Assessments of Landowners’ Willingness to Accept Compensation for Participating in Forest Certification in Shandong, China

Round 1

Reviewer 1 Report

In this manuscript, “Assessments of landowners’ willingness to accept compensation for participating in forest certification in Shandong China”, the authors analyze the factors that influence the compensation that would be accepted by landowners to get a sustainable forest management certification. An average compensation is also calculated. The conclusions are based on a thorough survey of landowners in the mentioned region in China. A couple of suggestions to cover this compensation are also briefly mentioned in the manuscript.

The manuscript provides an interesting analysis of the landowners in this Chinese region, and how different aspects (socioeconomic, type of ownership, land size, etc.) influence the compensation they would except for becoming certified. This analysis has potential for its use as basis for the development of specific measures or incentives to attract these landowners into getting certified. Nevertheless, there are several areas where the manuscript needs improvement:

1. A description of the context of the region where the surveys have been implemented is missing.

It would be useful to have a description of the region to better understand the background and the results obtained. Relevant aspects are missing, such as the predominant type and size of forests (natural forests or plantations; are they small plantations and young? Or close to harvest time?); any biodiversity hotspots in the region, and whether these are included under any protection figure; a description of the landscape, including the land uses and whether the forests are typically scattered or large portions of forests are present; etc.

A bit of background on the number of certified landowners in the region would also provide a better understanding of the context. Is this a region where forest certification is typically implemented or not? Why was this region chosen?

It would also be beneficial to include some of the benefits that forest certification brings to the landowners and the forests (e.g. environmental benefits, new forest knowledge, new business knowledge, more value from forests, higher profile). Is certification useful/used for market access? Why certification?

 

2. An indication of the type of forests owned by the landowners interviewed is missing.

Did all the landowners interviewed have plantations, except for those that indicate that their forests have a role in biodiversity conservation and recreational uses? What type of tree species are predominant in their forests?

 

3. An explanation of why the poplar value in markets is being used as basis for the calculation of the average compensation is missing.

Are poplar plantations the dominant forest type and use in the region? Do most of the landowners interviewed own poplar plantations? It´s not clear why the poplar market price is used as relevant reference to calculate the compensation average.

 

4. Discussion and Conclusion

Understanding it is difficult to obtain this information, having a reference on the average price premium obtained in the region thanks to forest certification would help understand whether the compensation calculated could be obtained based on previous price premiums for certified wood, or, as the paper indicates, some other incentive measures would have to be developed to incentivize the landowners´ participation in certification schemes. Were any certified landowners interviewed?

Additionally, one of the results of the analysis indicates that those landowners whose forests provide biodiversity protection or recreational uses expect a higher compensation to get certified.

It seems advisable to mention some existing mechanisms that can be implemented by landowners to get financial compensation for such biodiversity conservation and recreational services. There are examples that can be implemented together with forest management certification, such as the implementation of the Ecosystem Services Procedure from the Forest Stewardship Council (FSC-PRO-30-006), or examples that can be implemented separate from the forest certification, such as the creation of carbon credits or agreements with investors. The latter would serve as a previous step that would provide financial benefits and help achieve the forest certification.

 

5. Other issues

• APA guide and others suggest explaining the ‘local’ metrics, i.e. both USD and RMB could be used.
• Except in the abstract, there is no other reference to the specific certification scheme or specific questions in the survey: FSC, PEFC, CFCC (=PEFC)?.
• Line 48, and 274: it might not be clear to the readers that an interest to participate, or to pay for certification, does not automatically mean that a certificate will be granted. It would be wood to clarify this.
• Line 100: 100: what kind of landowners? And what kind of forest (plantation, species, size, rotation, ...) - and "heavily forested cities" - needs a bit more explanation (poplar plantation mentioned later - not enough).
• Line 111: this info "survey questions included .... socio-demographic information about landowners [16,17]" is not enough. The survey questions should be more easily available to the reader. In the “2. Materials and Methods / 2.1. Data Collection section” you should explain what your survey questions were, but not simply referring to earlier papers.
• Line 172: "the survey; and about 20% of the respondents indicated that even they heard forest certification.." – it is better to refer to ‘forest management certification’, to clarify that ‘the management is certified’, not ‘the forest’ or ‘the timber’.
• Line 220 – Table 2.: why ‘Occupation: --Forester/loggers/miner (%)’ in one group? Miner has not much to do with the forest management background (even logger is tricky).
• Line 327: This is a finding from an earlier Poudyal-study, but how far is it a quote, and how much is it the paper´s new finding? and why the text seems to indicate that owners will either get certified or convert? What´s the relationship between the Poudyal reference and the findings of the paper – why long term owners demand a WTA and how is this related to the potential for conversion?
• Line 345: "in Shandong" (not: "in China").

