Perception over Possession: How Farmers’ Subjective Tenure Security and Forest Certification Drive Sustainable Bamboo Management

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

My understanding of this manuscript is that it examines the formalisation of tenure in the forest sector of Fujian Province. While the manuscript has a forest focus, I have done my review from the perspective of tenure. The theme of the paper is of both Chinese and international relevance. The structure and flow of the manuscript are understandable. Below are detailed comments based on my impressions of reading the manuscript.

Title: Title looks good. However, it is somewhat tautological, as it repeats “perception” within it. Assuming the authors want to improve it, they could consider a title like “Perception over possession: How farmers’ subjective tenure security and forest certification drive sustainable bamboo management” to make it more succinct. This is a suggestion, and I am not insisting on it.

Abstract: The abstract represents the content of the manuscript. It reads as a standalone part of the manuscript and still captures the key parts of the study presented in the manuscript.

Section 1: The introduction, in my opinion, is well written and presented. It states the central question addressed and how.

Section 2: The caption “Literature Review and Theoretical Analysis” does not make this section read like a pre-methodological section. It gives the impression of an analytical section because of the “analysis” in the caption. Perhaps, the author can recaption that section simply as “Literature and theoretical review” or simply as “Literature review.” This way, readers are logically guided without having to be distracted by the “analysis” that makes it read like a post-methodological section.

Sections 3 & 4: This section is well-presented. The model is presented in detail, and its limitations are acknowledged. The variables and indicators are explained in detail and defined for clarity and comprehension. The results section is also well presented.

Section 5: This discussion section focused on “Perception over Possession: The Critical Pathway for Institutional Instrument Effectiveness” and “Institutional Types and Behavioural Responses: Divergence in Long-Term vs. Short-Term Strategies.” What about the levels of subjectivity in the perceptions? I did not see that coming out within the discussion. This is crucial as it is central to the topic. Readers would want to see it discussed more directly. I suggest that the authors move section 5.3 (Limitations and Future Research) to the conclusion.

Conclusion: This section is too lean and should be expanded to show the policy implications of the results in China and elsewhere. Like, I stated above, the authors could move section 5.3 (Limitations and Future Research) to the conclusion.

Comments on the Quality of English Language

The quality of English is understandable and comprehensible. I have only indicated "The English could be improved to more clearly express the research" to motivate the authors to improve it more during revision.

Author Response

We gratefully thank the Reviewer for your review of our paper. All your suggestions are very important to us. The detail responses to all the issues raised by the Reviewer are listed in the following, and all of your suggestions have been addressed in the revised manuscript (highlighted).

 

Response to the Reviewer#1 

1) Title: Title looks good. However, it is somewhat tautological, as it repeats “perception” within it. Assuming the authors want to improve it, they could consider a title like “Perception over possession: How farmers’ subjective tenure security and forest certification drive sustainable bamboo management” to make it more succinct. This is a suggestion, and I am not insisting on it.

R:

Thank you for your suggestion. We have revised the title of the manuscript as you suggested.

 

2) Section 2: The caption “Literature Review and Theoretical Analysis” does not make this section read like a pre-methodological section. It gives the impression of an analytical section because of the “analysis” in the caption. Perhaps, the author can recaption that section simply as “Literature and theoretical review” or simply as “Literature review.” This way, readers are logically guided without having to be distracted by the “analysis” that makes it read like a post-methodological section.

R:

Thanks for your valuable comments. We have removed the word “Analysis” from the heading of Section 2.

 

3) Section 5: This discussion section focused on “Perception over Possession: The Critical Pathway for Institutional Instrument Effectiveness” and “Institutional Types and Behavioural Responses: Divergence in Long-Term vs. Short-Term Strategies.” What about the levels of subjectivity in the perceptions? I did not see that coming out within the discussion. This is crucial as it is central to the topic. Readers would want to see it discussed more directly. I suggest that the authors move section 5.3 (Limitations and Future Research) to the conclusion.

R:

Thank you for this very insightful comment. We agree that the subjective level of perception is a crucial element, and we appreciate the opportunity to clarify the focus of our study.

 

You have correctly pointed out that the absolute level of perception (i.e., whether it is high or low) was not a central topic in our Discussion section. This was a deliberate decision based on the scope and primary goal of our research. The specific perception levels observed in our study are characteristic of our sample in Fujian, China, and we acknowledge that they may not be generalizable to other regions worldwide. For this reason, we presented this descriptive information in Table 1 to provide context for our sample, rather than treating it as a primary finding for discussion.

 

The main purpose of our study is to reveal the underlying relationship between the variation in perception levels (high vs. low) and the corresponding changes in management behaviors. We believe this pattern—how different levels of perception influence strategic choices—is more likely to hold true across different contexts and thus offers greater universal value for readers.

 

Therefore, our discussion in Section 5 was intentionally focused on “perception of possession as a key pathway” and “the difference between long-term and short-term strategies,” as these represent the core theoretical contributions of our work.

 

To better address your concern and make our rationale clearer, we have now added a brief clarification at the beginning of Section 5. We explicitly state that while the absolute perception levels are sample-specific (as presented in Table 1), our core analysis concentrates on the behavioral implications of the variation in these perceptions. We hope this revision adequately addresses your point and strengthens our manuscript.

 

4) Conclusion: This section is too lean and should be expanded to show the policy implications of the results in China and elsewhere. Like, I stated above, the authors could move section 5.3 (Limitations and Future Research) to the conclusion.

R: 

Thank you for this excellent suggestion. We completely agree that the original conclusion was too concise and that expanding on the policy implications of our findings is crucial for highlighting the study's contribution.

 

Regarding your suggestion to move Section 5.3 (Limitations and Future Research) into the conclusion, we have given it careful consideration. We believe that keeping these two sections distinct offers a clearer and more logical structure for the reader. Our intention is for the Conclusion section to provide a strong, focused summary of our study's key findings and their direct implications—essentially, the “take-home messages.” The Limitations and Future Research section, in contrast, serves a different purpose: to offer a critical self-assessment and a forward-looking roadmap for future inquiry. We feel that merging these two might dilute the impact of our main conclusions by immediately introducing caveats and future work.

