Consumer Willingness to Pay for Sustainable Wine—The Chilean Case

Round 1

Reviewer 1 Report

Dear Authors,

The manuscript entitled “Consumer Willingness to Pay for Sustainable Wine. The Chilean Case” deals with an interesting and current topic that fits in the journal’s scope but it has several flaws that need to be addressed.

The abstract should not be segmented based on the journal’s requirements. The results are missing from the abstract.

The literature review is weak; please, discuss previous research results on the factors influencing wine purchase/consumption decisions, with specific attention paid to the eco-certifications. Why did you choose those 5 labels? How are Chileans familiar with each of them? Moreover, wine purchase/consumption studies conducted in Chile also have to be discussed here.

In section 2 it is unnecessary to define MLR.

The questionnaire and the variables raise several questions. What is the difference between two variables mentioned in Table 1: “Have you bought eco-certified wine? (yes=1, no=2)” and “Did you purchase eco-certificated wine (yes=1_no=2)”? Why did you use only 2 categories for “How often buy sustainable wine” variable? Why did you use 16 USD as an upper limit for WTP?

Legend for “I” in Table 2 is missing.

The following comments apply to Tables 3-8. You provide a lot of models; however, it is unnecessary. Choose the model with the best fit for each case. Moreover, why didn’t you eliminate variables that are not significant? You should do that by e.g., the stepwise method, and then you have to show only the model with the remaining, significant variables. This, of course, changes the model significance and adjusted R-square, too. Please, include the multicollinearity and heteroscedasticity test results for the final models, too. Another question: Why did you build separate models for social, consumption, and interaction variables?

There are a lot of mistakes when discussing the MLR models. Related to Table 4, you state that “For the age range between 35 and 44 years, it shows that they consume 15.6% more than consumers aged 18 to 34” but it is significant only at the 10% level based on the table; you state that “consumers with a higher level of education consume 29.7% less than those consumers who have a lower level of educational attainment” but it is not significant based on the table; you state that “single people consume 17.7% less than married consumers” but it is significant only at the 10% level based on the table, and the correct value is 17.3%. Related to Table 5, you state that “when wine is bought over the Internet, consumers are willing to pay 12.6% less than when they buy from a wine store” but the correct value is 13.7% based on the table; you state that “consumers who have at one time bought sustainable certified wine are willing to pay 29.3% more per bottle” but it is not significant based on the table; you state that “those who buy wine more frequently are only willing to pay just 9.8% more per bottle” but it is significant only at the 10% level based on the table; you state that “consumers inclined to buy fairtrade-certified wines are willing to pay 8.49% less per bottle” but it is not significant based on the table. Related to Table 6, you state that “consumption decreases by 25.4% among those whose primary reason for drinking wine is to socialize with family” but it is significant only at the 10% level based on the table; you state that “respondents who gave this a score of 7, meaning that this consideration mattered to them less, consume 79.8% more wine than those who give it maximum importance” but it is significant only at the 10% level based on the table, moreover, values 4 and 5 are also significant at the 10% level which are not mentioned at all; you state that “their consumption increases by 31.2% and 25.2%, respectively” but the correct value is not 25.2% but 25.1% based on the table. Regarding Table 7, you state that “women are willing to pay 7.8% more per bottle of wine than men”, however, based on the table, women pay less than men; you state “consumers are willing to pay an additional 3.02% for a bottle of wine” but it is significant only at the 10% level based on the table. Concerning Table 8, you state that “consumers who buy from other retail stores consume 39.8% less than those who buy from wine stores” but the correct value is 33,8% based on the table; you state that “consumers who chose option (“It helps me relax”), consume 31.2% more than those who chose option (“It goes well with meals”)” but it is not significant based on the table; you state that “consumers between the ages of 35 and 44, who consume 15.7% more than younger consumers” but it is significant only at the 10% level based on the table.

Because of these mistakes, Discussion section also has to be rewritten. Moreover, in this section, results should not be repeated but they should be put in context, referring to much more previous research results. Also, theoretical consequences and practical recommendations have to be mentioned here as well as the limitations of the study. E.g., what could you say about the representativeness of the sample? Isn’t it a problem? What can this cause when interpreting the results? And finally, you say nothing about the various eco-certificates – what are your findings related to them?

In the Conclusions section you mention that “Consumption of eco-certified wines did increase.” Compared to what? To the previous time period? (You did not say anything about that.)

Finally, there are some spelling and formatting issues that need to be addressed. Line numbering is missing that makes extremely difficult for the reviewer to refer to the specific part of the text. Moreover, unnecessary spaces are used between sentences several times; in some cases punctuation marks are missing or they are unnecessary; and in Table 1 commas are missing between yes/no answers. The reference style does not comply with the journal’s requirements. And last but not least, there are some sentences that need to be revised, see, e.g., the first sentence of the abstract, the first sentence of the second paragraph of section 4.1.

