Examining COVID-19-Related Changes toward More Climate-Friendly Food Consumption in Germany

Round 1

Reviewer 1 Report

Dear Authors

Congratulations on authoring this article.
I consider the topic interesting, the study samples are robust but the difference between the samples is large.
I would like the concepts studied to be all addressed and deepened in the literature review.
They should also do a small review of English. A Title of example: "an 40 interruption" should be written: "a 40 interruption"; "empirical 79 findings seem to be more ambiguous", it should be written: "79 findings seem to be more ambiguous".

Best regards.

Author Response

Dear referee 1,

thank you very much for your appreciative feedback on our paper! We hope, that our revision meets your expectations. According to your recommendation, we deepened our literature review in the Introductions-Section (see e.g., Section 1.3 for details) as well as in the Discussion-Section. Although our first submission-version of the paper was already checked by a native English-speaking colleague, we also carefully revised the whole paper with regard to further language issues. Please let us know, if there is need for further revision from your point of view.

Please see the attachment (we uploaded a cover letter to explain all details of the revisions to the manuscript).

Author Response File: Author Response.pdf

Reviewer 2 Report

Please see attached. 

Comments for author File: Comments.pdf

Author Response

Dear referee 2,

thank you very much for all your effort in the review-process and for your appreciative feedback on our paper!

We revised the manuscript according to all of your recommendations/ comments. Here you find a detailled list with all revisions conducted according to your comments. Additionally, please see the attachment (with our cover letter describing all revisions, we conducted, in detail).

1. I believe the reader should see the full regression specifications. It is difficult to full understand the empirical approach without seeing the models and all variables involved.

Thank you for this important recommendation to which we fully agree! But we added all details on the regression analyses in Table B2 in Appendix B. Nonetheless, within the revision process, we moved this Table from Appendix B to the Results-section of our paper (see Section 3.4.1) in order to provide readers with all relevant information on the regression analyses all at once. But if there is information, that is still needed from your point of view, please let us know and we will also add this further information.

2. The complete surveys should be included as an appendix or in tables in the manuscript, depending on the editor s input. The reader should see the complete information shared with survey participants, and it is difficult reading the questions as included in paragraphs in the manuscript.

Thank you for mentioning this issue. We now integrated all information on the scales/ items used per survey into the Table B1 in the Appendix B and referred to it in the text as the new "Table 7". Furthermore, we shortened our description of the used measures in Section 2.2 as far as possible.

3. The research questions of this study would surely be better investigated using data, and I am confident the authors agree with me on this point. People are notoriously bad at recalling and estimating their own behavior with respect to food consumption, and there s a long line of literature showing that talk is cheap and we shouldn t put too much faith in people s predictions regarding how they will behave in the future. In many cases, people may be claiming they eat more organic foods or eat less meat, but this may be based on isolated memories of trying organic bananas or experimenting with meatless Mondays, with no real implications for overall food choices. The authors should be clear on these limitations, because as written, the limitations discussed include the cross-sectional nature of the data and some constraints on the content of the surveys. The conclusion section should strongly motivate more work on these research questions using dietary recall data or household scanner data.

Thanks again for such an important recommendation! Within the revision process, we extended the discussion of this relevant limitation of our results and added the recommended implication for future research (see Section 4.2.2 (a)).

4. To my read, the two major proxies for climate-friendliness in diets are organic food and meat consumption. Are these supported by the literature? Is there a clear consensus that eating more organic foods is better for the environment, and eating meat is worse? I see no discussion of dairy, which is established as a major contributor to climate emissions. The framework of the survey and the extent to which these two aspects of dietary choices intersect with the climate need to be much better motivated and clarified.

There seems to be a misunderstanding: Although, behavioral measurement in the COR1 survey was restricted due to survey lengths issues, we did not only measured participants' meat consumption and their consumption of organic food as most climate-relevant behavioral types. We also measured participants' household food waste in the COR1 survey, which also represents in important type of climate-relevant food consumption behavior. Furthermore, within the COR2 survey, we were able to integrate several additional types of climate-relevant/ -friendly food consumption behaviors (i.e., consumption of regionally produced food, of in-season food, of food with less packaging and of ready meals).