 

Therefore, while maintaining the separate structure, we have followed your primary advice and substantially expanded the Conclusion section. We have added a new, detailed paragraph discussing the specific policy implications of our findings, both for the ongoing institutional reforms in China and for similar contexts internationally. This new paragraph elaborates on how policymakers can leverage subjective perception to improve the efficacy of tenure and certification programs.

 

We hope this approach successfully addresses your concern and that the expanded conclusion now better showcases the value of our research.

Reviewer 2 Report

Comments and Suggestions for Authors

See attached document.

Comments for author File: Comments.pdf

Author Response

We gratefully thank the Reviewer for your review of our paper. All your suggestions are very important to us. The detail responses to all the issues raised by the Reviewer are listed in the following, and all of your suggestions have been addressed in the revised manuscript (highlighted).

Response to the Reviewer#2 

Introduction

 

1) L73-74 How have these models been used in the past to evaluate peoples perceptions?

R:

Thank you for this excellent question, which allows us to clarify the precise role of these models in our analysis and their grounding in the existing literature.

 

To be precise, the Double-Hurdle and Tobit models are not used to "evaluate perceptions" themselves. Rather, they are used to evaluate the impact of farmers' perception variables (as key independent variables) on their specific economic behaviors (the dependent variables).

 

We chose these models because our behavioral outcome variables (e.g., investment amount) are characterized by a large number of zero observations. The Double-Hurdle model, in particular, is well-suited as it allows us to analyze the two stages of a household's decision process separately: (1) the decision of whether to engage in an activity, and (2) how much to invest, conditional on engagement.

 

This methodological approach is a well-established and current practice in household-level micro-econometric analysis. For instance, the Double-Hurdle model has been effectively used to analyze household expenditure decisions in development contexts (e.g., Adusah-Poku & Takeuchi, 2019). Similarly, the Tobit model and its variants are standard tools for analyzing censored outcomes like household financial assets (e.g., Wang et al., 2022). More broadly, these types of models are frequently used in studies that, like ours, examine how various factors influence household participation and investment decisions (e.g., Du et al., 2024).

 

We hope this clarification and these examples from recent, relevant literature sufficiently demonstrate that our choice of methodology is robust and well-grounded in standard practice for this type of analysis. Thank you again for the opportunity to elaborate on this point.

 

1. Adusah-Poku, F., & Takeuchi, K. (2019). Household energy expenditure in Ghana: A double-hurdle model approach. World Development, 117, 266-277. https://doi.org/10.1016/j.worlddev.2019.01.018

 

2. Du, C., Ouyang, Y., Yang, Q., & Shi, Z. (2024). The impact of digital economy on household private insurance participation. International Review of Economics & Finance, 95, 103456. https://doi.org/10.1016/j.iref.2024.103456

 

3. Wang, X., Zhou, X., Li, B., Zhang, F., & Zhou, X. (2022). A bent line Tobit regression model with application to household financial assets. Journal of Statistical Planning and Inference, 221, 69-80. https://doi.org/10.1016/j.jspi.2022.02.008

 

2) L87- Not sure about this section, according to journal guidelines it should be an intro followed by the methods. The literature review should be incorporated into the introduction the hypotheses should also show up at the end of the introduction.

R:

Thank you for your thoughtful comment regarding the manuscript's structure and its alignment with the journal's guidelines. We have carefully considered your suggestion to incorporate the literature review and hypothesis development into the Introduction section.

 

We opted for the current structure, with a distinct 'Literature Review and Hypothesis Development' section (Section 2) following the Introduction, because we felt it provided the necessary space to thoroughly develop the theoretical framework and logically derive our hypotheses before presenting the methodology. This approach allows for a clearer and more systematic progression of our argument.

 

To ensure our structure is appropriate, we also reviewed recent articles published in Land. While we understand the "Introduction -> Methods" format is common, we found that the journal allows for flexibility to accommodate different types of empirical research that require a detailed theoretical foundation. For instance, we noted several recent articles that employ a similar structure, with a dedicated theoretical or literature review section between the Introduction and Methods.

 

Two prominent recent examples from Land include:

 

Liang, S. et al. (2025), "How Land Inflow Affects Rural Household Development Resilience—Empirical Evidence from Eight Western Counties in China," Land 14(6), 1251. (DOI: 10.3390/land14061251)

 

Han, X. et al. (2025), "Influencing Factors and Transmission Mechanisms of Pro-Environmental Behavior: Evidence from Tea Farmers in Wuyishan National Park," Land 14(7), 1367. (DOI: 10.3390/land14071367)

 

Both articles feature a dedicated theoretical analysis or literature review section after the introduction and before the methods.

 

Based on these precedents from the journal itself, we believe our current manuscript structure is consistent with Land's accepted practices and effectively serves the clarity of our argument. We hope this explanation clarifies our rationale.

 

3) L141: The hypotheses should be expanded a bit to tell us why the authors believe that they will be true e.g., H1: Stronger perceived tenure security positively correlates with higher household production input intensity because stakeholders confidence in policies has increased household production in other studies. These statements should be specific and follow directly from your literature review.

R:

Thank you for this excellent point. We completely agree that each hypothesis should be explicitly justified and follow directly from the preceding literature review and theoretical discussion, rather than being presented in a vacuum.

 

In our revised manuscript, we have structured the preceding sections (Sections 2.1 and 2.2) to systematically build the theoretical and empirical foundation for the hypotheses that are then formally presented in Section 2.3. We would like to briefly guide you to the specific justifications for each hypothesis, which are now embedded in the text leading up to them:

 

For Hypotheses H1 and H2 (related to Perceived Tenure Security): The rationale is developed throughout Section 2.1. This section first reviews literature establishing that stable tenure expectations increase long-term investment and promote sustainable harvesting patterns (e.g., extending harvesting cycles). It then highlights that this effect hinges on security being subjectively internalized by farmers, thus directly building the case for why perceived security is the key driver for H1 and H2.