Author Response

Dear Reviewer:

I hope you are very well, and first of all,  we have to say thank you very much for your valuable  feedback, that we incorporate totally  in this new version of our manuscript.

Best regards

Lionel V. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors,

 

Thank you for the opportunity to read and review your manuscript submitted to Sustainability. After reading the manuscript, I can see that you have accomplished relevant and comprehensive research on consumer willingness to pay for sustainable wine. The study undoubtedly gives valuable findings. However, some issues stop me from being convinced that the current version is suitable for publication. Therefore, I advise revision. Here is the list of recommendations:

1.      You have stated that there are more than 300 definitions of sustainability referencing the source prepared in 2006. 16 years have passed since 2006. Supposedly, during these 16 years, much more definitions have evolved.

2.      Please involve the recent scholarship in the Literature Review.

3.      What is the background for the measurement scales? You have clearly described the contents of your research instrument, but it lacks a strong rationale. Please strengthen the description of the research instrument with the scales used and references to relevant publications. What about the reliability and validity of the research instrument?  

4.      Table 1 looks pretty static. Even if it provides clear and relevant information, it is not worth providing it in the main text. My suggestion would be to move the table to the Appendix. In the main text, you could describe only the most relevant results.

5.      In the Discussion, I would like to recommend providing insights about the limitations of your research and possible trends for future research. Managerial implications would also be of high value.

 

Once again, thank you for the opportunity, and I wish you good luck in strengthening the manuscript.

Author Response

Dear Reviewer:

I hope you are very well, and first of all,  we have to say thank you very much for your valuable  feedback, that we incorporate totally  in this new version of our manuscript.

Best regards

Lionel V. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

The quality of the manuscript has increased significantly compared to the previous version. However, it still has serious flaws. First, I will go through your answers and then I will discuss additional issues.

Answers to your answers:

The literature review is still weak; please, discuss previous research results on the factors influencing wine purchase/consumption decisions, with specific attention paid to the eco-certifications. Why did you choose those 5 labels? On what you base your statement in your answer that “Labels considered are the most important ones in the market“? You have to discuss this topic in the manuscript as well as how Chileans are familiar with each of them. Moreover, wine purchase/consumption studies conducted in Chile are still missing; if it is true what you mention in your answer that “Chile is one of the top wine producers in the world,” a wealth of studies must be found on this topic.

Concerning the 2 categories for “How often buy sustainable wine” variable, please, include the explanation you gave in your answer in the manuscript as well.

I still do not understand (or it is not acceptable) why you used 16 USD as an upper limit for WTP; as you state, “This value was taken from previous studies of our research group” but you should refer to those studies in the manuscript to support your choice. Moreover, why did you use these two categories: 0-5 USD, 5-16 USD? (So why 5 USD was the separating WTP? It would be more logical to use, e.g., the following categories: 0-4.99, 5-9.99, 10-14.99, 15- .

It is still valid that in the Discussion section, results should not be repeated but they should be put in context, referring to much more previous research results. Also, specific theoretical consequences and practical recommendations have to be mentioned here. In this form, it seems that the results cannot be explained (since they are not explained, in fact); what are the consequences of your results – one by one?

You still say nothing about the various eco-certificates – what are your findings related to them? I mean are they perceived differently? Do they have different effects? Do they matter at all?

There are still a lot of unnecessary spaces between sentences and the reference style still does not comply with the journal’s requirements (please, check the formatting requirements carefully, it is the minimum that you follow all of them).

Additional comments:

In lines 78-96 the same thought is repeated at least four times; so this part needs rephrasing/restructuring.

Why are the results different on WTP according to various research studies? Could you detect any pattern or possible explanation? Does your study contribute to the understanding of this contradiction?

The date of data collection is missing.

Please, discuss all significant differences in purchase price frequencies among the different certification types, not only a few examples (it is required, since you do not provide the cross tables with related statistics).

What is the difference between C (Consumption) and I (Wine Consumption) variables (see tables 1 and 2) – except that the first one refers to categorical and the second one refers to continuous variables? I cannot see why you do not use the same type (e.g., Wine consumption – C) for both of them.

Do not highlight that you show only the significant variables in Table 3. First of all, it is unnecessary – insignificant variables should not be presented at all; moreover, the other tables’ headings do not contain this information.

What do P> (t) mean in your tables? You should simply use “p”.

When you state in tables that p=0, you make a mistake: p cannot reach zero, it can only approach it; it is only the statistical software that cannot show enough decimal places to see that it is not zero. The correct presentation of these kinds of values is p<0.001 (in tables: <0.001)

Interpretation of the results is problematic. You cannot make statements like x% increase in the independent variable causes y% increase in the dependent variable (based on the coefficients of the regression models), since your dependent variables are used as logarithms.

