**Understanding Attitudes towards Reducing Meat Consumption for Environmental Reasons. A Qualitative Synthesis Review**

## **Ruben Sanchez-Sabate 1,\*, Yasna Badilla-Briones <sup>2</sup> and Joan Sabaté <sup>3</sup>**


Received: 8 October 2019; Accepted: 1 November 2019; Published: 8 November 2019

**Abstract:** Meat-based diets are the norm in Western societies. This is a problem because meat production is a major contributor to global warming and environmental degradation. Despite the urgency to reduce meat consumption, quantitative studies have shown that there is only a small minority of consumers aware of the meat environmental impact, willing to halt or reduce meat intake for ecological reasons, or who have already stopped or reduced meat consumption because of environmental concerns. We conducted a qualitative synthesis reviewing studies that looked at attitudes towards changing meat consumption. Our focus was on the behavioral change process: Awareness, willingness, and change, aiming to enhance the current understanding of people's attitudes towards reducing meat consumption due to environmental concerns. The studies reviewed show that consumer awareness is hindered by beliefs about food, meat, and personal behavior. Nutrition, health, and taste were found to be both enablers and barriers with regard to willingness. Vegetarians and vegans perceive the environment as simply another reason, among others, to maintain a meatless diet. Based on these results, we offer recommendations for future dietary public health interventions, and for future research endeavors on this topic. This review employed a meta-aggregative approach and partially followed the Joanna Briggs Institute methodology for systematic reviews of qualitative evidence.

**Keywords:** consumer attitudes; meat consumption; environment; sustainability; meatless diets; meat avoidance; dietary behavior change; global warming; climate change

#### **1. Introduction**

Meat consumption is a major contributor to global warming and environmental degradation [1–5]. The livestock industry pollutes and depletes fresh water, contributes to the loss of biodiversity, and is a major source of anthropogenic greenhouse gas emissions [1]. Since the worldwide demand for meat is increasing due to rising incomes, growing populations, and other sociocultural factors [1,6,7], the health and well-being of the global population are every day at greater risk. A recent assessment estimates that climate change will increase future risks of armed conflict [8], while the recently published EAT-*Lancet* report "Food in the Anthropocene" [9] warns that unless red meat consumption is significantly reduced, it will be impossible to feed, in a healthy and sustainable manner, an estimated global population of 10 billion people by 2050. The latest Intergovernmental Panel on Climate Change (IPCC) special report stated that, given the current scientific evidence, there is a high degree of confidence in the potential of reducing animal-based product consumption to achieve significant mitigation of climate change [10]. The double food and environmental pyramid by the Barilla Center showcases at a glance this link between nutritional and environmental aspects of food [11].

Meat-based diets are the norm in Western societies. In countries like the United States and the United Kingdom, vegetarians account for less than 5% of their respective populations [12]. Rationalizations of meat consumption in the West are synthesized in the 4Ns: Meat is natural—it is what our biology has come to crave in the evolutionary process, eating meat is normal—it is a sociocultural practice and expectation in civilized societies, meat is necessary for humans to be healthy, and meat is nice—animal protein is tasty [13]. Consumption habits, culinary traditions, and what sociologists call cultural repertoires (culturally available unarticulated instructions loaded with values and understandings that guide people's actions) are also important factors to explain meat intake [14,15]. Given all these socio-cultural factors, it is clear that changing the current Western meat consumption pattern is challenging [16].

The Transtheoretical Model (TTM) of health behavior change [17] describes the behavioral change process as a three-stage progression that comprises awareness (precontemplation), willingness (contemplation and preparation), and change (action, maintenance, and termination). This process of change can only occur with the adoption of a positive attitude based on reasons and motivations [17]. Since climate change and environmental sustainability are pressing reasons or motivations to reduce meat intake [9], a growing body of research is looking at the influence of environmental concerns in Westerners' meat consumption. Systematic reviews of the quantitative evidence only have shown that environmental motives have a weak influence on meat consumption attitudes [18,19]. People are rarely aware of the meat environmental toll. They underestimate its impact compared with other behaviors or activities. Only a minority seem willing to alter meat consumption because of the environment, but even they would rather adopt other strategies to counter climate change. Consequently, only a small minority has altered its meat-eating patterns because of the environment, a motive more appealing to vegans and flexitarians than vegetarians. Those most aware and positively influenced by environmental concerns are female, young, and meat-reducers.

Quantitative studies generally provide little or no information on why a person is aware, or willing, or why they made a change in meat consumption habits because of the environment. Previous systematic reviews on this topic have acknowledged some of these limitations. For example, they have called for more in-depth studies on willingness [18], and for exploring the cultural and social factors attached to meat that impact willingness and dietary change [19]. Still, many more questions can be asked to qualitative evidence. Why are consumers not aware that meat consumption degrades the environment? Is it because they simply have not been educated on the topic? Could it be instead that their beliefs, perceptions, reasons or motivations conflict with the scientific evidence on meat consumption and the environment? Do they actually understand what environmentally-friendly food is? Similar questions can be raised regarding willingness. Is the low willingness to alter meat consumption for environmental reasons attributable only to their low awareness? Do environmental motives affect in any way the 4Ns (natural, normal, necessary, nice)? Is there something that can be done to increase willingness beyond educating people on the meat environmental impact either in a rational or emotional fashion? Regarding those who have changed their meat dietary patterns because of the environment, it is necessary to know how consumers understand their behavioral change. Do they see it as a must or as simply another way to help the environment? Why did non-ecologically oriented people alter their meat consumption because of the environment?

Syntheses of qualitative evidence have proved useful to complement and better interpret quantitative systematic reviews [20]. By qualitative review we mean "the synthesis or amalgamation of individual qualitative research reports (commonly called "primary research reports") that relate to a specific topic or focus in order to arrive at a new or enhanced understanding about the phenomenon under study" [21]. The value of "enhanced understanding" that qualitative evidence synthesis can provide is already widely recognized across the natural and social sciences within the evidence-based approach [20,22]. The Cochrane Collaboration Qualitative Methods Group, the Health Development Agency, The Economic and Social Research Council, and The Joanna Briggs Institute have made significant efforts to facilitate the use of qualitative evidence synthesis in disciplines such as public health and education, among others [20,23]. The efforts of these institutions are driven by the fact that qualitative evidence syntheses have the potential to contribute both to the scientific community and different kinds of institutions. Qualitative evidence synthesis may, indeed, serve to inform the creation of research instruments by identifying the significant characteristics of a phenomenon, as well as to develop actionable knowledge that can inform both policy and practice [20].

The purpose of this qualitative evidence synthesis is to enhance our understanding of attitudes towards reducing meat consumption because of environmental concerns. We expect to get a better grasp on the nature of individual and group awareness, willingness and actual change of dietary habits to protecting the planet. Hoping to inform both future dietary public health interventions in adopting healthy and sustainable diets, and future research endeavors on this topic, we have looked at what qualitative studies have to say regarding the already named stages of the behavioral change process: Awareness, willingness, and change. Our research question was: What are the enablers and barriers to increase awareness, willingness, and change when consumers are prompted to reduce meat consumption because of environmental concerns? A preliminary search of PROSPERO, MEDLINE, the Cochrane Database of Systematic Reviews, the JBI Database of Systematic Reviews and Implementation Reports was conducted and no current or ongoing qualitative evidence syntheses on this topic were identified.

#### **2. Materials and Methods**

#### *2.1. Search Strategy*

The articles considered for this qualitative evidence synthesis were identified through a literature search of the Web of Science (WOS) Core Collection conducted in March 2018. The search was aimed at finding studies on people's attitudes towards meat consumption in relation to planetary health. For each subtopic (awareness, willingness, and change), a separate query was conducted. Strings and search terms used to retrieve relevant literature are detailed in Table 1.




**Table 1.** *Cont.*

A single and separate query with one term of each column was created. A total of 48 search queries for awareness were created. A total of 60 search queries for willingness were created, a total of 42 search queries for change were created.

Two authors conducted the screening process independently in order to reduce bias. For each subtopic, a three-step procedure was performed. First, titles, abstracts, and keywords were screened. Relevant articles on either awareness, willingness, and/or change were imported to Zotero reference manager. Second, saved references were read in full with a twofold purpose: (1) To make sure that they met the inclusion criteria (Figure 1) and (2) to identify in their cited references section new articles not yielded by the WOS search. Third, these first two steps were repeated until citation redundancy was achieved. When a reference was declared eligible by one reviewer only, the two reviewers reached an agreement on its inclusion or exclusion. Flow charts of this process can be found in Appendix A.