 

For Hypotheses H3 and H4 (related to Perceived Certification Benefits): The justification is provided in Section 2.2. We first discuss literature showing that expected market benefits from certification motivate higher inputs (the basis for H3). Crucially, in response to feedback from another reviewer, we added a detailed paragraph at the end of Section 2.2 that explains the specific behavioral logic for H4: why a strong perception of benefits leads to a more sustainable harvesting pattern (more frequent, lower-volume selective cutting) driven by a desire for a stable income stream and an internalization of the certification's principles.

 

We believe that this structure, where the literature review in Sections 2.1 and 2.2 directly builds the case for the hypotheses presented in Section 2.3, now provides the clear, evidence-based justification that you rightly called for. Thank you for prompting us to confirm and clarify this logical linkage.

 

Methods

 

4) L263 Generally, the research area or site description comes at the beginning of the methods section. I would suggest restructuring the paper so that section 3.5 is moved to section 3.1. It is helpful to know about the stie before explaining most of the methods.

R: 

Thank you for this excellent and very logical suggestion regarding the structure of our Methods section. As you recommended, we have now restructured the paper by moving the “Research Area and Survey Methodology” section from its original position to become the new Section 3.1.

 

5) L263 - Also included in this section should a more detailed description of the people who were surveyed. How much of their livelihoods are based on bamboo harvesting? Also give us more of a sense of what the landscape mosaic looks like. What other land use types exist in this landscape?

R: 

Thank you for this excellent suggestion. We agree that providing a richer description of the surveyed households and the landscape context significantly enhances the reader’s understanding of our study.

 

Following your advice, we have added a new introductory passage to Section 3.4 (Descriptive Statistics) to provide this context before presenting the detailed statistics for our model’s variables.

 

This new passage now provides:

 

A more detailed profile of the surveyed households, drawing directly from Table 1. We now describe key demographic characteristics, including the average age, gender, and education level of the household heads, as well as household labor force size and participation in cooperatives.

 

A specific answer to your question about livelihoods: we now explicitly state that forestry income constitutes, on average, 15.14% of total household income. We also highlight the high standard deviation to note the significant variation in dependency among households.

 

A description of the landscape mosaic, explaining that it is dominated by bamboo forests but is also interspersed with other key land uses such as timber forests, tea plantations, and agricultural plots.

 

We believe this addition successfully paints the clearer and more detailed picture of our study's context that you recommended. Thank you for helping us improve the richness of our description.

 

• L281 The methods section is also missing a Data Analysis section where the specific models that you ran and why you ran those models are described.

R: 

Thank you for this excellent structural suggestion. We agree that having a clearly demarcated “Data Analysis” section significantly improves the readability and logical flow of the methodology.

 

In our previous version, the descriptions of our econometric models and the rationale for their use were detailed in separate sections (formerly 3.2 and 3.3). However, we see your point that grouping them under a single, explicit “Data Analysis” heading is a much clearer and more standard approach.

 

Following your advice, we have restructured our Methods section. We have created a new main section titled “3.2 Data Analysis Strategy”. The detailed descriptions of the Double-Hurdle model and the Tobit model, including their mathematical specifications and the justification for their use, are now presented as subsections 3.2.1 and 3.2.2 under this new heading. The subsequent sections have been renumbered accordingly.

 

We are confident that this new structure now provides the clear, dedicated Data Analysis section you rightly suggested and makes our analytical approach much easier for the reader to follow. Thank you for helping us improve the organization of our paper.

 

Results

 

7) L283-L287 moved to the methods section (in the data analysis section which still needs to be added).

R: 

Thank you for this excellent structural suggestion. We agree completely that the description of the multicollinearity test belongs in the Methods section as part of our data analysis strategy.

 

As you recommended, we have moved this entire section (formerly 4.1) from the Results to the Methods section. It is now presented as a new subsection, 3.2.3 Multicollinearity Test, under our main “Data Analysis Strategy” heading.

 

We believe this change improves the logical flow of the paper. Thank you for pointing this out.

 

8) L323 This robustness test section is also really not related to your actual results but is more something that you might include in the data analysis section of your results.

L329 - lf you aren't going to report the results of these models in the body of the ms than you need toinclude a supplementary file with those model results included.

R: 

Thank you for this excellent structural suggestion. We completely agree that the robustness test is part of our analytical strategy and that its description is better placed within the Methods section, with the detailed results available but not interrupting the flow of the main findings.

 

Following your advice, we have made two key changes:

 

We have moved the description of our robustness test strategy into a new subsection within our Methods, Section 3.2.4 ‘Robustness Test’.

 

The complete results tables for these robustness tests (formerly Table 6) have been moved to the Appendix.

 

We believe this new structure—explaining the method upfront in the “Data Analysis” section and presenting the detailed results in an appendix—significantly improves the logical flow and readability of the paper. Thank you for helping us to better organize our manuscript.

 

9) Throughout this section it would be good to include a bit more about each model as well including degrees of freedom and test statistics. Also, large tables are a bit hard to read. Would it be possible foryou to summarize these findings in a few well thought out figures/graphs?

R: 

Thank you for the excellent suggestion to enhance the statistical detail in our results tables. We agree that providing the complete test statistics, including degrees of freedom, is important for methodological rigor and transparency.

 

Following your advice, we have now updated our main regression tables (Tables 2-5) to include the degrees of freedom for our overall model fit statistics. The results for these tests are now reported in the standard, more complete format of Wald chi2(18) and LR chi2(18), as appropriate for each model.

 

We believe this change makes our statistical reporting more complete and professional. Thank you for helping us to improve the quality of our manuscript.