 

6. Minor issues

• Figure 1: different colors should be used to identify each category of the graphic. With the scale of grey that is currently used it is very difficult to identify each category.
• Regarding the English language and style, a revision of the manuscript is required, including, but not limited to, the following lines and words: 29 (‘polices’), 45 (‘"Although there are differences in the certify standards and process," ---> "in the certification schemes' standards and processes"), 47 (‘progressive’), 52 (‘improve’), 53 (‘motivate’), 55 (‘manage’), 58 (‘toward’), 59 (‘program’), 71 (‘environmental’), 80 (‘participate forest’), 105 (‘interest’), 148 (‘weather’), 179 (‘of’), 304 (‘signification’), 335 (‘profession’), etc.

Author Response

we provided a point-to-point response to the reviewers comments in the uploaded cover letter. 

Author Response File: Author Response.pdf

Reviewer 2 Report

The article attempts to carry out the assessments of landowners' willingness to accept compensation for participating in forest certification in Shandong China. Data and information for evaluation are obtained from a survey that was conducted in 2016. Here is the origin of the circumstances leading to the rejection of the article. The results of the survey were published in at least two journals (https://doi.org/10.1016/j.landusepol.2017.11.033; https://doi.org/10.3390/f9060361), which were cited in this revised article. Although the reviewed article seeks to provide a different perspective on the data found in the 2016 survey, it evaluates some areas of the survey that have already been published, such as Familiarity with Forest Certification presented here as Landowners' knowledge and attitudes towards forest certification. It is true that the same data/numbers are not presented again; however, the difference is due to the fact that other groups of the same sample of respondents are created and evaluated.

 

Regarding compensation for participating in forest certification in Shandong China, more detailed information is missing to quantify the following result: "Results of survey data revealed that landowners accepted minimum timber price increase was averaged 18% from current timber price for respondents to certainly have their forestland certified “. The information of this result should have been more detailed. The article does not include a question from the questionnaire related to this topic. The methodological procedure for this result and further use of this data is not developed at an acceptable scientific level. The situation is similar for the quantification of the "calculated expected / average poplar timber price". Was, for example, a simple average or a weighted arithmetic average of prices for the period 2005-2020 used? The 2016 survey on the one hand and the period of time considered for poplar wood prices (2005-2020) on the other hand are also problematic.

Author Response

we provided a point-to-point response to the reviewers comments in the uploaded cover letter. 

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper reports the results of a research assessing the landowners’ willingness to accept compensation for participating in forest certification programs in Shandong China based on an econometric analysis of the responses of 507 landowners. The topic is interesting and within the scope of Sustainability. However, the paper has some major shortcomings which must be fixed before it might be considered for publication.

General issues

(1) The paper conceptualize individuals as landowners. According to my knowledge there is no individual land ownership of forest land in China. Forest lands is either owned by the state or by local communities. Sure, there are individual use (management) rights, but I don’t think it’s correct to call it “ownership” in any legal or conceptually correct way. I might be wrong, but this issue needs clarification.

(2) It does not become clear how the WTA question was exactly asked and how the different program characteristics were introduced to the respondents. It’s very important to provide sufficient details of the data gathering procedure. Please clarify and add.

(3) The econometric model has severe flaws since the scales of the variables (nominal, ordinal or metric) are not adequately taken into account. Therefore all results and their interpretation need to be put in question. Variables must be properly reported and entered into the econometric model.

(4) The choice of the econometric model is not sufficiently reflected as well as the questions whether or not some typical post-estimation tests were performed (heteroscedasticity, normality of residuals, etc.). I would prefer to see not only one model specification in the paper (significant variables) but different specifications (full model) to be able to check for the robustness of the results. Also the overall model quality indicators should be presented in the tables. Furthermore the interpretation of the estimation results should not only focus on the sign of the influence (positive or negative), but should also reflect on the coefficient and the magnitude of the influence of determinants on the WTA.

(5) Please make a statement in the paper about the data availability. Actually, I would like to see the data set and the statistical commands to be able to verify your results. Thanks!