	-
	-


#### **Figure 1.** Eligibility criteria.

#### *2.2. The Meta-Aggregative Approach*

There are many methods for conducting qualitative evidence synthesis [21]. All of them entail various degrees of interpretation and aggregation, but in each method, either interpretation or

aggregation is more prominent than the other [21]. This review uses a meta-aggregative approach, a method "designed to model the Cochrane process of systematic reviews summarizing results of quantitative studies while being sensitive to the nature of qualitative research and its traditions" [21]. The focus of this approach is the practical consequences of the generalizations that can be drawn from the summary of common and competing findings. In other words, this method aims to produce the knowledge necessary to recommend actions in the field of study [21]. In the present case, our goal is to produce knowledge to make future diet-oriented public health intervention programs more effective, as well as to inform future research on consumers' attitudes towards meat intake reduction.

The process of meta-aggregation conducted followed, with minor variations, the Joanna Briggs Institute (JBI) methodology for systematic reviews of qualitative evidence [23]. It was conducted with the assistance of the JBI software called System for the Unified Management, Assessment, and Review of Information (SUMARI), which was designed to assist researchers and practitioners in conducting systematic reviews [24]. The JBI methodology includes developing a research question, conducting a literature search, doing a critical appraisal of selected studies, extracting the findings, categorizing them and, finally, generating synthesized findings [25]. The JBI SUMARI critical appraisal instrument evaluates the coherence between the theory, the methodology, the research question, and the representation of data and its interpretation, as well as identifying how the researchers' values and beliefs might influence the study. Research ethical procedures are also evaluated [23]. It was decided that, since a small number of studies were identified, all the studies would be included regardless of their quality. Even so, all selected studies were assessed by one of the authors (see Appendix B). The JBI protocol also recommends contacting authors of papers reviewed in order to ask for clarification when necessary. For this review, no further clarification was deemed necessary.

#### *2.3. Data Extraction*

Data extraction from the selected papers was carried out by two independent reviewers using the JBI-SUMARI extraction instrument [23] (see Appendix A). Extracted data included details about the population, context, geographical location, study methods, and the phenomena of interest relevant to the review questions. Findings and their illustrations were extracted by one reviewer only, since the meta-aggregative approach does not require two independent reviewers for this procedure. At this stage, the reviewer must stay as close as possible to the themes listed by the original authors and avoid interpretation of them [21]. A finding is defined as "a verbatim extract of the author's analytic interpretation of their results or data". An illustration is defined as "a direct quotation of a participant's voice, field-work observation or other supporting data from the paper" [23].

The JBI method requires the assignment of a credibility level for each finding depending on weather illustrations are provided in support of findings, and on how clearly an illustration supports a finding. Thus, "unequivocal" findings are clearly supported by the accompanying illustrations, "credible" findings come with an illustration lacking clear association with it, and "unsupported" findings do not provide supporting illustrations [23]. In the present review, however, only two levels of credibility are considered: "unequivocal" (U) for findings that come with illustrations and "credible" (C) for findings without illustrations. Two reasons justify this methodological decision. First, journal editors generally restrict the number of words per article, forcing authors to leave out illustration data. Secondly, while the JBI method allows the consideration of unpublished results [23], this review only considered published studies indexed by the Web of Science. Thus, it is assumed the papers included in this review went through a peer-review process that checked the credibility of the findings.

#### *2.4. Data Synthesis*

Qualitative research findings were pooled using JBI SUMARI with the meta-aggregation approach [23]. This involved the aggregation or synthesis of findings to generate a set of statements that represent that aggregation, through assembling the findings and categorizing these findings on the basis of similarity in meaning. These categories were then subjected to a synthesis in order to

produce a single comprehensive set of synthesized findings. These findings, then, should broaden our understanding of the relationship between environmental concerns and meat consumption and thus, inform future diet-oriented public health interventions and research efforts. Categories are defined as "a brief description of a key concept arising from the aggregation of two or more like findings and is accompanied by an explanatory statement that conveys the whole, inclusive meaning of a group of similar findings." [23]. A synthesized finding is "an overarching description of a group of categorized findings" [23]. This data synthesis process was conducted by two reviewers.

#### **3. Results**

#### *3.1. Studies Included and their Characteristics*

Following the inclusion criteria (Figure 1), a total of 10 studies were identified on either awareness of the meat impact on the planet, and/or willingness to stop or reduce meat consumption for environmental reasons, and/or consumers who have already altered their meat intake for ecological concerns. The number of studies identified and considered in this review is the minimum recommended to conduct a meta-synthesis [26].

The articles on awareness and/or willingness share a common phenomenon of interest, namely, people's perceptions of, and potential dietary response to the meat environmental impact. Their samples include a total of 328 participants, from teenagers to elders, defined as meat eaters. Very few vegetarians took part in a couple of studies. Reviewed papers on motivations that brought actual change focus only on vegetarians and vegans. We did not find any study that explored the reasons and motivations for consumers becoming flexitarians or simply reducing their meat consumption due to concerns for the environment. Four out of the five papers on dietary change explored the contexts and motivations for either becoming a vegetarian/vegan or keeping a meatless diet. The fifth article explored possible emotional relationships between vegetarianism and death. We considered it relevant because the research process included an inquiry on the reasons for adopting a vegetarian/vegan diet. These investigations on dietary change had a total of 301 vegetarian/vegan and former vegetarian participants aged from 14 to 85 years old. See Appendix A for the characteristics of the included studies.

Regarding methodology, a clear majority of the reviewed studies employed thematic analysis. Two of them worked with grounded theory. Focus groups were the most common way to obtain data. Three research projects interviewed or surveyed participants over the internet. The geographic context of the studies was very limited. All articles but one studied English-speaking populations mainly from the United Kingdom and Oceania. Only four articles mentioned some kind of covariates effects. Since reported covariates vary, or their effects are contradictory across studies, it was not possible to extract patterns of influence worthy of reporting. Overall, reviewed studies are of medium to high quality (see Appendix B).

#### *3.2. Awareness*

Figures A1 and A2 (see Appendix B) show the aggregative process from general findings extracted from four papers to synthesized findings on people's awareness of the environmental impact of meat production and consumption. Extracted findings and illustrations are reported in Appendix C.

Figure A1 presents the findings and categories that sustain the first synthetized finding on awareness: Consumers do not bear in mind that food in itself has an environmental dimension. Extracted findings show that consumers believe that unsustainable food has to do with food production and distribution activities like deforestation, pollution, transport, and excessive packaging [27–30], for example, but not with the type of food in itself, be it animal or vegetable. In fact, we found evidence that consumers are confused or simply ignorant when it comes to defining environmentally friendly foods [28]. This is consistent with findings about consumers not thinking of the environment when making food purchase decisions [27,28,30]. Therefore, we classified all these findings into three categories: (1) Consumers attribute food environmental impact to food systems activities only, (2) lack

of clarity among consumers on what ecological food is, and (3) in the same way that consumers consider price and healthiness, among others, food environmental impact is not a food property for them. These three categories have in common an understanding or conceptualization of food as detached from the environment.

If the first synthetized finding on awareness has to do with food in general, the second, presented in Figure A2, is exclusively about meat: Consumers' perceptions of meat consumption are a barrier for them to acknowledge or accept that reducing meat intake would alleviate negative impacts on the environment. In other words, consumers do not seem ready to fully integrate the idea that a sustainable diet must have little or no meat. Extracted findings show that people believe it to be irrelevant, or they disagree that reducing their own meat intake would significantly impact the environment [29]. Some consumers associate the meat environmental impact with cattle methane gas and deforestation only [27,29], making no link between meat consumption and climate change [27–30]. In general, they have a romanticized image or conception of meat, imagining animals freely grazing in the field and not in packed stables typical of intensive farming [28]. Still, some people are sensitive to animal welfare issues [28]. These findings were classified into three categories: (1) Disbelief that altering personal meat consumption can alleviate climate change, (2) meat environmental impact restricted to methane gas and deforestation, and (3) consumers have an overall good perception of meat.

#### *3.3. Willingness*

Four papers on people's willingness to reduce meat consumption because of the environment were identified and reviewed. Extracted findings and illustrations can be found in Appendix C. Figures A3–A6 (see Appendix B) show the aggregative process from general findings to a total of four synthesized findings.