 

Discussion

10) In this section, you need to do a much greater contextualization of how your work fits into past work. You need to draw on more citations and further consider how your work fits into the larger body of work. You need to return to each hypothesis you have and specifically address how your findings relate to other work for each hypothesis.

R: 

Thank you for your very clear and constructive feedback on how to strengthen the Discussion section. We completely agree that a systematic discussion that explicitly connects our findings back to each hypothesis and the broader literature significantly enhances the paper's contribution.

 

Following your detailed guidance, we have substantially revised the entire Discussion section (Sections 5.1 and 5.2) to improve its structure, contextualization, and argumentation.

 

Specifically, the revised section now does the following:

 

Explicitly Addresses Each Hypothesis: The discussion is now structured around our key findings as they relate to each of our four hypotheses (H1, H2, H3, and H4). We directly state whether each hypothesis was supported by our empirical results.

 

Provides Deeper Contextualization: For each finding, we now provide a more detailed discussion of how it fits into the larger body of work. We have integrated additional references to past studies to show how our work confirms, extends, or adds nuance to the existing literature.

 

Ensures All Statements are Supported: We have carefully reviewed the entire section to ensure that every statement is now either explicitly supported by our own empirical results or by a citation to previous research, as you rightly suggested.

 

We are confident that this thoroughly revised Discussion section now provides the clear, well-supported, and structured analysis that you called for. Thank you again for your insightful guidance, which has been instrumental in helping us to significantly sharpen our paper's core arguments and contributions.

 

Reviewer 3 Report

Comments and Suggestions for Authors

This study focuses on the topic of evaluating the effectiveness of land and forestry policies, with a particular emphasis on whether the "subjective perception" of farmers is more important than the "objective possession" of policy tools. Against the backdrop of China's collective forest tenure reform and forest rights certification, it systematically deconstructs the impact paths of clear property rights (forest tenure certificates) and market-oriented incentives (bamboo forest certification) on farmers' production behaviors. This is a topic of significant academic value and practical significance. Overall, this is a high-quality paper with publication potential, but there are still some areas that can be improved:

1. For the literature review and theoretical analysis section, first of all, the logic of H4 hypothesis needs further elaboration: H4 proposes that a "stronger certification perception" will lead to "shorter harvesting intervals" but "reduced harvesting volume per extraction". This hypothesis is interesting and aligns with the logic of sustainable management (more frequent and less intense "selective cutting"). However, in the theoretical analysis section, the author did not elaborate on this logic in detail. Why would stronger market confidence lead to this specific shift in the harvesting pattern? Is it because the certification standards themselves require this, or because farmers expect to obtain continuous and stable income through this approach? It is suggested to add a short paragraph in Section 2.3 to provide a more detailed explanation of the behavioral logic behind H4 to enhance its theoretical foundation. Regarding the theoretical positioning of "perception outweighs possession", although the author mentioned this logic in Section 2.3, it is suggested to further briefly relate it to broader social science theories (such as "mental accounting" in behavioral economics, "informal institutions" in institutional economics, etc.). It is believed that this will enhance the theoretical height and universality of this article.

2. For the section on methods and data sources, this research mentioned the "perceived benefit index constructed via Likert-scale". To enhance reproducibility, it is suggested that a brief explanation of how this index was constructed be provided in the appendix. For instance, was it obtained by averaging several Likert-scale questions, or through factor analysis or principal component analysis for dimension reduction? Additionally, the dependent variable "Total inputs per mu" was quantified as "integrated land, labor, and capital inputs". It is recommended that a brief explanation be given in the main text or in a footnote on how these three types of inputs were monetized and integrated. For example, how was labor input converted into monetary value (was it based on local wage rates)?

3. Regarding the research results section, first of all, the presentation of the tables can be optimized to a certain extent to highlight the core comparison of this study - "possession" vs "perceived". For instance, the regression results of "investment intensity" in Table 2 (property rights) and Table 3 (certification) can be combined into one table. Similarly, Table 4 and Table 5 can be merged. This way, the coefficient sizes and significance of the four core explanatory variables (holding property rights certificate, perceived property rights security, participation in certification, and perceived certification benefits) can be compared intuitively in the same table, thereby more directly verifying the conclusion that "perception is more important than possession". Additionally, although the core explanatory variables are the focus, the significant results of some control variables are also worth briefly mentioning, which can increase the richness of the research. For example, Table 2 shows that "risk preference" and "perceived environmental value" have significant impacts on investment behavior. Briefly mentioning these findings in the result analysis section can make the analysis more comprehensive.

4. As for the discussion part, the author put forward the viewpoint that policy design should focus on "perception construction", but it could go further by presenting some specific and operational policy suggestions. For instance, how should the government and certification institutions build farmers' positive perception through "strategic information dissemination, cognitive guidance and value shaping"? Should it be through more training, setting up demonstration households, leveraging village social networks, or through media publicity? Offering more specific suggestions will greatly enhance the policy reference value of this article. Additionally, this study found that "holding a forest tenure certificate" and "participating in certification" themselves were not significant. The discussion section attributed this to the fact that "perception" is more important. This is indeed the core conclusion, but it could also explore why "holding" itself fails. Regarding the forest tenure certificate, is it because local government enforcement is insufficient, or due to historical policy fluctuations, causing farmers to lack trust in this "piece of paper"? As for certification, is it because the market premium brought by certification is not obvious in reality, or because middlemen have seized most of the profits, leading to farmers' poor actual participation experience?

Author Response

We gratefully thank the Reviewer for your review of our paper. All your suggestions are very important to us. The detail responses to all the issues raised by the Reviewer are listed in the following, and all of your suggestions have been addressed in the revised manuscript (highlighted).

