Offshore Wind Farms and Tourism Development Relationship to Energy Distribution Justice for the Beibu Gulf, China
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsLand- 2989650 Can tourism landscape development of offshore wind farms effectively contribute to energy distributive justice? Experimental evidence from the Beibu Gulf Region
This is very interesting research on the relationship of offshore energy development, perceived impacts, and distributive justice on tourists and critical sensitive groups such as fisherman. I have several suggestions to improve the article.
The title is very long. I would suggest changing it to something like- “Offshore windfarm and tourism development relationship to distributive justice for the Beibu Gulf China.”
Literature review line 118- Please explain “distance decay” or use other terms.
Data sources and research methods- should note that research methodology assumes “rationale choice” in decision making. It should be noted that some social science researchers have critiqued respondents a ability to make rational choices especially regarding willingness to pay versus actual behavior.
Lines 370-371 – Please explain how “cheap talk” works.
Lines 376- 371- since CVG’s are randomly stratified sampled and tourist are convenience sampled- does this pose any problems for sample data comparability and representation?
Results and analysis
For lines 488-489 I would argue that greater energy distributive justice for CVG’s could be achieved as we don’t know if they could actually do the OWF tourism development. I would also argue that this may or may not balance out the distributive justice for CVG’s.
Discussion
Same point as above for lines 495-524. The point is that CVG’s may not perceive or operationalize the benefit offset potential of Offshore wind tourism development. Maybe this is a future research agenda.
Comments on the Quality of English Language
The English is good and needs minor copy editing.
Author Response
Manuscript ID: land-2989650 (“Land”)
Point by point response to the manuscript titled “Can tourism landscape development of offshore wind farms effectively contribute to energy distribution justice? Experimental evidence from the Beibu Gulf region”.
Dear reviewer,
Thank you very much for your time and comment on our manuscript. On behalf of the co-authors, I am submitting the revised manuscript along with the responses to the comments made by you.
Below we document our responses to your comments. Our responses start with “Response:”. The line numbers used here refer to the ones in the Word document (with Track Changes turned to “All Markup”). All line numbers are based on the revised manuscript.
Reviewer #1:
This is very interesting research on the relationship of offshore energy development, perceived impacts, and distributive justice on tourists and critical sensitive groups such as fisherman. I have several suggestions to improve the article.
Response: Thank you very much for your strong support of our work, you have provided us with very valuable suggestions to improve the quality of this paper! We have used your comments to revise the manuscript and have attached a point-by-point response to your comments.
1.The title is very long. I would suggest changing it to something like- “Offshore wind farm and tourism development relationship to distributive justice for the Beibu Gulf China.”
Response: Thanks for your advice on the title. We have revised the title with reference to your suggestion, and the revised title is “Offshore wind farms and tourism development relationship to energy distribution justice for the Beibu Gulf China”。
2.Literature review line 118- Please explain “distance decay” or use other terms.
Response: I would like to express my gratitude for your valuable suggestion. Instead of using the term “distance decay”, we have interpreted it to mean the same thing. The revised version is presented below (lines 120–127).
Revision: Coastal fishing communities in China have optimized their household livelihood structures through tourism models such as "Fisherman's Family Tourism", thus enhancing family income and reducing vulnerability [54]. However, this will inevitably be affected by the externalities of OWFs, and the closer the Fisherman's Family Tourism operator is to the coastal zone, the greater the impact (vision, noise, etc.), as tourists are often willing to rent accommodation closer to the coast in order to reduce their transportation costs [47,54].
3.Data sources and research methods should note that research methodology assumes “rationale choice” in decision making. It should be noted that some social science researchers have critiqued respondents a ability to make rational choices especially regarding willingness to pay versus actual behavior.
Response: Yes, choice experiments are based on the assumption of "rational choice", but respondents often have only limited rationality. For this reason, this study uses cut-offs to overcome the estimation error caused by respondents choosing options that deviate from their actual preferences, and cheap talk to try to bridge the gap between willingness to pay and actual behavior.
(1)Cut-offs
The inclusion of Cut-offs somewhat overcame the model estimation bias caused by respondents' information overload. Choice experiments present respondents with a series of decision scenarios and derive preference information based on their selections. However, when confronted with extensive choice data, limited by cognitive capacities, respondents frequently resort to non-compensatory strategies to reduce decision-making costs [74]. Traditional choice models that are grounded in linear compensatory principles do not account for the likelihood of respondents using non-compensatory strategies, possibly compromising the accuracy of estimation outcomes.