The first synthesized finding is that environmental reasons and motives can be an enabler for reducing meat consumption to a certain extent. Extracted findings indicate that environmental concerns can prompt consumers to minor reductions in meat intake [31]. Some people, though, believe more scientific evidence is needed on the environmental impact of meat before they would make a dietary decision [29]. These findings were classified into two very similar categories summarized here in a sentence: Environmental concerns may motivate reductions in meat intake but not significantly alter meat-eating patterns.

The second synthesized finding is that sociocultural, culinary, and physiological reasons are barriers for consumers to reduce meat consumption on environmental concerns. Under the category "barriers to a large meat intake reduction," we grouped extracted findings that show that consumers argue social reasons, tradition, and lack of culinary skills to resist a substantial reduction of meat in their diets [30]. In fact, some consumers perceived a 70% meat reduction to be the same as becoming vegetarians, something they were not willing to do [30]. Extracted findings from papers that simply asked consumers about their willingness to reduce their meat intake without specifying the amount were grouped under the category "barriers to an indeterminate meat intake reduction". Again, sociocultural [27,29,31] and culinary [29,31] reasons were barriers. Dietary habits also prompt resistance to meat intake reduction [27,29,31]. Other findings have to do with physiological reasons [27,29,31], like achieving satiety and the pleasure of eating meat. These two barriers might be reinforced by the finding that consumers believe that a proper meal must have meat [30]. Consumers also may refuse to curtail meat consumption to mitigate climate change because they say that, driven by other reasons, they have already reduced their meat consumption [29]. Other consumers may resist a reduction in meat consumption for no reason whatsoever [27], or they choose to change other behaviors to help the environment instead of altering their intake of meat [29].

The third synthesized finding captures the ambiguous potential of health, nutrition, taste, and economic reasons to act as both enablers or barriers in consumers' willingness to reduce meat consumption because of the environment. Extracted findings show that some consumers may perceive a reduction of meat as healthy and nutritionally sound [29,31], while others as unhealthy

and nutritionally unsound [27,29–31]. In the same way, some consumers argued the lack of palatable alternatives to meat as a resistant factor to changing meat consumption [27,29–31], while others, faced with pictures of tasty looking vegetarian dishes by the research team, were inclined to eat less meat [31]. Extracted findings regarding the influence of economic reasons show that pricey meat prompts lower consumption of it [29–31]. However, consumers associate the lack of meat in their main daily meals with poverty [30].

The fourth synthetized finding is that consumers demand nutritional and culinary education in order to adopt a low or meatless diet to alleviate harm to the environment. People have difficulties imagining an alternative diet with low or no meat to their current dietary patterns [29–31]. Having understood the pressure of environmental concerns, they demand nutritional and culinary education to adopt a sustainable diet [30]. Consumers do not appreciate dietary interventions focused on what not to eat. They would rather listen to advice on how to lead a healthy and tasty low or meatless diet [30].

#### *3.4. Change*

A total of five articles were identified and reviewed regarding meat reduction or avoidance due to environmental concerns. All reviewed articles studied vegetarians and/or vegans. No qualitative evidence has been found regarding people who simply reduced their meat intake because of the environment. Extracted findings and illustrations can be found in Appendix C. Figures A7 and A8 (see Appendix B) show the aggregative process from general findings to a total of two synthesized findings.

The first synthesized finding states that environmental concerns can be both a trigger for adopting [32–34], and reinforcement for sustaining [34–36], a meatless diet. Even though only for a minority, there are consumers for whom environmental reasons were the initial motivation to become vegetarian. In some cases, environmental motivations prompted vegetarians to go vegan [35]. Other extracted findings were grouped under the category "environmental concerns as part of a non-ecological web of motivations to adopt or sustain a vegetarian diet". These findings show that care for the environment is another reason to justify a vegetarian diet already adopted because of health or ethical reasons, or one of many reasons to avoid animal products. One paper reported that some consumers argued that the environmental inefficiency of meat production contributed to world hunger, making the adoption of a vegetarian diet a human rights issue [36].

The second synthesized finding is that consumers involved in environmentalism or that simply love nature consider a meatless diet just another behavior to care for the environment. Extracted findings show that love for nature can prompt consumers to become vegetarian [33]. Other consumers see a meatless diet as a way to reinforce their commitment to life on Earth [34]. People may also adopt a vegetarian diet as a consequence of their environmentally friendly lifestyle [35]. However, one study on reasons for leaving vegetarianism showed that among ex-vegetarians, the majority had originally adopted a vegetarian diet because of the environment. It seems that with time, they opted for other ways to alleviate the environmental impacts, such as eating limited amounts of meat or only organic meat [32].

#### **4. Discussion**

In this qualitative evidence synthesis, we looked at three groups of behavior stages of change regarding meat consumption: (1) Awareness, which refers to subjects in the precontemplation stage, (2) willingness, which refers to individuals in the contemplation and preparation stages, and (3) change, which includes people in the action, maintenance, and termination stages [17]. Below, we discuss how the reviewed qualitative evidence complements the quantitative studies systematically reviewed on awareness, willingness, and change [19].

Quantitative research on awareness has shown that aware consumers are a minority, that they underestimate or ignore the potential of reducing meat consumption to mitigate climate change, and that consumers ignore that a meatless diet is more environmentally friendly than one including

meat [19]. Reviewed qualitative evidence reaffirms these findings, and synthesized findings add profundity to them. As long as consumers perceive food as detached from the environment, it is going to be hard for them to, first, make sense of the fact that food in itself, apart from transport and packaging, has an environmental impact, and, second, start distinguishing foods according to their environmental impact. This is clearly a barrier to increase awareness, and it may help explain why sustainability messages in favor of meat reduction are difficult for consumers to understand [37]. A second barrier is that consumers have an overall positive image of meat that does not seem to be affected by scientific evidence regarding the environmental degradation caused by meat production. In other words, the values consumers attach to meat trump the value of environmental protection and the value to mitigate climate change. All this indicates that the problem is not (only) a matter of knowledge but a matter of internal dispositions and mental frameworks that could make it hard for consumers to learn and accept scientific evidence regarding the meat environmental impact. De Boer and Aiking [37] have recently proposed several communication strategies based on the psychological theory of frames that may help bridge mental frameworks against meat intake reduction.

Quantitative studies on willingness have shown that consumers willing to reduce their meat intake to alleviate the environment are a minority. Among suggested strategies to curb climate change, meat curtailment is the least preferred [19]. The reviewed qualitative evidence reaffirms these findings and sheds further light. In general, there is a strong resistance to alter meat consumption because sociocultural, culinary, and physiological reasons trump environmental motives. This finding is consistent with the experience of actual meat-reducers. A qualitative study conducted in the UK demonstrated that meat reduction is conditioned by determinants beyond consumers' ethical stance towards the environment or animal welfare. Social pressure and understandings of nutrition constrain consumers' meat-reducing practices [38]. This might explain why consumers would rather adopt any other strategy to counter climate change before eating less meat: People have many strong reasons to continue eating meat. However, the reviewed qualitative evidence indicates that there are several important reasons, like health, nutrition, and taste, that can be allies of environmental motives in increasing consumers' willingness to reduce meat intake. Consumers who have an already positive image of low-meat diets may find in environmental protection a trigger to make a dietary change. This could mean that out of the aforementioned 4Ns, the Ns for necessity (need to eat meat to be healthy) and nice (meat tastes good) have the potential to become strong allies of meat intake reduction if consumers understand that planetary health is essentially linked to human health, and if consumers are helped to generate positive taste expectations for plant-based meals, as the results of this review show and another review suggested [39]. In this sense, probably the most useful finding is what could be consumers' fundamental reason for unwillingness to alter their meat consumption: Their acknowledged lack of nutritional and culinary knowledge that hinders them from being able to imagine an alternative way of eating with little or no meat. This barrier has also been identified by a review on "capability, opportunity, and motivation" to reduce meat intake and adopt plant-based diets [39], and by a review of influence factors on meat reduction [40]. Given the fact that consumers demand nutritional and culinary education, increasing willingness might be less a matter of strong reasons to resist meat intake reduction and more a matter of knowledge and practical skills to actually adopt an appealing low or meatless diet. Stoll-Kleemann and Schmidt [40] have also argued for the need to stop underestimating the importance of providing food-related skills (i.e., how to cook, know where to find affordable and tasty meat-free food) in getting consumers to reduce meat intake.

