer's concern regarding clarity, we have added a concise, synthesizing statement at the outset of Chapter 2 and have streamlined the theoretical model section by removing redundant text, making it more succinct and systematic.

Comments 2. The empirical design does not convincingly establish causal effects. Endogeneity between resilience and land policy remains unresolved despite superficial GMM attempts. The chosen instruments are inappropriate and insufficiently justified, making the causal claims unreliable.

Response: We acknowledge the reviewer's concern. To more rigorously address endogeneity in the econometric models, we have augmented the original System GMM approach with an Instrumental Variables (IV) strategy. As detailed in Section 4.2.1: (1) We utilize the lagged dependent variable as an instrument to construct a System GMM model, integrated within a spatial econometric framework as a Han-Phillips GMM model to mitigate endogeneity bias. (2) Following established methodologies in the literature, we employ city terrain slope and plot ratio as instrumental variables, applying a Two-Stage Least Squares (2SLS) estimator for robustness testing. These instruments demonstrate a strong correlation with land misallocation and meet the exclusivity condition, which is confirmed by standard instrument validity tests. These additional analyses substantiate the robustness of our empirical findings.

Comments 3. The operationalization of the core constructs (land misallocation; economic resilience), especially resilience, relies on highly simplified assumptions and non-standard transformations. These decisions lack justification and may generate artificially strong results.

Response: We appreciate the reviewer's observation. In response, we have provided a more detailed description and deeper analysis of the operationalization of these two core concepts. Recognizing that the modeling of economic resilience might be influenced by regional industrial structure, we have incorporated an additional robustness check in Section 4.2.2. Here, we refine the resilience measure to focus on industrial resilience and re-conduct the empirical tests, which continue to support our main conclusions.

Comments 4. The authors use multiple spatial models (SEM, SAR, SDM, Dynamic SDM) without clear justification for model choices or interpretation of cross-model inconsistencies.

Response: We thank the reviewer for this suggestion. Following this advice, we have added a detailed explanation in Section 3.2.1 regarding the reasons for potential discrepancies among the results of different spatial econometric models. Furthermore, in Section 4.1 (the benchmark analysis), we have elaborated on the process and criteria used for selecting the most appropriate spatial model.

Comments 5. The spatial weight matrix based on GDP differences is poorly motivated and may bias results.

Response: We concur with the reviewer's comments and have accordingly refined the spatial weight matrix. The flow of production factors between cities is influenced by both economic scale similarity and geographical proximity. Therefore, an optimal spatial economic weight matrix should account for both dimensions. Heeding the reviewer's suggestion and referencing extensive literature, we have revised the matrix. The new matrix is constructed using the product of the geographical distance and the economic disparity (in terms of GDP) between city pairs. All spatial econometric models have been re-estimated using this updated matrix.

Comments 6. The mediation analysis fails patents do not significantly mediate the relationship, yet the authors continue to assert the mechanism as if it were validated. This is a major flaw: the mechanism is central to the paper’s contribution, yet the evidence contradicts it.

Response: We thank the reviewer for highlighting this point. To clarify our analysis, we have added a summarizing statement at the beginning of Section 4.3: (1) The initial mediation test using total patent count (Patent) as the mediator yielded insignificant results. This lack of significance, however, does not invalidate the theoretical mechanism but may stem from measurement issues with the mediator variable. (2) Subsequent tests disaggregating patents into invention patents (Invention) and utility model/design patents (Design) revealed significant mediation effects for invention patents. We argue that Invention patents are a more precise proxy for regional technological capability than the aggregate Patent count. Therefore, the significant results from this refined test provide evidence supporting the proposed theoretical mechanism.

Comments 7. Several datasets (e.g., patent counts, land price ratios) are not fully described, and key details such as city selection criteria, missing values, or data cleaning procedures are absent. Reproducibility is hindered by incomplete methodological transparency.

Response: We appreciate the reviewer's comment. In accordance with this suggestion, we have added a comprehensive description of the sample selection and data processing procedures prior to the empirical analysis in Section 4.1. This includes detailed explanations of city selection criteria, handling of missing values, and the rationale behind the chosen sample's representativeness.

Comments 8. The manuscript is overly long, repetitive, and poorly structured. Many sections include redundant explanations of basic concepts, while important empirical interpretations are brief and unclear. Figures and tables are not adequately integrated into the narrative.

Response: We agree with the reviewer's assessment and have undertaken significant revisions to address these issues. We have condensed and restructured the overall content. Redundant explanations of basic concepts and less critical passages have been removed from Chapters 2 and 3 (theoretical model and research design). Conversely, we have expanded the interpretation of key empirical results in Chapter 4 and bolstered the discussion of robustness checks. Additionally, we have more effectively integrated figures and tables into the main narrative.