For example, the maximum amount that a tourist is willing to pay for OWFs is 100 yuan; however, the long interview and a series of choice sets caused cognitive fatigue for the respondent, leading him to choose the option of paying 150 yuan in the choice set, which clearly exceeds his maximum willingness to pay and will result in a biased final estimate. The Log likelihood becomes higher, confirming this.
(2)Cheap talk
Emphasize to respondents the importance of answering carefully and repeatedly confirming the reliability of their answers. The research team conducted the study with the support of the government (which objectively reflected the importance of the study and made respondents more conscientious in their answers), and we used the following techniques to motivate respondents to fill out the questionnaire conscientiously and bridge the gap between willingness to pay and actual behavior as much as possible. For tourists, we asked rhetorically, "Are you really willing to pay this amount? If you choose this amount, after the OWFs are built, you will be asked to pay the corresponding amount when you visit the area". For CVGs, we would say, "Your choices will affect the compensation you can receive in the future''. At the same time, we would also ask, "What is the minimum amount of compensation you think the government needs to make up for the loss of livelihoods caused by OWFs?". These types of questions make respondents rethink the options given and can increase the reliability of responses.
4.Lines 370-371 – Please explain how “cheap talk” works.
Response: Thanks for your suggestions. We have further elaborated on how cheap talk works in the manuscript. Due to space constraints, we have refined the flow of cheap talk in the manuscript as follows (lines 403 to 408).
Revision: Furthermore, we added cheap talk during the CE process, which improved the precision of the measure by giving full disclosure, providing a free discussion environment, and conducting multiple response confirmations to make the respondents' MWTA/MWTP closer to the true value [69,99,100].
The detailed explanation is as follows.
Cheap talk in this study is divided into three main steps as follows.
(1) All the possible positive and negative impacts of OWFs mentioned in Table A1 and Table A2 were disclosed to the respondents through videos, texts and other materials. For example, we informed tourists that OWFs would lead to better recreational fishing experience but at the same time generate noise; we informed CVGs that the placement of OWFs could lead to no-fishing zones but at the same time might attract tourists and increase tourism income.
(2) Emphasize to respondents that there is no right or wrong answer, and that it is their true feelings and perceptions that matter. This can reduce the influence of social expectations, ethical elements, etc., and allow participants to express their true preferences more freely. In this scenario, tourists can choose the option that best enhances their tourism experience, which is more in line with their real recreational choices; CVGs can also choose compensation that is more in line with their losses without being labeled as "hindering the development of green energy".
(3) Emphasize to respondents the importance of answering carefully and repeatedly confirming the reliability of their answers. The research team conducted the study with the support of the government (which objectively reflected the importance of the study and made respondents more conscientious in their answers), and we used the following techniques to motivate respondents to fill out the questionnaire conscientiously and bridge the gap between willingness to pay and actual behavior as much as possible. For tourists, we asked rhetorically, "Are you really willing to pay this amount? If you choose this amount, after the OWFs are built, you will be asked to pay the corresponding amount when you visit the area". For CVGs, we would say, "Your choices will affect the compensation you can receive in the future. At the same time, we would also ask, "What is the minimum amount of compensation you think the government needs to make up for the loss of livelihoods caused by OWFs?". These types of questions make respondents rethink the options given and can increase the reliability of responses.
5.Lines 376- 371- since CVG’s are randomly stratified sampled and tourist are convenience sampled does this pose any problems for sample data comparability and representation?
Response: Thank you for your question. Overall, the samples in this paper remain comparable and representative for two specific reasons.
(1) This study discusses whether energy distribution justice can be achieved between tourists and CVGs in the Beibu Gulf region, which requires studies of both groups to be obtained through field research in order to be comparable. By working with the local government, we have obtained a stratified random sample of CVGs. If we want to discuss whether OWF tourism can achieve local energy distribution justice, we need tourists who are willing and able to travel to the area, which means that if we use an online approach, we will create a large gap between willingness to pay and actual behavior. Willingness to pay online does not mean that tourists are actually willing to travel, especially given the high cost of travel time and money due to the large size of China. Therefore, both groups need to be studied through offline field research, which improves the comparability of sample data.