Quantitative evidence on change has shown that environmental concerns are the main reason to adopt a meatless diet for only a minority among the general population. The influence of environmental motives to drive change is more prevalent among meat-reducers or flexitarians than among vegetarians or vegans [19]. In light of quantitative and qualitative evidence on willingness, these results are a logical consequence. They are also consistent with the fact that the most prevalent reasons to become vegetarian [35,36,41–43] or reduce meat consumption [44] are health and animal welfare.

The reviewed qualitative evidence on change studied only vegetarians and vegans, and not meat-reducers or flexitarians. Still, it may prove useful to explain why environmental protection is a more prevalent primary driver among meat-reducers or flexitarians than among vegetarians and vegans when it comes to change. Qualitative evidence shows that for the majority of vegetarians and vegans, environmental concerns are a secondary or subsidiary reason to health or animal welfare motives. For them, the environment is less a motive to change and more a reason to sustain and further justify their meatless diet. In the case of flexitarians, the lack of qualitative evidence on change [38,45] makes it difficult to understand how environmental concerns influence their dietary choices. Based on current evidence on flexitarian awareness and willingness, however, we hypothesize that environmental concerns have a similar degree of influence in both vegetarians/vegans and flexitarians. In other words, we suggest that both groups of people value the environment arguably the same. The reason why environmental concerns are more prevalent drivers for change among flexitarians than vegetarians may have to do with attributed value to animal welfare and the understanding of what constitutes a healthy diet. While many vegetarians and vegans place a high value on animal welfare to the point of adopting a countercultural diet in the Western context, flexitarians would give a low value to such ethical concerns. Regarding health, vegetarians associate it to a meatless diet while flexitarians to a diet with meat, as the reviewed qualitative evidence and other quantitative evidence [46] has shown. This would support the idea that flexitarians, indifferent to animal welfare and convinced that a certain amount of meat is indispensable to meet human nutritional needs, may find in environmental protection the only driver to eat less meat. Both a review on the psychology of vegetarianism that also considered the literature on flexitarians, and another review on influence factors for reducing meat consumption reinforce this hypothesis [40,47].

A general recommendation for future research on awareness, willingness, and change is to address the geographical and cultural limitations of the reviewed studies. As noted above, they were conducted in North America, the United Kingdom, Australia, and New Zealand only. Since quantitative evidence has already shown that willingness is significantly higher in southern European countries than in northern ones [48], qualitative research is necessary in order to understand why that is the case. Moreover, since meat consumption is increasing in the low and medium-income countries of Asia and Latin America [1,6], qualitative research in these countries is urgent in order to stop such increase before meat consumption reaches the level of the high-income countries. Another limitation of the reviewed qualitative evidence is that cultural and ethnic factors were rarely taken into account in the research design and sample selection. Given that quantitative evidence has shown that there are differences in willingness across ethnicities inhabiting the same country [49], qualitative studies could further illuminate the role of different cultures in awareness, willingness, and change. Finally, since several reviewed studies are a few years old, and environmental concerns have received abundant media attention in the last two years, consumer attitudes towards meat consumption because of the environment might have changed.

#### *Recommendations*

The findings of this review suggest the following recommendations for future public health dietary interventions, and for future research on awareness, willingness, and change.

Public health interventions on awareness:

• Prepare consumers to understand that meat has an environmental impact by (1) informing that food, in general, has an environmental dimension, (2) addressing the positive image of meat consumption among consumers (strategies need to be found in order to persuade consumers that meat is not as good as they believe, (3) addressing skepticism towards the effectiveness of personal dietary change.

Scientific research on awareness:


Public health interventions on willingness:

• Consumers need to feel nutritionally safe and enjoy their meals. Nutritional and culinary education on meatless diets may increase consumers' willingness to reduce their meat consumption.

Scientific research on willingness:


Public health interventions on change:


Scientific research on change:

• Conduct qualitative investigations on meat-reducers and flexitarians.

## **5. Conclusions**

The few qualitative research studies on consumers' attitudes towards meat consumption in relation to planetary health has proven useful to deepen our understanding of people's awareness of environmental impact of meat consumption, of consumers' willingness to eat less meat to alleviate the negative impacts on the environment, and of dietary change motivated by environmental concerns. Awareness is hindered by certain consumers' beliefs on food, meat, and personal behavior. Thus, public health interventions aimed at increasing awareness should rely less on disseminating information and more on fact-based persuasive communication strategies. Willingness is easily hindered by different reasons and motivations. Therefore, satisfying consumer demands for nutritional and culinary education may significantly increase people's willingness to help the environment by reducing meat consumption. Change is especially understudied because there is no qualitative evidence on flexitarians. Environmental vegetarians and vegans need further education in order to be aware that giving up meat is not just another behavior to help the environment but the most impactful individual practice.

**Author Contributions:** Conceptualization, R.S.-S. and J.S.; basic data extraction and data synthesis, R.S.-S. and Y.B.-B.; writing—the original draft preparation, R.S.-S.; writing—review and editing, R.S.-S., Y.B.-B. and J.S.

**Funding:** This research received no external funding.

**Acknowledgments:** A special thanks to Joanna Briggs Institute for their technical support with this qualitative synthesis review.

**Conflicts of Interest:** The authors declare no conflict of interest.







**Table A2.** People willingness to stop or reduce meat consumption because of its environmental

 impact.




**TableA3.***Cont.*

ethnographic

adopting a vegetarian or

and prayer meetings).

affective and philosophical

reasons evoking and spirituality.

transcendence

vegan diet.

method.


**TableA4.**Criticalappraisalresults.

**Appendix B**


**Table A6.** People's awareness of the environmental impact of meat production and consumption.

**Appendix**

 **C. List of Study Findings with** 

**Illustrations**

**A7.**People'swillingnesstostoporreducemeatconsumptionbecauseofitsenvironmental


**TableA7.***Cont.*

**TableA7.***Cont.*




**Table A8.** *Cont.*

#### *Sustainability* **2019**, *11*, 6295

#### **Appendix E. Meta-aggregative Flowcharts**

**Figure A1.** Meta-aggregative Flowchart for the first synthesized finding on awareness.

**Figure A2.** Meta-aggregative Flowchart for the second synthesized finding on awareness.

*Sustainability* **2019**, *11*, 6295

**Figure A3.** Meta-aggregative Flowchart for the first synthesized finding on willingness.

**Figure A4.** Meta-aggregative Flowchart for the second synthesized finding on willingness.

**Figure A5.** Meta-aggregative Flowchart for the third synthesized finding on willingness.

**Figure A6.** Meta-aggregative Flowchart for the fourth synthesized finding on willingness.

**Figure A7.** Meta-aggregative Flowchart for the first synthesized finding on change.

**Figure A8.** Meta-aggregative Flowchart for the second synthesized finding on change.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Consumers Demand for Social Farming Products: An Analysis with Discrete Choice Experiments**

**Tiziano Tempesta 1, Daniel Vecchiato 1,\*, Federico Nassivera <sup>2</sup> Maria Bugatti <sup>1</sup> and Biancamaria Torquati <sup>3</sup>**


Received: 3 October 2019; Accepted: 19 November 2019; Published: 28 November 2019

**Abstract:** This paper analyses the demand for social farming (SF) products. In particular, we investigate the preferences of consumers who buy their products from large retailers, rather than from solidarity purchasing groups or other niche markets using a sample of 225 consumers. In this regard, a discrete choice experiment (DCE) was carried out to estimate the willingness to pay (WTP) a premium price for the purchase of a common product (i.e., eggs) from farms that employ disabled people. The attributes considered in our DCE design are the employment of disabled people and two additional attributes which may have ethical implications for the choices. The results indicate that consumers are interested in buying SF products, with about 74% of the sample willing to buy the eggs produced by social farms and the average WTP being equal to -1.36 for a pack of six eggs. Moreover, the average WTP for the use of labour of disabled people attribute amounted to -0.69 for a pack of six eggs.