Please, do not refer to a table like “the following”; the final place of the table will depend on the final editing of the manuscript so it is possible that the table will precede the text referring to it.

Why does the fact that the sample included 75% males explain that it is possible to identify other factors that influence consumer behavior? (lines 287-289)

In line 300, it is not the wine price that is influenced by the variables but the WTP.

There are a lot of grammar issues and typos, see the following lines: 5, 21, 133, 154, 237, 268, 270, 271, 272, 283-284, 289, 292, 295, 300, 301, 302, 322, 328. It is highly recommended to get the text proofread by a native English speaker.

Author Response

Dear Reviewer:

We hope you are doing well. First, we want to thank you for the revision of our manuscript and the valuable feedback you provided. Below, please find the responses.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

 

Thank you for the improvements. It seems that the majority of recommendations were fully regarded. Aiming for a more superior contribution to the scholarship, I suggest you to make some more corrections that were previously mentioned:

1.      What is the background for the measurement scales? You have clearly described the contents of your research instrument, but it lacks a strong rationale. Please strengthen the description of the research instrument with the scales used and references to relevant publications. What about the reliability and validity of the research instrument?  

 

2.      Table 1 looks pretty static. Even if it provides clear and relevant information, it is not worth providing it in the main text. My suggestion would be to move the table to the Appendix. In the main text, you could describe only the most relevant results.

Author Response

Dear Reviewer:

We hope you are doing well. First, we want to thank you for the revision of our manuscript and the valuable feedback you provided. Below, please find the responses.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Dear Authors,

The quality of the manuscript has increased significantly compared to the previous version. However, it still has some problems. I try to be even more specific to help you to correct them.

In lines 54-55, the sentence “These factors motivate consumer behavior” needs to be clarified, since consumer behavior cannot be motivated; it would be correct if you said “These factors motivate [what kind of] consumer behavior” or if you rephrased the sentence.

Why is the age cohort 27-35 older? (line 96) They are rather young.

Renaming the variable group to Drivers for wine consumption is fine, but you should use this new name consequently throughout the text (see lines 197, 201, 206, 256, 279).

In line 219, it seems that your whole sample consist of males with high education attainment, etc. Be more precise with wording.

The word “organic” is missing from the list in parentheses in lines 222-223.

Elaboration of significant differences in purchase price frequencies among the different certification types is still missing – in contrast to your answer “discussion has now been included (Lines 328-378).” You have to show explicitly where you found significant differences with the related statistics (Chi-square value, p value at least, and preferably statistics showing the strength of the relationship, e.g., phi, uncertainty coefficient, lambda, etc., depending on which one is applicable in this case) around lines 224-228, and not in the Discussion section (where it is not discussed, neither).

At the end of Table 1, in the WTP variables, there is no need for > signs.

You are right, when both dependent and independent variables are ln, the coefficient is %/%. But when dependent is ln and independent is as it is, a change in X (independent variable) by one unit (∆X=1) is associated with a (exp(coefficient) - 1)*100 % change in Y (dependent variable). So your interpretations need to be revised.

I am a little bit confused; you write in lines 265-267 “In the age analysis, the oldest consumers are willing to pay less for eco-certified wine compared to those aged 18 to 34 years.” But isn’t lnprice for a conventional bottle of wine the dependent variable (the last variable in Table 2)? Or if not, which label’s WTP did you use as dependent variable? Please, clarify this in the manuscript, preferably before the Results section.

In Tables 3-8, I don’t understand the Breusch-Pagan test for heteroscedasticity values; why you use p>number? Isn’t it equal?

In line 309 you state that “Respondents are willing to pay an additional 0.0831% for a bottle of wine.”, but it is not clear what is the basis for comparison (in other words, compared to what?).

In line 343, why do you say “Conversely”, when the statement that follows is in line with the previous one?

In line 348, it seems that marital status means age and sex. Rephrase the sentence.

Answering one of your answers, I cannot see how you have included a new section in lines 379-385 that specifically discusses the concern about WTP threshold values but I don’t think that it should be included among the limitations, though.

There are still several grammar issues and typos, see, e.g., line 80 (consumers’); there is no need for comma after citations (line 104, 120, 128, 353); a full stop (.) is missing in line 125; there is an excess space in line 155; use capital letter in line 186 (Chi-square); there is no need for parentheses in lines 307-309; in line 328 “20 or 30 years” is correct; in line 340 there is no need for capital letters and a semi-colon is missing; in lines 347 and 350 there are excess full stops; and in line 358 “Internet” is the correct spelling.

Formatting issues: different line spacing in lines 76-79 and 112-115; inconsistent formatting of subsection title in line 301 (plus the word “variables” is repeated).