 

Specific issues

L 41: “Therefore, China began…” Who in China? The Chinese government? Forest companies? Be more specific.

L41-43: “…China has three major certification programs including Forest Stewardship Council (FSC), Programme for the Endorsement of Forest Certification schemes (PEFC), and China Forest Certification Council (CFCC).” Please clarify for the reader how FSC, PEFC and CFCC are interrelated in China. Are these programs independent from each other or interlinked?

L 49: “…exceeded 9 million hectares in 2016 [4].” The 9 million hectares should be put into relation. What is the share of certified forests in the total?

L 53-54: “…motivate them to participate in this program also…” Which program? There are different programs right? Clarify.

L 56-80: In the literature review, attention should be paid to the geographical and institutional background of the studies (location). The different background of the cited studies are currently not mentioned at all. Please revise.

L 94: Please state clearly your contribution to the literature in answering these questions. Where do you see the value added?

L 100: “…survey of randomly selected landowners.” How did you manage to randomly select landowners? Please provide more details on the selection procedure. Who was considered as landowner? Only individual households or also companies? Clarify and provide more details.

L 108-110: “Specifically, respondents were asked to indicate the percentage increase in stumpage price (relative to current price) for them to consider certainly participating in certification program.” Please add the exact wording of the question asked. How did you introduce the different attributes of the certification program. That’s not clear. Clarify and provide more details.

L 115: “2.2 Empirical Model” might better read “2.2 Econometric Model”

L 128: “The econometric estimation approach employed in this research was multivariate ridge regression…” Please provide a justification for the choice of the multivariate ridge regression. Why was this necessary? Please also mention the software you used to implement the multivariate ridge regression.

L 130: “…participating in forest certification…” Shouldn’t it read “…participating in a hypothetical forest certification…”? I guess none of the land owners actually participate in forest certification. Please clarify.

L 142-144: “Compensation for adopting forest certification = f (Sociodemographic, Ownership size, Tenure,  Acquisition Mode, Management Objectives, Harvest history, Owners Familiarly and Interest in Certification Program, Percentage of Return from Forest)” Actually, I don’t think that this equation represents the econometric model. Clarify.

L 160: Table 1: Please add the number of observations to the table heading: (n=xxx)

The variable ACQIZ is a multi-nominal variable. “Acquisition mode of forestland (1 = purchase, 2 = inherited, 3 = rent/traded)” It does not make any sense to present the mean and SD. There is also no meaningful order. It should be better presented as a set of dummies, fixing a reference.

The variable GENDER should better be coded as a dummy, that is “Gender of landowners (1 = male, 0 = female). Much easier to interpret.

The variable EDU is a multi-nominal variable: “Highest grade of school that landowners completed (1 = High school degree/vocational training, 2 = Did not complete high school, 3 = Bachelor/higher)”. It does not make any sense to present the mean and SD. There is also no meaningful order. It should be better presented as a set of dummies, fixing a reference.

The variable OCCU is a multi-nominal variable: “Occupation of landowners (1 = Forester/loggers/miner, 2 = Professional manager, 3 = Government employee, 4 = Farmer, 5 = Retired/businessman/others)”. It does not make any sense to present the mean and SD. There is also no meaningful order. It should be better presented as a set of dummies, fixing a reference.

The variable LIVAREA can be considered as a multi-nominal variable or as an ordinal variable. “Landowners living community (1 = Metropolitan area/suburb area, 2 = town/county, 3 = rural area/village)”. You need to make an argument for an ordinal scale.

L 180-182: “To test the hypothesis that landowners might be more interested in adopting forest certification if they were more aware of this concept, we used Kendall's rank correlation tau to examine their correlation.” This is an ad-hock hypothesis which was not introduced before. How did you derive at the hypothesis?

L 186: “Figure 1. Familiarity with forest certification among respondents” Please add the number of observations: (n=xxx)

L 220: Table 2. Please indicate the meaning of the superscripts a and b in notes to the Table. Please also indicate the number of observation of each group in the table. Why are not all relevant variables included in this comparison? LIVAREA is missing. Why? Why are some forest resource characteristics missing? SIZE and FURHA is included by not ACQIZ and other. Why?

L 244-246: “Similarly, the positive coefficient before living area (p < 0.05) suggested that respondents who lived in rural/village area demanded a greater WTA compensation to have their forestland certified.” This can’t be concluded from the results. The variable is not a dummy variable indicating precisely rural/village.