**Keywords:** social farming; food; agriculture; consumer demand; disability; discrete choice experiment; eggs; sustainable agriculture

#### **1. Introduction**

In addition to the socio-economic contribution of agri-food production, societal expectations toward agriculture focus on its role in promoting and proposing activities aimed at bolstering environmental and landscape services such as water management, social care and cohesion [1]. The reason is that agriculture is still involved in the largest use of land by far [2] and that rural areas are increasingly shifting from being productive areas to what can be called consumptive areas [3] that need to perform social, recreational, and maintenance functions. Multifunctionality, as a core item in the common agricultural and rural development agenda, refers to the different functions that agriculture performs in society, functions that are not only oriented to the production of food and fibres, but also include, for instance, the creation of new job opportunities and the development of rural areas to attract other new customers of rural services [4]. These new opportunities for farms within the multifunctional paradigm are diversely detected, with the common target to reconsider their predominant orientation towards primary production and profit maximization and to rearrange the adoption of a more socially responsible pattern of production [5–9]. Therefore, multifunctionality can be the new unifying paradigm to bring post-modern agriculture in accord with the new societal demands [4]. Among the various multifunctional practices, social farming (SF) allows the farms to reach their scope of activities [10,11]. SF has appeared as a multifunctional innovative strategy [12] that contributes to social care through the production and processing of agricultural products by

incorporating direct social benefits in employment, training, and therapy or rehabilitating groups at risk of social emargination [13]. Following this logic, many European countries and regions have tried to promote SF among local communities to create and popularize innovative work practices [14].

The term "social farming" is also called care farming, farming for health, green care, and connective agriculture [15], and is used to describe farming activities when they are oriented toward the promotion of the rehabilitation and care of disadvantaged people and the integration of people with low job-holding capacity, i.e., people with psychophysical disabilities, convicts, drug addicts, minors, and immigrants. Besides rehabilitation and sheltered employment, other examples of SF services include therapy, lifelong education and other opportunities to contribute to enhance social inclusion [16–18]. SF is the result of a new, widespread positive perception of agricultural and rural resources, enhancing the interest in the beneficial effects of nature and agricultural activities on the social, physical, and mental wellbeing of people [19], thus linking elements of healthcare system to agriculture (social and therapeutic horticulture), landscape or nature conservation (eco-therapy), animal keeping (animal-assisted intervention), or animal husbandry (care farming) [20].

SF and its relevance is now emerging in Italy, as well as in other European Countries and other parts of the world [16,20–22]. It is paving the way for a new chance to diversify rural activities, to enhance the role of a renewed agriculture in society, and to strengthen the economic and social viability of farms and rural communities as a whole [10,23–26]. Although there is great interest for SF, social farms still suffer the need to find adequate funding [27]. Analysing social co-operatives, which represent one of the most relevant SF initiatives in Italy, Fazzi [17] pointed out their important involvement in the management of social health services outsourced and financed by local healthcare boards, and at the same time their need to have a profitable activity to reduce their risks of dependence on public funding only. Besides the public sector, new opportunities for social farms rely on, for example, the private demand for a social service, e.g., clients or client representatives that contact a care farm directly without connecting their needs to a public institution [27].

In recent years, a new phenomenon is spreading in many industrialised countries, where consumers reacted to the global standardisation and homologation of agricultural products [11] developing a special sensitivity to their ethical attributes. Such penomenon is the so-called critical consumption or the anti-consumerism movement. Among the ethical characteristics of products, "critical" consumers are very interested in issues such as environmental sustainability, social justice and inclusion, income distribution, economic diversification, and preservation of small and local firms [28]. The new needs of these consumers, not satisfied by the standard products, resulted in the development of new markets. These markets mainly rely on short value chains [29–32] or direct purchase from the producers. In this way the consumers can satisfy their need for local, typical, and very often organic products, characteristics that are a proxy for values such as sustainability, solidarity with small farmers, fair trade and personal health. The increasing interest about the ethical content of agricultural products, may also develop a notable opportunity for social farms, given that their products comply with the ethical requirements of these new consumers. In fact, the growing demand for these products suggests that the ethical content of the SF produce could be remunerated by the market [11,33].

Many studies highlighted that consumers are willing to pay a premium price for socially responsible products [34], namely, for products produced by companies that adopt production and organizational techniques capable of providing social benefits that are not incorporated in other goods, and therefore, do not bring any direct advantage to the buyer.

Analysing the studies that estimated the consumers' willingness to pay (WTP) for socially responsible products Tully and Winer [34] highlight that the social benefits usually considered in the literature concern the environment, animals, or other people. With regard to the benefits of other people, previous studies considered the WTP related to fair trade for products imported from tropical or sub-tropical countries such as coffee, chocolate, and bananas [34,35].

To our knowledge, only one study has analysed the demand and WTP for SF products [36]. However, in this study, people who buy food from alternative food networks (AFNs) were mainly

interviewed. In general, although these people currently constitute the reference market for SF in Italy, they have a greater sensitivity towards socially responsible products, thus being more attentive to the use of short supply chains and the consumption of organic and low environmental impact products. Nevertheless, we should notice that only a small fraction of Italian consumers buy food from AFNs, as the large majority of consumers usually buy food from supermarkets; thus, it would be useful to understand if and to what extent they are sensitive to the ethical and environmental aspects related to food production. In particular, the increase in the consumers' knowledge on SF could favour its diffusion in the future, increasing social benefits and at least partially reducing the costs borne by public structures for social inclusion of people with disabilities or other social disadvantages. Additionally, it should be noted that large-scale distribution has sometimes shown itself to be sensitive to the support of products with a high degree of ethical attributes, favouring and encouraging their sale.

The objective of this paper is to analyse the demand for SF products, but contrary to Torquati et al. [36] that focused on AFNs, we considered a sample of consumers who buy their products from large retailers. We want to test whether consumers not belonging to any specific niche market have a positive WTP for SF products. Therefore, a discrete choice experiment (DCE) was carried out to estimate the willingness to pay a premium price for the purchase of a common product (eggs) from farms that employ disabled people.

The attributes considered in our DCE design are the employment of disabled people and two additional attributes which may have ethical implications for the consumers' choices. The first is the possibility that the eggs are produced by organic farming, while the second considers the location of the producing farm, more precisely, if the producing farm is located near the place of purchase. The purchase of organic products can be promoted both by the benefits that can be obtained from it on a personal level (consumption of foods without chemical residues or with better organoleptic characteristics), and by the environmental benefits resulting from the adoption of more environmentally friendly production methods. The proximity of the production site enables the consumption of fresh products, and simultaneously, reduces the environmental impact of transporting goods.

The rest of the paper is organized as follows. Section 2 describes the methodology. In Section 3, we present the results, while in Section 4, we discuss the results and summarise the conclusions.

#### **2. Material and Methods**

#### *2.1. Questionnaire Structure*

To analyse the potential demand for eggs produced by companies that practice SF, we designed a questionnaire of four parts.

The first part presents the research and provides some information on SF. The interviewees were advised that during the interview they would be asked to indicate their preferences regarding the purchase of eggs produced by different methods, including the employment of disabled people, that is, companies that practice SF, organic farming respecting animal welfare, and production in farms located near or far from the place of purchase.

The second part was structured to collect information on the buying habits of the respondents. First, the interviewees were asked to indicate where they usually buy food (type of commercial structure and distance from home) and how often. Afterwards, the interviewees were asked to rate the importance of the following elements when buying food according to a five-point Likert scale (ranging from not relevant at all to highly relevant): the brand, price, organoleptic characteristics (taste, smell, etc.), produced by organic farming, quality certifications (protected designation of origin (PDO), protected geographical indication (PGI), etc.), processed products (products prepared or ready to use), health and hygienic safety, place of production (Italy or abroad), convenience (offers), and trust in the seller. Some questions were asked to verify the degree of SF knowledge and to identify which factors could favour the purchase of SF products. The third part included the DCE and the six choice sets that are described in the following paragraphs.

Finally, the socio-economic characteristics (gender, age, educational level, and employment position) were collected in the last part. Since people in Italy are very reluctant to provide data on family income, we asked them to judge their family standard of living which somehow provides a more accurate measure of the real well-being level of the family.

The questionnaire was initially tested on a small group of people to check the comprehensibility and duration before being administered to the interviewees. Following this preliminary analysis, changes were made to simplify some questions and provide a greater clarity in others.