In order to make this aspect clear as well as to provide further information for our choice of behavioral measures, we extended the according note in Appendix A: "Please note: Data acquisitions during the COR1 survey reported in this paper were part of a more extensive online survey for examining diverse changes in people’s sustainability-relevant behavioral patterns and beliefs due to the pandemic (see [50] for an over-view). That is why we had to restrict our behavioral measures to only three types of climate-friendly food consumption patterns in the COR1 survey (due to issues with survey length). Thereby, we chose several behavioral measures/ types of cli-mate-friendly food consumption based on previous research on relevant  categories of different types of sustainable food consumption behaviors proposed by Verain and colleagues [53]: On the one hand, we captured data referring to consumers’ sustainable product choices (i.e., concerning the way products are produced). Thereby, we focused on the choice of organically produced food since organic food production is perceived as highly relevant for sustainability (see e.g. [54] for an overview). On the other hand, we also captured consumers’ sustainability-relevant dietary patterns concerning dietary composition and consumption curtailment (i.e., reduced quantity) with regard to one highly climate-relevant food products (i.e, meat consumption) as well as with regard to food products at all (represented by household food waste (see e.g., [55] for further information on the high climate-relevance of peoples’ meat consumption and household food waste in Germany).

With regard to the COR2 survey, no such restrictions were relevant. Thus, we were able to extend our behavioral measures to include further types of climate-friendly food consumption.”

Furthermore, we integrated the relevant limitation of not considering dairy product consumption in our surveys in the discussion (see Section 4.2.2 (b)).

5. The practical or policy implications of the work are not clear until the end of the paper and even then are not fleshed out. In my view, the notion that diets became more climate friendly during COVID is interesting, but what of that finding? The authors mention the idea of promotion in the conclusion and on this point I would like to read more. How can governments or trade associations act on this change in consumer spending and attitudes? What can or should be done to capitalize?

Thank you very much for putting our attention to this important issue. We now discuss the practical implications of our findings in the new Section "4.2 Practical and/ or policy implications of our findings".

6. With respect to education, the COR2 survey is not in sync with the general population. Is this important in interpreting the results?

According to this recommendation, we now discuss possible consequences of the higher education levels of the COR2 sample in Section 4.2.3.

7. The explained variance of the estimated regressions is very low. This is potentially worrisome as a model diagnostic. Readers could better interpret these numbers when evaluating the complete regression specifications. The authors should discuss why this might be.

According to this recommendation, we added a discussion of the small amounts of explained variances in our regression analyses in Section 4.2.2 (c).

8. The reduction in meat consumption is, as I understand the paper, the only means by which it can be said that the results support the notion that COVID-19 resulted in climate-friendly changes to diets in Germany. Moreover, and importantly, I am not sure how consumers attitudes towards the climate interact with this finding. I do not understand this paragraph from lines 618-631:

Although we did not find any strong empirical evidence of a linear moderating effect, there was some empirical evidence of relevant interaction effects between COVID-19-related behavioral changes toward more climate-friendly food consumption and consumers personal norms: Thereby, most of the empirical evidence of such an interaction effect was found for reductions in consumers meat consumption during the time period in which the COVID-19 restrictions were in place as well as for their intended meat consumption in the post-COVID-19 period. So, although we referred to the consumption of other types of climate-friendly foods that were captured in our surveys, our empirical findings remained ambiguous. However, we still provided initial empirical evidence that consumers personal climate-protection norms in addition to promoting the main/direct effect on consumers engagement in climate-friendly food consumption behaviors can also interact significantly with temporal changes in external factors/behavioral contexts [28,29], as such changes resulted from the COVID-19 pandemic and its restrictions in Germany.

Therefore, to my reading, the finding with respect to meat consumption is observationally consistent with the fact that meat is relatively expensive and unemployment was up and consumer confidence was down during the pandemic. Moreover, meat is relatively difficult to prepare, among food groups, and many consumers ate less of it during the pandemic these reasons.

This leads to two questions for the authors. One is: can the results demonstrate that the observed decrease in meat consumption is related attitudes towards sustainability, or could this finding be driven by the economic impacts of the pandemic and lockdown? And the second is: does this distinction matter, in terms of the implications of the findings?

Thank you very much for putting our attention to this important issue! In order to directly examine the mentioned effects of economic considerations on our data, we optimized our statistical analyses by considering participants' sociodemographic features (also including their income) as covariates within our analyses (see Section 3.4.2 for an overview on the adjusted analytical procedure and the adjusted results). As shown by our results (especially referring to results of the COR1 sample), also controlling for effects of participants' income did not change our results fundamentally. That is why did not modify the according interpretation of our results in the mentioned section. Nonetheless, since the considerations on possible alternative explanations for our results based on economic issues, which were mentioned by the reviewer, seem to be quite interesting for future research, we also added an according implication for future studies at the end of Section 4.2.2 (c).

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Thank you for your careful revision of the paper. It is much improved.