L 248-250: “A similar result was observed for the acquisition mode of forestland (p < 0.05), indicating that if a landowner rented/traded for his/her current managed forestland, he/she needed a higher WTA amount than those who inherited it.” This can’t be concluded from the results. The variable is not a dummy variable indicating precisely rented/traded.

Author Response

we provided a point-to-point response to the reviewers comments, please see attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

The article is of high significance and topicality.

I would like to give some suggestions for article improvements:

-the paragraph of 67-80 lines seems like a part of discussion part rather the part of introduction. Could you revise it.

-the data collection presented very briefly, could you revise it and make mode in details.

-it is not really clear why multivariate ridge regression is selected. There is lack of method substantiation. Where are other quite suitable models, explored in WTA/WTP analysis as logit, MNL or etc. Why that regression was chosen?

-maybe you can revise the formula 2. It could be better to code it and make as formula.

 

Author Response

we provided a point-to-point response to the reviewers comments in the uploaded cover letter. 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The addition and modification of the text of the article led to a significant increase in the quality of the reviewed article. It is clear now that the article builds on previous published research and at the same time brings new (unpublished) information in a significant way.

I have the following comments on the article:

The sentence on lines 45 and 46 cannot be accepted. Both systems (FSC and PEFC) are internationally recognized certification schemes. The difference is that wood processors (in addition to other entities) were involved in the origin of PEFC. However, this does not mean that PEFC is intended for industrial forests.

Link the new separate paragraph on lines 166-173 to the information on line 164. At first sight, it is not clear that the new paragraph signifies a survey question.

Author Response

The addition and modification of the text of the article led to a significant increase in the quality of the reviewed article. It is clear now that the article builds on previous published research and at the same time brings new (unpublished) information in a significant way.

Response: Thank you so much for your review and comments.

I have the following comments on the article:

The sentence on lines 45 and 46 cannot be accepted. Both systems (FSC and PEFC) are internationally recognized certification schemes. The difference is that wood processors (in addition to other entities) were involved in the origin of PEFC. However, this does not mean that PEFC is intended for industrial forests.

Response: As suggested, we revised this sentence into following: (Line 45-49)

“Both FSC and PEFC are internationally recognized certification schemes and the difference is that wood processors (in addition to other entities) are more involved in the origin of PEFC. System of CFCC is a program developed by China itself and is endorsed by PEFC. ”

Link the new separate paragraph on lines 166-173 to the information on line 164. At first sight, it is not clear that the new paragraph signifies a survey question.

Response: Good point. As suggested, we added a sentence to link this paragraph (Line 176)

“The specific WTA question included in the survey was presented as:”

Reviewer 3 Report

Thank you very much for the revision of your paper. Actually, I’m quite puzzled by some of your responses and need to request further major clarifications and revisions.

(1) You argue that because you want to be consistent with your former publications you keep the terms “landowners”, “forestland owner households” and so on. However, although there was a forest tenure reform in China, household forestland ownership doesn’t exist so far in China. Chinese colleagues I asked confirmed this fact/information. You can’t insist that you want to be consistent with a false claim! You need to correct it. Otherwise you will further increase confusion. Clarify and revise.

(2) I’m very skeptical about the way in which the WTA question was asked. The question is so fuzzy that I doubt that answers are reliable. Further, if I now understood correctly, some households already participated in certification programs. However, the information on this participation is completely missing except the information on the “certification program design characteristics”. Is it correct that 65% or more of the households already enrolled in a certification program? At least this is what the variable RPFMH suggests. What is then the sense of asking the WTA question is the way you did? I don’t understand. It does not make sense. Also the information that 65% enrolled in a certification program is not consistent with the claim that “almost 38% of the respondents reported that they had never heard about the concept” (line 249-250) before the survey. Actually, it is still not possible to get a good understanding of the information you gathered. Please clarify.

(3) The data is a partly inconsistent. In the first version, averages of multi-nominal variables were calculated. However the values do not match the values for the derived dummy variables, of example for the former variable LIVAREA the mean 2.49 (1 = Metropolitan area/suburb area, 2 = town/county, 3 = rural area/village) does not match the set of dummy variable means 0.23 - Metropolitan area/suburb area, 0.44 - town/county, 0.33 - rural area/village, since 0.23*507*1+0.44*507*2+0.33*507*3 = 2.1. The same holds for the variable ACQIZ and EDU and the respective set of dummies. As a consequence, I have great difficulties to trust your calculation and your data set. Please clarify, if possible.