#### *2.2. The DCE Methodology*

We applied DCE to estimate the consumers' preferences with regard to SF products. Given that the DCE approach is well known nowadays and widely applied in marketing, agribusiness, environmental valuation, health economics, and transportation studies, our presentation of DCE will not be exhaustive, providing only a broad overview of the methodology and introducing the reader to its specific terminology. We invite the readers not very familiar with such approach and interested in a deeper understanding to read Hensher et al. [37], Hauber et al. [38] or Ben-Akiva et al. [39]. The DCE is one of the stated preference methods used in economics [40], having its solid theoretical foundations in the Lancaster's consumer theory [41], where random utility models are applied [42,43]. The popularity of DCE is due to its ability to estimate both the value of a good/service as a whole and the implicit value of its attributes [44]. A further aspect that deserves attention is its applicability to ex-post and ex-ante valuations. In fact, given that the DCE presents a hypothetical scenario to respondents, it allows us to value future scenarios/products, thus performing ex-ante valuations.

Data are collected using either a paper-based or a digital questionnaire. The DCE goal is to analyse the respondents' preferences with regard to a hypothetical scenario/product. Respondents are presented with such hypothetical scenario and then requested to choose -often repeatedly- from a bundle of different options (such bundle is technically called the 'choice set') according to their preferences. The different choice options included in the choice set usually present the same good/service, characterised by a set of attributes, with the levels of such attributes varying in each choice option. For example, considering eggs, an attribute could be the 'egg size', and its levels could be 'small', 'medium', and 'big'.

A key aspect of the DCE preparation is the 'experimental design' [45], namely, the process that allows the researcher to build the choice sets that will be presented to respondents. The first phase of the 'experimental design' is the selection of the key attributes of the good/service presented to the respondents, with the levels of attributes being determined in this phase. To reduce all possible combinations of attributes and levels that will be presented to respondents, the researcher creates a subset of them using the 'experimental design' according to different statistical criteria (see Johnson et al. [45] for more details). The final subset of all potential choice sets is then included in the questionnaire and presented to the respondents, who are expected to choose their preferred alternative among those options presented in the choice sets. According to the random utility theory, each respondent assigns a certain utility to each choice option in the choice set and performs his/her choice maximizing his/her expected utility. Although it is not possible to measure the respondents' utility directly, the researcher can observe their choices. Thereby, the part-worth utilities of the attribute levels are estimated linking the probability of choice to the respondent utility function. Such estimates can be performed applying different models, including a popular model which is McFadden [46] conditional logit model, often referred to as the multinomial logit (MNL) model in the DCE literature. MNL models have been widely applied when studying the sample mean preferences, while the random parameter logit models (RPL) [47–49] and latent class models (LCM) [50] are applied to investigate heterogeneity across respondents. Although both models are suitable, they differ. RPL takes the preference heterogeneity into account in a continuous fashion, considering it random with a specific density function, whereas LCM can be considered a semiparametric variant of MNL [47], as the

probability of choosing an option is conditional on two aspects: the good characteristics and the individual belonging to a cluster of people with common preferences.

#### *2.3. Attributes Selection and Experimental Design*

As highlighted in the introduction, the purpose of the survey was to assess the consumers' WTP for a pack of six eggs produced using different production methods. In this regard, following Torquati et al. [36], four different attributes were considered:


The lower price level for the attribute price (-2.4) was chosen based on the average price of the eggs sold in the supermarkets in the area where the interviews were conducted. Since the total number of profiles (choice options) derived from all the possible combinations of attributes and levels was too high (2<sup>3</sup> × 31 = 24 profiles), through an experimental orthogonal design, 12 profiles were selected from which six choice sets were created, with each choice set being composed of two profiles and the opt-out (or no choice option). Figure 1 shows a choice set used in the DCE. Data were analysed using a RPL model [39,46,49], which has the advantage of considering the sample heterogeneity, treating it in a continuous fashion and allowing the estimation of the individual WTP for each attribute investigated. The RPL model examined, along with DCE attributes, both the socio-economic characteristics and motivational aspects of respondents for choosing the eggs. To measure the effect of individual characteristics on the demand for eggs, three interaction variables were included in the model: age over 60 years, high or medium-high living standards, and consumer attitudes with respect to the purchase of food products, with the latter obtained through a cluster analysis.


**Figure 1.** A choice card presented in the choice experiment.

The interviewees were characterised as belonging into two clusters by means of k-means cluster analysis performed on the scores given to the seven elements considered when buying food. The k-means cluster analysis was estimated using the SPSS statistical package, version 25. A dummy variable referring to the respondents that belong to the second cluster was interacted with the eggs' attributes in the utility function used in the RPL model. The following utility function (Equation (1)) was used to estimate the model:

$$\begin{split} \mathcal{U}(\mathbf{x}\_{i}) &= \sum \mathcal{B}\_{i} \cdot A\_{i} + \sum \mathcal{B}\_{A\_{i} \cdot A \cdot \mathbf{G}E} \cdot A\_{i} \cdot A \mathbf{G}E + \sum \mathcal{B}\_{A\_{i} \cdot HST} \cdot A\_{i} \cdot HST \\ &+ \sum \mathcal{B}\_{A\_{i} \cdot CLIST} \cdot A\_{i} \cdot CLIST + \mathcal{B}\_{\text{price}} \cdot PRICE + \mathcal{B}\_{\text{OPT}OIT} \cdot OPTUIT \end{split} \tag{1}$$

where *A* represents all attributes except price, *AGE* is a dummy variable assuming the value 1 if the respondents have an age greater than 60 years, *HST* is a dummy variable assuming the value 1 if the respondents have a standard of living high or very high, *CLUST* is a dummy variable assuming the value 1 if the respondents belong to cluster number 2, *PRICE* is a continuous variable for the attribute price, and *OPTOUT* is a dummy variable that assumes value 1 when the choice option is 'Not Buying' (see Figure 1). All independent variables in the utility function, except for *PRICE* and *OPTOUT*, were effect coded. Using Equation (2), it is possible to estimate the average WTP for each attribute level as follows:

$$\text{WTP}\_{\text{i}} = -\frac{2\beta\_{i}}{\beta\_{price}}\tag{2}$$

Looking at Equation (2), *β<sup>i</sup>* was multiplied by 2 to take into account that the variables in the model were effect coded [51].

Data were analysed with the NLogit version 6 software. We applied RPL models to take into consideration preference heterogeneity. RPL models were estimated using 1000 Halton draws and the random parameters were assumed as normally distributed, incorporating into the model only interaction terms that had a significant result after an exploratory analysis. To derive the demand function it is important to consider the relationship between the premium price and the quantity sold. This enables producers to define a marketing strategy for their product, given that they know the market share that their product will have at each price level. To derive such information we analysed respondents heterogeneity by means of the RPL model, and in particular their individual WTP for the attribute levels considered in the DCE. From the individual WTPs we derived their complementary cumulative frequency distribution. The latter can be considered a proxy for the market demand function, under the assumption that respondents state their WTP for a single product unit. To evaluate the complementary cumulative frequency distribution, the following logistic function was estimated for each attribute (Equation (3)):

$$MarketShare = \frac{1}{\frac{1}{u} + b\_0 \cdot b\_1^{Price}}\tag{3}$$

where *u* is the upper bound of the market share (equal to 1), and *Price* corresponds to the individual WTP.

Once known, the complementary cumulative frequency distribution can help producers in optimising their marketing strategies choosing the price and quantity that maximise their profit, in accordance with their market power.

#### *2.4. Data Collection*

Data collection was carried out in the period from 10 October to 15 November 2017. A total of 225 questionnaires were collected by direct interviewing at the entrance or at the supermarkets of the municipality of Brescia, Italy. We could thus analyse the behaviour of the population as a whole and not of particular segments of demand. In fact, in Italy, people usually buy food at large retail chains. To avoid distortions, the interviews were carried out every day of the week, at different times of the day and in front of different supermarkets in the Brescia municipality to collect a highly representative sample of the consumers.

#### **3. Results**

#### *3.1. Socio-Economic Characteristics of Respondents*

The average age of respondents was 55 years, being higher than the national average (44 years), since generally the younger ones do not do the shopping. In fact, respondents younger than 20 years of age were only 5.3% of the sample, and those with an age between 21 and 30 years were 10.7% (Table 1), whereas 44.0% were over 60 years of age. Regarding gender distribution, the sample was not distributed in a balanced way between males and females, with females constituting 71.6% of the sample. This imbalance is because women usually deal with food shopping in most Italian families.