(2) Tourists in field research do not have the conditions of stratified random sampling, while the larger sample size ensures a better representation of the tourist sample. In field research, we can not know the personal information of tourists in advance, and therefore can not carry out stratified random sampling, but this problem can be overcome by increasing the sample size and according to the actual distribution of tourists consciously control the sample composition. The minimum sample size required for a representative sample for the choice experiment method is 300 (Orme, 2010; Assele et al., 2023), and the sample of 1010 tourists in this study far exceeds the minimum requirement. In addition, from the results of the descriptive statistics in Table 2, the distribution of socioeconomic characteristics of the sample is basically consistent with the characteristics of the actual distribution of tourists in the Beibu Gulf Region Statistical Yearbook, which means that the sample has good representativeness.
References
Orme, B., 2010. Sawtooth Software Sample Size Issues for Conjoint Analysis Studies. Research Paper Series.
Assele, S.Y., Meulders, M., Vandebroek, M., 2023. Sample size selection for discrete choice experiments using design features. Journal of Choice Modelling.
6.For lines 488-489 I would argue that greater energy distributive justice for CVG’s could be achieved as we don’t know if they could actually do the OWF tourism development. I would also argue that this may or may not balance out the distributive justice for CVG’s.
Response: Thank you for pointing this out. Indeed, focusing on individual CVG, OWF tourism development does not necessarily lead to energy distribution justice for each CVG, and this study only suggests that there is a potential to achieve energy distribution justice, but that we can use financial resources to underwrite the CVG individuals who are not able to be adequately compensated for OWF tourism development. Each CVG has a different ability to utilize the tourism resources of the OWF, and in practice there is a possibility that tourism revenues will not compensate for their loss of livelihood. However, the government can guide them to optimize their livelihood structures by introducing policies that encourage integration with the tourism industry. This integration ensures that tourism revenue from OWFs directly benefits CVG communities. Further, the government can identify groups or segments of the livelihood conversion process that are unable to realize offsets and compensate them monetarily accordingly, and the MWTA derived from this study can be used as a reference. Overall, the development of OWF tourism is still more feasible than the government's direct use of RMB 1.366 billion (see Table 5) in fiscal funds for livelihood compensation (the annual fiscal revenue of Beihai City in 2023 is RMB 7.360 billion).
To address this issue, we have made the following changes in section 6.2. Policy recommendations (lines 626 to 635).
Revision: For CVGs, the government should guide them to optimize their livelihood structures by introducing policies that encourage integration with the tourism industry. This integration ensures that tourism revenue from OWFs directly benefits CVG communities. Further, the government can identify groups or segments of the livelihood conversion process that are unable to realize offsets and compensate them monetarily accordingly, and the MWTA derived from this study can be used as a reference. For tourist groups, attractions should incorporate recreational activities linked with OWFs, such as leisure fishing and promoting clean energy awareness. The direct benefits from OWFs manifest mainly in tourism growth, with tourists paying for the enjoyed positive externalities.
We have also modified the specific formulation to emphasize that OWF tourism development has the potential to achieve energy distribution justice, rather than necessarily achieving it (lines 538 to 557).
7.Same point as above for lines 495-524. The point is that CVG’s may not perceive or operationalize the benefit offset potential of Offshore wind tourism development. Maybe this is a future research agenda.
Response: Thank you for your suggestion. In response, we have added the following section to the limitations and future research directions section of this paper (lines 598 to 601).
Revision: Overall, this study concludes that OWFs tourism landscape development has the potential to contribute to energy distribution justice, but how CVGs can fully exploit the potential of offshore wind tourism still deserves further research.
Finally, thanks again for reviewer’s kind comments and encouragement on this manuscript.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsI like how that paper surveys two groups and is able to compare and contrast those estimates. Below are comments on your paper in order of appearance.
1. Two citations you should add that find neutral or positive impacts of offshore wind are Carr-Harris and Lang (2019) and Trandafir et al. (2020). Carr-Harris and Lang is a revealed preference study, which makes it very valuable in this literature. Trandafir et al. is a survey, but incorporates prior exposure to the turbines, which turns out to have a positive effect on valuation of recreation with turbines present. In contrast, respondents in other survey-based work rarely or never have experience with offshore wind (eg, Parsons et al.). 2. Another article to cite is Bennear (2022). This article offers a relevant overview of energy justice issues related to renewables that can help frame your research, but also you seemingly point to a new channel of energy injustice (impact to local community) that she does not discuss, which helps frame your contribution.
3. You need to be more careful with how you phrase findings of stated preference studies. For example, you write “Close construction distance can negatively affect nearby property rents, and a loss in rental value of up to 8% has been reported when OWFs are located within 10 miles of the coast [8].” But [8] is a survey. There are no actual rent declines reported. They are hypothetical.