Author Response

Dear Reviewer:

We hope you are doing well. First, we want to thank you for the revision of our manuscript and the valuable feedback you provided. Below, please find the responses.

Answers to your answers:

• In lines 54-55, the sentence “These factors motivate consumer behavior” needs to be clarified, since consumer behavior cannot be motivated; it would be correct if you said “These factors motivate [what kind of] consumer behavior” or if you rephrased the sentence.

 

Answer: Thank you, your suggestion was incorporated (Lines 54– 55).

 

• Why is the age cohort 27-35 older? (line 96) They are rather young.

 

Answer: Thank you, your suggestion was incorporated (Line 96).

 

• Renaming the variable group to Drivers for wine consumption is fine, but you should use this new name consequently throughout the text (see lines 197, 201, 206, 256, 279).

 

Answer: Thank you, your suggestion was incorporated (Lines 196, 200, 205, 257,282).

 

• In line 219, it seems that your whole sample consist of males with high education attainment, etc. Be more precise with wording.

 

Answer: Thank you, your suggestion was incorporated (Line 218-219).

 

• The word “organic” is missing from the list in parentheses in lines 222-223.

 

Answer: Thank you, your suggestion was incorporated (Line 222).

 

• Elaboration of significant differences in purchase price frequencies among the different certification types is still missing – in contrast to your answer “discussion has now been included (Lines 328-378).” You have to show explicitly where you found significant differences with the related statistics (Chi-square value, p value at least, and preferably statistics showing the strength of the relationship, e.g., phi, uncertainty coefficient, lambda, etc., depending on which one is applicable in this case) around lines 224-228, and not in the Discussion section (where it is not discussed, neither).

 

Answer: Thank you, your suggestion. We incorporated the p value (Line 224).

 

• At the end of Table 1, in the WTP variables, there is no need for > signs.

 

Answer: Thank you, your suggestion was incorporated (Line 249-250).

 

 

• You are right, when both dependent and independent variables are ln, the coefficient is %/%. But when dependent is ln and independent is as it is, a change in X (independent variable) by one unit (∆X=1) is associated with a (exp(coefficient) - 1)*100 % change in Y (dependent variable). So your interpretations need to be revised.

Answer: Thank you for your suggestion. We incorporated the exact percentage in each case according to your indication. (Lines 267, 270, 277,279,288, 296, 298, 300,310,312, 316, 317,318,323,329-331).

• I am a little bit confused; you write in lines 265-267 “In the age analysis, the oldest consumers are willing to pay less for eco-certified wine compared to those aged 18 to 34 years.” But isn’t lnprice for a conventional bottle of wine the dependent variable (the last variable in Table 2)? Or if not, which label’s WTP did you use as dependent variable? Please, clarify this in the manuscript, preferably before the Results section.

 

Answer: Thank you, your suggestion was incorporated (Line 267)

 

 

• In Tables 3-8, I don’t understand the Breusch-Pagan test for heteroscedasticity values; why you use p>number? Isn’t it equal?

 

Answer: Thank you, your suggestion was incorporated (Lines 272-332)

 

• In line 309 you state that “Respondents are willing to pay an additional 0.0831% for a bottle of wine.”, but it is not clear what is the basis for comparison (in other words, compared to what?).

 

Answer: Thank you, your point was clarified in the text (Lines 312-315)

 

• In line 343, why do you say “Conversely”, when the statement that follows is in line with the previous one?

Answer: Thank you, your suggestion was incorporated (Line 348-349).

 

• In line 348, it seems that marital status means age and sex. Rephrase the sentence.

Answer: Thank you, your suggestion was incorporated (Line 354).

 

• Answering one of your answers, I cannot see how you have included a new section in lines 379-385 that specifically discusses the concern about WTP threshold values but I don’t think that it should be included among the limitations, though.

 

Answer: Thank you, with your suggestion we relocate our answer (Lines 343-350, 364-370, 384-388).

 

• There are still several grammar issues and typos, see, e.g., line 80 (consumers’); there is no need for comma after citations (line 104, 120, 128, 353); a full stop (.) is missing in line 125; there is an excess space in line 155; use in line 186 (Chi-square); there is no need for parentheses in lines 307-309; in line 328 “20 or 30 years” is correct; in line 340 there is no need for capital letters and a semi-colon is missing; in lines 347 and 350 there are excess full stops; and in line 358 “Internet” is the correct spelling.

 

Answer: Thank you, your suggestion was incorporated (Lines 80, 104, 120, 124, 127, 129, 154, 185, 310,312-315,329, 334, 345, 346, 350, 353,358,362).

 

• Formatting issues: different line spacing in lines 76-79 and 112-115; inconsistent formatting of subsection title in line 301 (plus the word “variables” is repeated).

Answer: Thank you, your suggestion was incorporated (Lines 76-79, 112-115).