(4) I really regret that you can’t make the data set available. Please be aware that it’s then impossible to verify or replicate your analysis. However, could you please send at least a copy of the approval letter of your university (with the allowed and excluded terms of data use) and the original English version of the questionnaire. Thanks!

(5) Still only one model specification is presented instead of multiple ones and no summary statistics on the model fit (quality) are presented. I can’t believe that the model was estimated without an intercept. Clarify. Please also report the base group clearly if you include a set of dummy variables. Extend and revise with a consistent data set.

 

Author Response

See the response in the cover letter

Author Response File: Author Response.pdf

Round 3

Reviewer 3 Report

Thank you very much for the revisions, explanations and additional documents. I appreciate very much that you documented the IRB approval procedure and the questionnaire. However, at the current stage I can’t support the publication of your manuscript since significant revisions are still necessary (if they are possible).

(1) It is not enough to replace “forestland owner” just by “forest household”. You need to say something in the introduction about the forest land tenure in China to make it understandable and explain your terms. Also in your questionnaire your first question was about the ownership status of the forest (either state or collective. By the way, why don’t report the results for this question?). Then you continue with question “3. How many hectares of forest lands do you own?” Now, who owns the forest? How is it possible to purchase, inherit or trade households’ forest lands when households don’t own the forest? (Question 5). You can only solve this dilemma of terms by explaining that households in China can have time limited forest use and management rights (30 to 70 years). These rights can be transferred in different ways, they even can be mortgaged. This is a kind of ownership or property, but not in forests or forest lands, but in time limited forest use and management rights. Thus, there is no other option than to explain the reader the forest tenure system in China and justify then the use of terms!

(2) After reading your questionnaire, I’m completely lost with the variable group “Certification program design characteristics” in your manuscript. First, I realized that the “Certification program design characteristics” refer to hypothetical situations, but also that they are in no way linked to the WTA question. Second, I wonder how the variables like RMNP, ROPI, RPFMH, PICP RIRP could be constructed from the questionnaire. Let’s take the example of RPFMH which should represent “Dummy variable, whether required to use a profession forester to manage and harvest forestland (1 = Yes, 0 = No)”. How could this variable be derived from question 16 f where you asked “Would you participate if you were: (not) required to you a professional forester when managing your forest for harvesting timber”, with answers from “Not at all likely” to “Extremely likely”? I don’t think it’s possible to construct the variable RPFMH from the answer to the question 16f. There is something deeply flawed here. Any correction might affect your results and its interpretation significantly.

(3) Reading your questionnaire, I even have less trust in your WTA question. It is now obvious that you never introduced the FSC, PEFC and CFCC certification schemes to the respondents. But then you asked “please write-in an answer about your minimum acceptable timber price increase percentage (%) from the current price for you to surely consider having your forestland certified under available schemes of FSC, PEFC, and CFCC. ” How should the respondents have any understanding of what FSC, PEFC, and CFCC certification means?

(4) You claim that you double checked the data. However, the data is still partly inconsistent. Just compare the Acquisition mode of forestland in Table 1 with Table 2. Inconsistent. Also in Table 1 “Yea = 1” must read “Yes=1” for the education levels.

(5) Still only one econometric model is presented (instead of multiple) and the logic of which variables were included and which one not  (model specification) is not explained. Thus, readers can’t check the robustness of the results depending on the model specification. (Since you can’t make the data set available to the reader, this information is very important). As regards the variables: Why did you include for instance include for Living area only Rural Village and not Rural Village and Town/country? This might be due to reasons of multicollinearity, but it should be mentioned.

(6) The meaning of the base group doesn’t seem clear to you. In your model (Table 3) the base group are female households, that did complete high school, are living either in metropolitan/suburb or town/country areas, are either businessmen, retired, loggers ore others, and have inherited the forest! You always need to take into account that if you include a set of dummies, the remaining 0 are always the based group which is represented in the intercept and the influence of the dummy variables are always estimated relative to that base group. This must be observed in the interpretation of the results. Also the base group for each category should be clearly indicated in the Table 3.

Overall the data and the manuscript need significant revisions. In any case, if the paper ever will become publishable I request you to publish the questionnaire as a supplementary material together with the paper. That should be possible.