4. Figure 2: it would be a better design if you label the Cindicator and Tindicator variables with their name instead of just numbers.
5. In Section 3.3, your discussion of cut-offs is general. Can you give an example in this context of how that will improve modeling, or how modeling will better reflect the real world?
6. Tables 3-4: don’t write CX1, CX2, etc. Instead, write out the variable names. You should also add MWTP/MWTA to these tables in a new column. Or what would be really great is a new table that provides the MWTP/MWTA estimates for CVG and tourists side by side to facilitate comparison.
7. Why don’t you estimate Table 3 using the cutoffs?
References Bennear, L. S. (2022). Energy Justice, Decarbonization, and the Clean Energy Transformation. Annual Review of Resource Economics, 14, 647-668. Carr-Harris, A., & Lang, C. (2019). Sustainability and tourism: The effect of the United States’ first offshore wind farm on the vacation rental market. Resource and Energy Economics, 57, 51-67. Trandafir, S., Gaur, V., Behanan, P., Uchida, E., Lang, C., & Miao, H. (2020). How are tourists affected by offshore wind turbines? A case study of the first US offshore wind farm. Journal of Ocean and Coastal Economics, 7(1).
fine
Author Response
Manuscript ID: land-2989650 (“Land”)
Point by point response to the manuscript titled “Can tourism landscape development of offshore wind farms effectively contribute to energy distribution justice? Experimental evidence from the Beibu Gulf region”.
Dear reviewer,
Thank you very much for your time and comment on our manuscript. On behalf of the co-authors, I am submitting the revised manuscript along with the responses to the comments made by you.
Below we document our responses to your comments. Our responses start with “Response:”. The line numbers used here refer to the ones in the Word document (with Track Changes turned to “All Markup”). All line numbers are based on the revised manuscript.
Reviewer #2:
I like how that paper surveys two groups and is able to compare and contrast those estimates. Below are comments on your paper in order of appearance.
Response: Thank you very much for your strong support of our work, you have provided us with very valuable suggestions to improve the quality of this paper! We have used your comments to revise the manuscript and have attached a point-by-point response to your comments.
1.Two citations you should add that find neutral or positive impacts of offshore wind are Carr-Harris and Lang (2019) and Trandafir et al. (2020). Carr-Harris and Lang is a revealed preference study, which makes it very valuable in this literature. Trandafir et al. is a survey, but incorporates prior exposure to the turbines, which turns out to have a positive effect on valuation of recreation with turbines present. In contrast, respondents in other survey-based work rarely or never have experience with offshore wind (eg, Parsons et al.).
Response: Thank you for your literature recommendation. The literature you recommend is very valuable in providing evidence of the positive impact of OWFs on the vacation rental market and in raising the possibility that a prior experience increases the willingness to pay for OWFs landscapes. Accordingly, we have made the following changes to the 2.3. Impact of offshore wind farms on tourism section of the manuscript, citing the relevant literature. These are as follows (lines 140 to 166).
Revision:
Lines 140 to 150: Meanwhile, OWFs have two opposing views of the vacation rental market. A stated preference study in coastal North Carolina showed that rental value losses are likely to be as high as 10% when OWFs are within eight miles of the coast [8], but Block Island's difference-in-differences study found that OWF construction significantly increased nightly bookings, occupancy, and monthly income.
Lines 155 to 166: As offshore wind technology advances and tourists' prior experience with OWFs increases, tourists are likely to pay more for locations where they can see the turbines. The negative impacts of OWFs on the tourism industry will become more manageable and, with the help of certain management measures, will gradually turn into positive impacts.
2.You need to be more careful with how you phrase findings of stated preference studies. For example, you write “Close construction distance can negatively affect nearby property rents, and a loss in rental value of up to 8% has been reported when OWFs are located within 10 miles of the coast [8].” But [8] is a survey. There are no actual rent declines reported. They are hypothetical.
Response: Thank you for pointing out the lack of rigor in the literature citation. Accordingly, we have made the following changes (lines 140 to 150).
Revision: Meanwhile, OWFs have two opposing views of the vacation rental market. A stated preference study in coastal North Carolina showed that rental value losses are likely to be as high as 10% when OWFs are within eight miles of the coast [8], but Block Island's difference-in-differences study found that OWF construction significantly increased nightly bookings, occupancy, and monthly income.