Response to the Reviewer#3 

1) For the literature review and theoretical analysis section, first of all, the logic of H4 hypothesis needs further elaboration: H4 proposes that a "stronger certification perception" will lead to "shorter harvesting intervals" but "reduced harvesting volume per extraction". This hypothesis is interesting and aligns with the logic of sustainable management (more frequent and less intense "selective cutting"). However, in the theoretical analysis section, the author did not elaborate on this logic in detail. Why would stronger market confidence lead to this specific shift in the harvesting pattern? Is it because the certification standards themselves require this, or because farmers expect to obtain continuous and stable income through this approach? It is suggested to add a short paragraph in Section 2.3 to provide a more detailed explanation of the behavioral logic behind H4 to enhance its theoretical foundation. Regarding the theoretical positioning of "perception outweighs possession", although the author mentioned this logic in Section 2.3, it is suggested to further briefly relate it to broader social science theories (such as "mental accounting" in behavioral economics, "informal institutions" in institutional economics, etc.). It is believed that this will enhance the theoretical height and universality of this article.

R:

Thank you for these two excellent and highly constructive suggestions. We completely agree that both the behavioral logic behind H4 and the theoretical foundation of our "perception-over-possession" framework required further elaboration to strengthen the manuscript. We have revised the paper accordingly.

 

On the Behavioral Logic of Hypothesis H4: We agree that the connection between stronger certification perception and the specific harvesting pattern (shorter intervals, reduced volume) needed a more detailed explanation. Following your advice, we have added a new paragraph at the end of Section 2.2 to elaborate on this. The new text explains that this sustainable harvesting strategy is driven by two key factors: (1) the economic rationale of shifting from one-time profit maximization to securing a continuous and stable income stream, and (2) the internalization of sustainable management principles and selective cutting requirements often mandated by the certification standards themselves. We believe this addition provides a much stronger theoretical foundation for H4.

 

On the Theoretical Positioning of “Perception Outweighs Possession”: Thank you for the insightful recommendation to connect our core concept to broader social science theories. As you suggested, we have revised the first paragraph of Section 2.3 to explicitly link our “perception-over-possession” logic to foundational theories. We now connect our framework to the concept of informal institutions from institutional economics (North, 1990), where subjective beliefs and norms can be more influential than formal rules. Additionally, we relate it to mental accounting from behavioral economics (Thaler, 1999), where the subjective value an individual assigns to an asset dictates their behavior more than its objective form. We are confident that these additions better situate our study within a wider theoretical context and enhance its universality.

 

2) For the section on methods and data sources, this research mentioned the "perceived benefit index constructed via Likert-scale". To enhance reproducibility, it is suggested that a brief explanation of how this index was constructed be provided in the appendix. For instance, was it obtained by averaging several Likert-scale questions, or through factor analysis or principal component analysis for dimension reduction? Additionally, the dependent variable "Total inputs per mu" was quantified as "integrated land, labor, and capital inputs". It is recommended that a brief explanation be given in the main text or in a footnote on how these three types of inputs were monetized and integrated. For example, how was labor input converted into monetary value (was it based on local wage rates)?

R:

Thank you for your very valuable comments aimed at improving the rigor and reproducibility of our methods section. We completely agree with your points and have made additions to the manuscript based on your excellent suggestions.

 

Regarding the construction of the “perceived benefit index”: Thank you for highlighting the need for this clarification. We agree that explaining the index construction is crucial for reproducibility. Following your advice, we have added a new, detailed explanation of this process in Section 3.3. This new text specifies that the index was obtained by averaging the scores of several Likert-scale questions. While you suggested an appendix, we felt this methodological detail was central enough to include directly in the main text for immediate reader access, and we thank you for prompting us to add it.

 

Regarding the monetization and integration of inputs for “Total inputs per mu”: This was another excellent suggestion. As you recommended, we have added a new explanation in the main text. This new text clarifies how land, labor, and capital inputs were integrated into a single monetary measure, specifying that labor input was converted into a monetary value based on the prevailing local wage rates for hired labor.

 

3) Regarding the research results section, first of all, the presentation of the tables can be optimized to a certain extent to highlight the core comparison of this study - "possession" vs "perceived". For instance, the regression results of "investment intensity" in Table 2 (property rights) and Table 3 (certification) can be combined into one table. Similarly, Table 4 and Table 5 can be merged. This way, the coefficient sizes and significance of the four core explanatory variables (holding property rights certificate, perceived property rights security, participation in certification, and perceived certification benefits) can be compared intuitively in the same table, thereby more directly verifying the conclusion that "perception is more important than possession". Additionally, although the core explanatory variables are the focus, the significant results of some control variables are also worth briefly mentioning, which can increase the richness of the research. For example, Table 2 shows that "risk preference" and "perceived environmental value" have significant impacts on investment behavior. Briefly mentioning these findings in the result analysis section can make the analysis more comprehensive.

R:

Thank you for these two very thoughtful suggestions on how to optimize the presentation of our results. We have carefully considered your advice on merging the tables and discussing the control variables.

 

Regarding merging the regression tables (Tables 2 & 3, and Tables 4 & 5): We completely agree with the spirit of your suggestion. Placing the coefficients for “possession” and “perception” side-by-side would indeed offer a more direct and intuitive comparison. This was a format we initially considered. However, we encountered a significant practical challenge during typesetting. Merging the tables results in a table that is excessively wide, containing numerous variables. To fit such a table onto a page, it would almost certainly require a landscape (horizontal) orientation. We felt that forcing the reader to turn the page to read a table creates a disruptive and difficult reading experience. Therefore, to prioritize reader-friendliness and a smooth narrative flow, we opted to keep the tables separate. We hope you will agree that this trade-off is reasonable in favor of better readability.

 

Regarding the discussion of significant control variables: This is another excellent point. We acknowledge that some control variables, such as “risk preference” and “perceived environmental value,” show significant effects that are interesting in their own right. However, our decision to not elaborate on them was a deliberate narrative choice. The significance of these control variables is largely consistent with established findings in the literature, and their primary role in our models is to mitigate omitted-variable bias. To maintain a sharp and clear focus on the central theme of our paper—the "perception outweighs possession" hypothesis—we chose to streamline the results section to concentrate exclusively on our core variables. We were concerned that discussing secondary findings might dilute the main argument and distract the reader from the key takeaway message.