**Table 1.** Socio-economic characteristics of the sample.

With respect to the educational qualification, 9.8% of the interviewees had an elementary school certificate or no educational qualifications, 17.8% had a lower middle school certificate, 43.1% had a high school diploma, and 29.3% had a college degree. Compared to the average Italian situation, in the sample, people with a diploma were more likely to be represented than those with only a primary school certificate. Nevertheless, it should be noted that in urban and metropolitan areas, the population has higher educational qualifications than the average of the country.

Regarding the employment status, 42.7% said they were employed, 4.9% were unemployed, 9.3% were housewives, 4.4% were students, and 38.2% were retired.

Most respondents lived alone (27.6%) or with another person (35.1%), while a minority belonged to families with four or more members (20.4%). This datum could be associated with age, given that single-person families often constitute elderly people who live alone.

As noted above, considering the difficulty encountered in Italy in collecting information on the family income of respondents, to have at least an indication of the economic status of the sample, people were asked to assess their standard of living. Most respondents (56.0%) considered it medium-high, while 34% considered it low-medium, 7.5% low, and only 2.2% high (Table 1). Therefore, in the majority of cases, the interviewees considered themselves quite well-off; however, it can be assumed that for psychological and privacy reasons, there has been a tendency not to declare themselves belonging to either of the two extreme categories.

#### *3.2. Buying Habits of Respondents*

The majority of respondents (94.7%) shopped in supermarkets, 16.9% in stores, and 11.6% in discount stores, with the percentage of those attending local markets (8.4%) or members of solidarity purchasing groups (5.8%) being much lower (Table 2). This last figure is confirmed by the fact that only 1.6% stated that they are members of consumer associations. In most cases, the place where the shopping is done is less than 1 km from home (68.0%), while only 6.2% buy food in shops located more than 5 km from home (Table 2). Since the interviewees live in an urban area, it can be assumed that, due to their buying habits, they could hardly consider the opportunity to buy food directly at the place of production. In these cases, the purchase of social agriculture products could take place only if they were also offered for sale in large-scale distribution or local markets located in urban areas.


**Table 2.** Interviewees buying habits.

\* Multiple choice question.

Considering the purpose of this study, it is beneficial to understand the behaviour of the interviewees regarding the purchase of eggs. As reported in Table 2, 43.6% of respondents buy eggs once a week, 31.6% once every two weeks, and only 4.0% do not buy eggs regularly (less than once a month), indicating that eggs are a common ingredient of the respondents' diet.

67.6% of respondents said they buy organic products; however, this does not mean that they consume exclusively organic products, but that these products are bought occasionally. This is confirmed by the fact that 53.8% stated that they sometimes bought organic eggs. This observation is important because it suggests that the interviewees are aware of the good under investigation. Although 120 respondents (53.3%) stated that they had already heard about SF, this knowledge seems to relate more to the existence of the phenomenon rather than to the real meaning of SF in normative terms. In other words, we can assume that the knowledge of SF is quite superficial because it is based mainly on hearsay and not on direct experience. In fact, with reference to the Italian context, in a study conducted by Nassivera et al. [52], only 22% of respondents stated that they had an adequate knowledge of SF. With reference to the factors most considered during food shopping, we can see (Table 3) that the greatest importance is attributed to health and hygienic safety (4.04), followed by sales discount (3.70), Italian production (3.66), organoleptic characteristics (3.60), and price (3.45). On the contrary, processed food (2.02), the brand (2.52), and organic production (2.57) are of little importance. This last datum somehow contradicts the statements concerning the purchase of organic products, but is derived to a large extent from the presence of two categories of interviewees within the sample who attribute a markedly different importance to organic food.

Starting from the scores given by the interviewees to the factors considered in the purchase of food, using a cluster analysis, the interviewees were divided into two clusters with quite distinct characteristics (Table 4). Only in the case of the purchase of processed products, the members of the two groups seemed to have similar preferences. Conversely, in all other cases, average scores differed at least with a 95% probability. To get a more precise idea of the greatest differences between the two groups, the Cohen's d coefficient was calculated.


**Table 3.** Average scores of the factors affecting the food purchase by the interviewees.

† Based on a five point Likert scale ranging from 1 (not relevant at all) to 5 (highly relevant). \* Conf. Int. = Confidence Interval. \*\* Inf. = Inferior limit; Sup. = Superior limit.


**Table 4.** Average scores of the factors affecting the food purchase of the interviewees belonging to Cluster 1 and Cluster 2.

\* Sign. = Significance.

Based on the values assumed by Cohen's d, it should be noted that people belonging to the first cluster (N = 104) when doing food shopping give a much higher importance to organic farming products, the presence of labels or quality cues that certify the origin of the product (PDO, PGI, etc.), Italian production, and health security. However, they tend to attach less importance to price. Conversely, the second cluster (N = 121) includes subjects who are less attentive to the problems connected to the environmental impact of production techniques or have less confidence in quality certification marks. However, they pay more attention to the price and the presence of sales discount. The members of the two clusters differ in a statistically significant way (Pearson's Chi-squared probability < 0.05) with respect to some socio-economic and behavioural characteristics. The members of the first cluster have a higher educational qualification, belong to a greater extent to cultural or environmental associations, buy more organic products or fair trade certificated, declare themselves to a greater extent to be aware of the meaning of SF, and have a higher frequency of shopping in specialized shops or farm markets. To understand which factors could increase the sales of SF, the interviewees were asked to indicate which elements, not actually present, could make them consider in the future the possibility of buying SF products. Most respondents (56.0%) stated that more information about SF could help them consider buying these products, 52.9% believed that a greater availability of products would help the purchase, while for 42.7%, price policies would be important, namely, it would be useful if the SF products had a price similar to others. Finally, 33.8% underlined that the certification of such products could favour their purchase, guaranteeing consumers the origin of the product. With regard to the attention paid to the ethical aspects of consumption, it is interesting to note that 45.7% stated that they had purchased fair trade products in the past.

#### *3.3. The DCE Results*

The DCE estimated model has a good interpretative capacity (McFadden's Pseudo R-squared: 0.50) according to the standards of these models [37]. The attributes considered in the model have a significant degree of heterogeneity, indicating that the interviewees attribute different degrees of importance to these characteristics of the product under analysis. All estimated parameters (Table 5) proved significant at the 95% level, except for the coefficient of the interaction term: disabled workers × belonging to cluster 2 (*p* = 0.051).

The WTPs of the coefficients not interacted correspond to the premium price that the interviewees who have a low standard of living, belong to cluster 1, and are aged less than 60 years, are willing to pay. The members of this group were on average willing to pay -0.83 more for the eggs produced in farms that also employ disabled people, -0.50 for products made near the place of purchase, and -1.10 for organic eggs. Age presents a positive correlation only for the social agriculture attribute level (disabled workers), while it does not seem to have any effect on the WTP for organic and local products. The WTP of people who declared a medium-high or high standard of living, which constitute 58.2% of the sample, was considerably higher for all the attributes considered. Further, their average WTP for eggs produced by SF was -1.71, for local products -0.85, and for organic products -2.00. The average WTP of the members of the second cluster for each of these characteristics was always very low and ranged from -0.10 to -0.16.

The effect of age is significant only for the eggs produced by the employment of disabled people. People over 60 were willing to spend -1.63 more on average than younger people. Considering the sample as a whole, the average WTP for eggs produced by SF is equal to 1.36 -/pack of six eggs (95% confidence interval (CI) is 1.09-–1.63-); for organic productions, it is 1.13 -/pack of six eggs (95% CI is 0.93-–1.33-); and for local productions, it is equal to 0.49 -/pack of six eggs (95% CI is 0.45–0.53). Considering the average price of -2.4 for a pack of 6 eggs (free-range poultry farming) in the survey area, the premium price would tend to be 56% for eggs from SF, 47% for organic eggs, and 20% for local products. These percentages are comparable to those obtained by Loke et al. [53] in a study that used the hedonic price method to analyse the factors affecting the price of eggs in Hawaii (organic + 64%; local + 40%).

However, it should be noted that the standard deviation of the estimated coefficients is much higher for eggs produced by SF (s.d. = 2.70) than for the organic products (s.d. = 2.03) and local production (s.d. = 0.47). This highlights that for the SF products, there is a greater uncertainty among consumers, probably, because they are not familiar with them.