3.Figure 2: it would be a better design if you label the Cindicator and Tindicator variables with their name instead of just numbers.
Response: Thank you for your suggestions. We have modified the labeling design of Figure 2 (lines 293 to 294) and updated the Table A1 and Table A2 (lines 653 to 656).
4.In Section 3.3, your discussion of cut-offs is general. Can you give an example in this context of how that will improve modeling, or how modeling will better reflect the real world?
Response: Many thanks to you for your suggestion. We have added examples to the manuscript to further explain how cut-offs can better reflect the real world.
The detailed explanation is as follows.
The inclusion of Cut-offs somewhat overcame the model estimation bias caused by respondents' information overload. Choice experiments present respondents with a series of decision scenarios and derive preference information based on their selections. However, when confronted with extensive choice data, limited by cognitive capacities, respondents frequently resort to non-compensatory strategies to reduce decision-making costs [74]. Traditional choice models that are grounded in linear compensatory principles do not account for the likelihood of respondents using non-compensatory strategies, possibly compromising the accuracy of estimation outcomes.
For example, the maximum amount that a tourist is willing to pay for OWFs is 100 yuan; however, the long interview and a series of choice sets caused cognitive fatigue for the respondent, leading him to choose the option of paying 150 yuan in the choice set, which clearly exceeds his maximum willingness to pay and will result in a biased final estimate. The Log likelihood becomes higher, confirming this.
The following are the specific changes we made in Section 3.3. Modified choice experiment and research design (lines 328 to 332).
Revision: For example, the maximum amount that a tourist is willing to pay for OWFs is 100 yuan; however, the long interview and a series of choice sets caused cognitive fatigue for the respondent, leading him to choose the option of paying 150 yuan in the choice set, which clearly exceeds his maximum willingness to pay and will result in a biased final estimate. Essentially, the new model amends observable effects through the cut-offs and recalibrates them into a composite of two components: one is the utility respondent n derives from the combination of OWF attributes, and the other is the negative utility stemming from the breach of attribute cut-offs.
5.Tables 3-4: don’t write CX1, CX2, etc. Instead, write out the variable names. You should also add MWTP/MWTA to these tables in a new column. Or what would be really great is a new table that provides the MWTP/MWTA estimates for CVG and tourists side by side to facilitate comparison.
Response: We appreciate your comments and suggestions. We changed Table 3, Table 4, and Table A3 in the manuscript directly with variable names, and we added Table 5 to facilitate comparison of MWTP/MWTA estimates between CVG and tourists. We have also removed the explanation of the labels CX1, CX2, etc. in the manuscript (lines 437 to 477).
6.Why don’t you estimate Table 3 using the cutoffs?
Response: Thank you very much for your question. Instead of estimating Table 3 with cut-offs, this paper uses a traditional choice experiment model for the following three reasons.
(1) Incorporating cut-offs when measuring MWTA will overestimate the MWTA of CVGs.The research team found in the actual research process that, in the context of measuring MWTA, the CVGs will choose extreme cut-offs to gain as much benefit as possible, such as the highest compensation (even if the compensation has been much higher than their losses), lowest risk of fish losses (even if changes in fish populations do not affect their livelihoods), etc. Once such cut-offs are included in the model can lead to an overestimation of the MWTA of the CVGs.
(2) Measuring MWTP can be done using cut-offs, however, in the actual study respondents refused to pay for programs that affect their livelihoods. When we used the question "What is the maximum amount you are willing to pay for the placement of OWFs?" respondents said, "Why should I pay for a program that I have suffered a loss?" . When we asked with the question, "What is the maximum amount you would be willing to pay to avoid negative impacts of OWFs?" respondents said, "Isn't it the government's responsibility to reduce the negative impacts of OWFs on us, so why should we have to pay for it?" . Measuring the MWTP of CVGs in this scenario would result in a large bias.
(3) The use of a traditional choice experiment model does not affect the robustness of the core argument of this study. The MWTA measure itself will be several times higher than the MWTP, and if the MWTP of tourists offsets the MWTA of CVGs, it can also be a surrogate for the MWTP, which could still argue for the central thesis of this study-that landscape development of OWFs has the potential to achieve energy distribution justice between tourists and CVGs.
Therefore, instead of estimating Table 3 with cut-offs, the study uses the traditional choice of experimental model for estimation.
Finally, thanks again for reviewer’s kind comments and encouragement on this manuscript.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for making changes to your manuscript. Please make sure the final version includes those changes, including the citations that you needed help inserting.