 

We hope this explanation clarifies our rationale. Thank you again for your insightful feedback, which prompted us to re-evaluate and confirm our presentation strategy.

 

4) As for the discussion part, the author put forward the viewpoint that policy design should focus on "perception construction", but it could go further by presenting some specific and operational policy suggestions. For instance, how should the government and certification institutions build farmers' positive perception through "strategic information dissemination, cognitive guidance and value shaping"? Should it be through more training, setting up demonstration households, leveraging village social networks, or through media publicity? Offering more specific suggestions will greatly enhance the policy reference value of this article. Additionally, this study found that "holding a forest tenure certificate" and "participating in certification" themselves were not significant. The discussion section attributed this to the fact that "perception" is more important. This is indeed the core conclusion, but it could also explore why "holding" itself fails. Regarding the forest tenure certificate, is it because local government enforcement is insufficient, or due to historical policy fluctuations, causing farmers to lack trust in this "piece of paper"? As for certification, is it because the market premium brought by certification is not obvious in reality, or because middlemen have seized most of the profits, leading to farmers' poor actual participation experience?

R: 

Thank you for this incredibly valuable and constructive feedback. Following your excellent advice, we have added a new, detailed paragraph outlining these specific policy strategies. You will notice that we have incorporated this new section into the Conclusion rather than the Discussion. This was a thoughtful structural decision made in consideration of feedback from both reviewers.

 

Another reviewer suggested that our original conclusion was too brief and should be expanded. By placing the detailed policy recommendations in the Conclusion, we were able to address both suggestions effectively in a single, cohesive section. Moreover, we believe this structure offers a stronger narrative flow. The Discussion section remains focused on interpreting our findings in the context of academic theory and prior literature, while the Conclusion now serves as a more powerful final section. It not only summarizes our key findings but also translates them into impactful, real-world applications, providing a strong, forward-looking closing for the manuscript.

 

Additionally, We completely agree that a deeper exploration into why formal instruments like “holding a certificate” fail to produce significant effects is a critical and fascinating area of inquiry. The potential reasons you’ve raised—such as insufficient government enforcement, historical policy distrust, lack of obvious market premiums, or profit capture by middlemen—are indeed highly plausible and important explanations.

 

While we find these questions extremely compelling, we believe that rigorously investigating these specific institutional and market-based mechanisms would require a different research design and data sources (e.g., qualitative interviews, detailed market-chain analysis, or longitudinal institutional data) that are beyond the scope of our current quantitative study. Our study was primarily designed to test the relative importance of subjective perception versus formal status, and our dataset is well-suited for that specific question. Adding a discussion on these underlying causes without direct empirical support from our data would amount to speculation, which we feel could detract from the scientific rigor of our core findings.

 

However, we believe your suggestions are too valuable to be ignored. Therefore, in light of your excellent feedback, we have explicitly incorporated these potential explanations into our “Limitations and Future Research” section (Section 5.3). We now state that future research should focus on these very questions to build upon our findings and untangle the complex reasons behind the failure of formal institutions.

 

Thank you again for pushing us to think more deeply about our findings. Your feedback has been instrumental in helping us better frame the boundaries of our current work and outline a clear agenda for future research.

 

Reviewer 4 Report

Comments and Suggestions for Authors

“Households’ management decisions are 17 driven more strongly by their subjective perceptions than by objectively held policy instruments.” I am wondering whether there is somewhere in the world a household deciding on the basis of objective tools and not on subjective assessments or criteria. Try to let your basic hypothesis seem less self-evident and really needing research.

 

“it deconstructs the effects”. I suggest “it disputes the effects”

Fig.1 should be further clarified concerning “perceptions” as well as the rest variables (some examples would help)

“mu (0.067 hectares)”. If this is a unit of area, clarify in tables as well.

“integrated land”. Clarify

“Will own forestland in 10 years”. Is it perceived tenure or a perceived acquirement?

Clarify exactly what you mean with “tenure” in this study.

“1,772 yuan/mu”. If you want to provide such info, then provide also some more financial info (e.g. average income per capita annually or some other characteristic parameter) so that someone can compare.

 

In table 1, exemplify in terms of % the classification “1=Flat; 2=Moderate; 3=Steep”

 

“integrating social network analysis”. That sounds interesting. Exemplify however what could be e.g. the nodes and the links in this case.

“This unique combination of resource endowment 269 and institutional advancement renders it an ideal sample…”. I suppose this makes this sample preferable as a purposive sample?

“Using combined purposive and random sampling”. Explain. Do you mean, you selected the area or a certain wide group as a purposive sample due to certain criteria (which ones?) and then, within this, you selected the individual householders randomly (on the basis of a total list? Was is an official valid list? Where did find it?), or what?

Author Response

We gratefully thank the Reviewer for your review of our paper. All your suggestions are very important to us. The detail responses to all the issues raised by the Reviewer are listed in the following, and all of your suggestions have been addressed in the revised manuscript (highlighted).

Response to the Reviewer#4 

1) “Households’ management decisions are driven more strongly by their subjective perceptions than by objectively held policy instruments.” I am wondering whether there is somewhere in the world a household deciding on the basis of objective tools and not on subjective assessments or criteria. Try to let your basic hypothesis seem less self-evident and really needing research.

R:

Thank you for this truly excellent and deeply insightful point. You have raised a crucial question that pushes us to sharpen the fundamental premise of our research. We absolutely agree that on a basic level, all human decisions are filtered through subjective assessments. Our goal is not to prove this self-evident truth, but to challenge a dominant paradigm in policy-making, which your comment has helped us to articulate more clearly.

 

As stated in our introduction, much of contemporary policy (such as tenure reform) is designed and evaluated based on the distribution of objective, formal instruments (e.g., certificates). This approach implicitly assumes a strong, direct link between the instrument and the desired behavioral outcome. Our study is motivated by the “significant ‘perception gap’ ” that often exists between this institutional supply and households’ actual responses.

 

To address your excellent point and better frame our contribution, we have revised the end of the second paragraph of our Introduction. We now explicitly state that our research moves beyond the seemingly self-evident premise that “perception matters” to ask a more nuanced and critical empirical question with profound policy implications: What is the relative explanatory power of the objective policy instrument versus the subjective perception in driving management behavior?

 

We clarify that by quantifying and comparing these effects, our research provides a non-trivial finding: an empirical validation of a critical, yet often overlooked, flaw in conventional policy logic. This reframing, prompted by your comment, better highlights why our research is necessary and what its core contribution is.

 

Thank you again for this superb comment. It has been instrumental in helping us make our paper's central argument much more compelling.

 

2) “it deconstructs the effects”. I suggest “it disputes the effects”

R:

Thanks for your valuable comments. We have revised the corresponding expression in the manuscript.

 

3) Fig.1 should be further clarified concerning “perceptions” as well as the rest variables (some examples would help)

R:

Thank you for your very helpful and constructive suggestion regarding Figure 1. Following your advice, we have substantially revised Figure 1 to make the model more concrete and to better visualize our core research framework and findings.We have also updated the figure caption to explain this new, more informative structure.

 

4) “mu (0.067 hectares)”. If this is a unit of area, clarify in tables as well.

R: 

Thanks for your valuable comments. As suggested, we have added a clarification in Table 1.

 

5) “integrated land”. Clarify

R: 

Thanks for your valuable comments.We have revised the confusing phrasing in the latest version.

 

6) “Will own forestland in 10 years”. Is it perceived tenure or a perceived acquirement?

R: 

Thank you for this excellent and very sharp observation. You are absolutely right that the phrasing used in our measurement table was ambiguous and could be misinterpreted as a perceived future acquirement. We are grateful for the opportunity to clarify this.

 

Our intention with this variable was to measure the respondent's confidence in the long-term stability of their existing tenure. The actual survey question was, “Do you believe you will be able to continue to hold your family's current forestland for the next 10 years?”

 

The ambiguous description appeared only in our table detailing the survey items. To completely remove this ambiguity and accurately reflect what was measured, we have corrected the variable's description in that table. As per your feedback leading to this clarification, the variable is now precisely described as: “Expectation of continued possession (10 years)”.

 

Our intention with this variable was to measure the respondent’s confidence in the long-term stability of their existing tenure. The actual survey question was, “Do you believe you will be able to continue to hold your family’s current forestland for the next 10 years?”

 

The ambiguous description appeared only in our table detailing the survey items. To completely remove this ambiguity and accurately reflect what was measured, we have corrected the variable’s description in that table. As per your feedback leading to this clarification, the variable is now precisely described as: “Expectation of continued possession (10 years)”.

 

7) Clarify exactly what you mean with “tenure” in this study.

R: 

Thank you for your very constructive feedback regarding the clarity of our conceptual framework and the definition of our key terms. Your comments prompted us to significantly refine this section to better articulate our study's core logic.

 

Following your suggestions, we have revised and integrated our core definitions into a single, comprehensive paragraph in the revised manuscript.

 

This new paragraph now provides the precise, operational definitions you requested. It explicitly disentangles both 'tenure' and 'forest certification' into their two core dimensions: Objective Status (i.e., holding a certificate, formal participation) and Subjective Perception (i.e., expectation of continued possession, perceived benefits).

 

This revised text now works in direct concert with the updated Figure 1, providing a clear and unified narrative for our conceptual model and creating a much stronger foundation for the hypotheses that follow.

 

We are confident that this concentrated revision addresses your points effectively and significantly improves the clarity of our theoretical framework. Thank you again for your valuable guidance.

 

 

8) “1,772 yuan/mu”. If you want to provide such info, then provide also some more financial info (e.g. average income per capita annually or some other characteristic parameter) so that someone can compare.

R: 

Thank you for the very thoughtful suggestion to provide a comparative context for the "1,772 yuan/mu" investment figure. We agree that helping the reader understand its scale is important, and your comment has prompted us to add a critical clarification about the nature of this variable.

 

We would like to clarify the specific composition of this key variable. The 'total inputs' figure of 1,772 yuan/mu is a comprehensive economic measure that includes not only direct cash outlays (e.g., for fertilizer and hired labor) but also the imputed opportunity cost of the household's own land and labor.

 

Because a significant portion of this 'cost' is a non-cash opportunity cost, directly comparing the full 1,772 yuan figure to a household's annual cash income would be highly misleading. It would substantially overstate the actual financial burden and cash expenditure required from the farmer within the production cycle. A reader might incorrectly conclude that farmers are spending a very large fraction of their cash income on bamboo, when much of that 'cost' is a non-cash, economic calculation for which no money actually leaves the household.

 

Therefore, to provide context in the most rigorous and accurate way, we have revised the manuscript to explicitly state the composition of this input variable. The text now clarifies that it includes both direct outlays and these imputed opportunity costs. We believe this clarification is the most methodologically sound way to address your point, as it explains the nature of the figure itself, rather than risking a flawed comparison.

 

Thank you again for prompting this critical clarification, which has certainly improved the transparency of our methodology.

 

9) In table 1, exemplify in terms of % the classification “1=Flat; 2=Moderate; 3=Steep”

R: 

Thank you for this sharp observation and for prompting us to clarify the nature of this variable. You are correct that the coding “1=Flat; 2=Moderate; 3=Steep” suggests a categorical variable at first glance.

 

We would like to clarify that in our analytical framework, 'Average slope' is treated as a continuous variable, representing an ordered score of steepness. This approach is a common practice in applied econometric analysis within our field. Given this treatment, the mean of 1.819 provides a meaningful description of the central tendency for our sample.

 

To eliminate any potential ambiguity for our readers and to address your valid point, we have added a note at the bottom of Table 1. This note explicitly states that this variable, despite its coding, is treated as a continuous scale in our analysis.

 

10) “integrating social network analysis”. That sounds interesting. Exemplify however what could be e.g. the nodes and the links in this case.

R: 

Thank you for your interest in this point and for the excellent suggestion to exemplify how social network analysis could be applied in this context. We agree that providing concrete examples makes our suggestion for future research much more tangible.

 

Following your advice, we have expanded this part of our “Limitations and Future Research” section (Section 5.3) to provide these specific examples.

 

In the revised text, we now clarify that in such a study, the nodes would be the individual households or farmers within the community. The links could represent various types of social ties that are crucial for the diffusion of information and perceptions. We provide examples such as information sharing networks (i.e., ‘who discusses forestry management with whom’) or trust networks (i.e., ‘whose advice on policy is considered reliable’).

 

We further elaborate that a future study could then analyze how a household’s structural position within these networks—for instance, their centrality or their proximity to key opinion leaders—influences the formation of their subjective perceptions regarding tenure security and certification.

 

We believe this added detail makes our suggestion for future research much more concrete and actionable. Thank you for pushing us to develop this idea further.

 

11) “This unique combination of resource endowment and institutional advancement renders it an ideal sample…”. I suppose this makes this sample preferable as a purposive sample?

“Using combined purposive and random sampling”. Explain. Do you mean, you selected the area or a certain wide group as a purposive sample due to certain criteria (which ones?) and then, within this, you selected the individual householders randomly (on the basis of a total list? Was is an official valid list? Where did find it?), or what?

R: 

Thank you for these excellent and detailed questions regarding our sampling methodology. As they are closely related, we will address them together here. Your feedback has prompted us to provide a much more transparent and detailed account of our sampling process in the manuscript.

 

To clarify, you were right to infer that our process began with purposive sampling at the macro level, followed by random sampling at the household level. We employed a multi-stage sampling design, which we have now explicitly detailed in the revised Section 3.5.

 

The stages were as follows:

 

Stage 1: Purposive Selection of the Case Study Area. We first selected Sanming City as our research site. The criteria for this purposive choice, as you asked, were its dual status as China's “Bamboo Capital” (ensuring high resource endowment and industry relevance) and as a “National Comprehensive Reform Pilot Zone for Collective Forestry” (ensuring advanced institutional innovation). This unique combination makes it a particularly suitable site for our research questions.

 

Stage 2: Purposive Selection of Counties & Townships. Within Sanming, we then purposively selected six key counties known for their highly developed bamboo industries. Within these counties, we further selected townships that are central to the local bamboo economy.

 

Stage 3: Random Selection of Households. This addresses your crucial question about the random component. For the final stage, we obtained official and complete rosters of all bamboo-farming households from the respective local village committees. These lists served as our sampling frame. From these valid lists, we then employed a systematic random sampling method to select the final household sample for our face-to-face interviews.

 

We have rewritten the methodology paragraph in Section 3.1 to clearly articulate this multi-stage process, detailing the criteria for each purposive stage and the basis for the final random selection. We are confident this comprehensive explanation now provides the full methodological transparency you rightly requested. Thank you again for pushing us to be more precise.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for addressing my concerns!

Author Response

Thank you very much for your valuable suggestion.

Reviewer 3 Report

Comments and Suggestions for Authors

Firstly, compared with the previous version, the new manuscript has achieved substantial improvements in theoretical depth, methodology, and policy implications. The authors have carefully addressed all the review comments, and the current version of the manuscript is relatively mature and close to the publishable level. Below are some suggestions for minor adjustments in details:

1 In the revision of the literature review and theoretical analysis section, the dual behavioral logic behind Hypothesis H4 has been clearly expounded, greatly enhancing its theoretical persuasiveness. At the same time, the core framework of "perception-over-possession" has been ingeniously linked to the classic theories of institutional economics (North) and behavioral economics (Thaler). This revision has effectively elevated the theoretical level of the article. I have no further suggestions on this.

2 Regarding the research methods and data sources section, this version clearly explains the construction method of the "perceived benefit index" (by averaging the Likert scale questions) and explicitly explains how to monetarily integrate land, labor, and capital inputs. This revision has effectively addressed my doubts. Thank you very much!

3 For the presentation of results, I accept the author's response to not merge the tables. Additionally, regarding the supplementary explanation of control variables, to further enhance the persuasiveness of the model without deviating from the core narrative, I suggest that the author briefly mention it in the analysis result paragraphs. For example, after analyzing the core variables, a sentence can be added: "It is also worth noting that the significant influence directions of control variables such as risk preference and education level are consistent with existing theories and literature findings, which increases the robustness and validity of our model setting. This way, it can reflect the richness of the results while strengthening the reliability of the model without distracting the readers' attention.

Author Response

Response to the Reviewer#3 

1) For the presentation of results, I accept the author's response to not merge the tables. Additionally, regarding the supplementary explanation of control variables, to further enhance the persuasiveness of the model without deviating from the core narrative, I suggest that the author briefly mention it in the analysis result paragraphs. For example, after analyzing the core variables, a sentence can be added: "It is also worth noting that the significant influence directions of control variables such as risk preference and education level are consistent with existing theories and literature findings, which increases the robustness and validity of our model setting. This way, it can reflect the richness of the results while strengthening the reliability of the model without distracting the readers' attention.

R:

Thank you very much for your valuable suggestion. We completely agree with your view that briefly mentioning the findings related to the control variables in the results analysis section effectively enhances the persuasiveness of our model and the rigor of the research, without detracting from the core narrative. Following your suggestion, we have added a corresponding description in the empirical results analysis section of the manuscript, after the analysis of the core variables.