The analysis of individual WTPs estimated with the RPL model (Figure 2) shows that the WTP complementary cumulative frequency distribution function for eggs produced using disabled labour and organic farming is very different from that of products produced near the site of purchase. First, the percentage of those who are willing to pay a minimum premium price for organic or SF products is considerably lower relative to the place of production. Furthermore, 26% of respondents are not willing to pay any amount for SF products, with 23% having a negative WTP for organic products. The percentage of those who are not willing to pay more for products produced near the place of purchase is just over 8%. It can also be seen that the WTP complementary cumulative frequency distribution function for this attribute has a much greater slope. From Figure 2, it can be observed that the percentage of people willing to pay a premium price higher than -0.30 for a pack of six eggs is always greater for both organic and SF products compared to the proximity of the place of production. For example, while more than half of respondents would be willing to pay 1 more for eggs produced by SF and for organic products, this percentage drops to just over 6% in the case of local products.



† Significance levels: \*\*\* significant at the 99% level; \*\* significant at the 95% level; \* significant at the 90% level. ‡ Random parameters were assumed to be normally distributed.

**Figure 2.** WTP complementary cumulative frequency distribution functions.

#### **4. Discussion and Conclusions**

Following De Pelsmacker et al. [54] and Doane [55], ethical consumerism can be defined as "the purchase of a product that concerns a certain ethical issue (human rights, labour conditions, animal well-being, environment, etc.) and is chosen freely by an individual consumer" [54]. According to Zander and Hamm [56], the ethical attributes considered in the purchase of food can be numerous, but, in general, they can be summarized as follows: animal welfare, good working conditions for farm workers, support for family farms, preservation of local cultural landscape, regional and local production, protection of biodiversity, fair prices to farmers, no air freighting, production in care farms, revival of traditional processing methods, and fair product prices.

Numerous studies have been carried out in the past to analyse the consumer demand and WTP for these attributes; in particular, the studies in the fields of fair trade (fair prices and good working conditions) [35], organic production [57,58], animal welfare [59], and local production [60,61]. However, only a few studies have dealt with eggs [36,62–64] or SF products [36,52,56], with only one study estimating the WTP for food products produced by social farms.

Our results highlight that consumers are interested in buying SF products, with about 74% of the sample willing to buy the eggs produced by social farms with an average WTP equal to -1.36 (+56% considering an average price of -2.4) for a pack of six eggs. Torquati et al. [36] carried out a DCE to analyse the WTP of members of solidarity purchase groups in the case of eggs and zucchini produced by SF. With regard to eggs, two attributes were considered similar to those present in our survey, namely, the use of labour from disabled people and the proximity of the place of production to the consumption market. The average WTP for the use of labour from disabled people attribute amounted to -0.69 for a pack of six eggs, while the WTP for purchasing products produced near the place of purchase was -0.51. These numbers are in some respect comparable to those estimated in this survey. Thus, the average WTP of the consumers interviewed is similar to that of the members of the solidarity purchase groups considered by Torquati et al. [36].

For a correct interpretation of our results, we have to consider the possibility that other ethical attributes may influence consumer behaviour. In this regard, Zander and Hamm [56] examined the effect of additional ethical attributes of organic food on the consumers' purchase decisions. Using an information-display-matrix approach, they found that "being produced in a care farm" is one of the ethical attributes less considered when buying organic products. Conversely, the most important attributes were animal welfare, regional production, and setting fair prices to farmers. However, it should be emphasized that, in Italy, organic farms must implement animal welfare

policies; therefore, the organic attribute largely covers animal welfare. Our findings indicate that the respondents attribute the same importance to animal welfare and SF.

Our study has shown the presence of a remarkable variability in the attitudes of the interviewees regarding all three attributes considered. This variability is related to some socio-economic characteristics of the respondents (age and standard of living) and the factors generally considered in the purchase of food products. On the one hand, a higher standard of living corresponds to a higher WTP for all attributes considered; on the other hand, belonging to cluster 2 reduces it. It should be remembered that the most important factor considered by members of cluster 2 when purchasing food is the price of the product, which tends to assume a higher importance than its organoleptic characteristics. For this group, the presence of labels that are somehow a proxy for the quality of the product (PDO, PGI, or organic) is of a little importance. Therefore, we can conclude that for about half of the interviewees the ethical factors play a marginal role in addressing the purchase choices of food products.

Our research also indicates the presence of a considerable segmentation in the demand for eggs with regard to some ethical attributes (organic and animal welfare). Similar results were found in previous literature; for example, Gerini et al. [62], with reference to Norway, who have shown that only consumers who purchase organic products with high frequency are willing to pay a premium price for organic eggs or those produced in compliance with animal welfare standards. Mesias et al. [65] in a study carried out in Spain identified the presence of four groups of consumers for whom the breeding method has a distinctly different effect with regard to the propensity to purchase eggs. Andersen [63] pointed out that the WTP for organic eggs is greater for people with higher incomes or who live in urban areas.

Even in the case of other foods and other ethical attributes, numerous studies have highlighted the presence of many factors that determine the market segmentation. For example, Feldmann and Hamm [60] reviewed the studies that analysed the effect of the place of production on food demand, finding that the importance of the place of production (origin) depends on demographics, knowledge, context, attitudes, and behaviour. With reference to demographics, old wealthy people, living in rural areas have more supportive attitudes towards local food. Attitudes and opinions on local productions also seem to be particularly important. Hemmerling et al. [58] pointed out that there are numerous factors that can influence WTP for organic production, with many studies demonstrating that the WTP for organic products increases with income and educational level, whereas the age and family size seem to have an inverse effect.

Our results indicate that the SF products could be bought by a fair number of consumers, even if the actual size of the market segment may depend on numerous factors, not explicitly tested in our study and that should be taken into consideration in future research.

First, many studies in the past have pointed out a considerable discrepancy between the purchase intentions expressed by consumers and their real behaviour [66]. The presence of this intention-behaviour gap may depend on both individual factors and organizational and structural factors [67]. Thus, we must first emphasise that consumers are often guided by their habits in purchasing decisions. Second, consumers are much more attentive to the price in real life than in the case of surveys that refer to purely hypothetical situations. In addition, taste plays an important role in the real consumption of food, as it can significantly change the importance of credence attributes in purchasing decisions [68]. Therefore, ultimately, the demand analysis based on declared preferences may overestimate the actual WTP for people and their actual propensity to purchase [69].

Second, we should note the considerable discrepancy between the quantity of ethical products placed on the market and their potential demand. The low availability of these products on the shelves means that they are generally not considered in the consumer-shopping basket, which most of the time ignores their existence. For example, in a fair trade coffee experiment, Hainmueller et al. [70] observed that the addition of the fair trade label increased sales by 15%. Hence, it should not be overlooked that the SF production constitutes a credence attribute, and therefore, a certification by an independent third party is necessary to attest the characteristics of the product. In the absence of such certification or in the presence of a lack of knowledge of certification procedures by consumers, they may be reluctant to purchase SF and other ethical products.

In conclusion, despite the limitations mentioned above, it can be assumed that, similar to other ethical attributes, social farms' production can be an attribute appreciated by consumers. Nevertheless, according to the interviewees opinions, two factors emerge as critical in ensuring the future development of SF, namely, an adequate certification system and a proper communication of the added value of SF products.

**Author Contributions:** T.T. and B.T. conceived the study; T.T. oversaw the data collection; M.B. data collection; D.V. and T.T. developed the DCE experimental design; F.N. contributed Section 1; T.T. contributed Section 2.1; D.V. contributed Sections 2.2, 2.3 and 2.4; T.T. contributed Sections 3 and 4; writing–review and editing, D.V.; supervision T.T. and B.T.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **Abbreviations**

The following abbreviations are used in this manuscript:


#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Sustainability in the Beverage Industry: A Research Agenda from the Demand Side**

**Carla Rodriguez-Sanchez and Ricardo Sellers-Rubio \***

Department of Marketing, Faculty of Economics, University of Alicante, Crta San Vicente del Raspeig, San Vicente del Raspeig, 03690 Alicante, Spain; carla.rodriguez@ua.es

**\*** Correspondence: ricardo.sellers@ua.es

**Abstract:** Sustainability has become one of the most important challenges for the beverage industry over the last few decades. In fact, many producers have implemented environmental, social, or economic aspects of sustainability at several stages of their production process. One of the reasons that might explain this interest in sustainability is that consumers are changing their behavior to integrate sustainable and environmental considerations into their purchase behavior. Accordingly, some consumers' purchasing decisions are based not only on how well products satisfy their needs but also on how these products affect the environment or society at large. Within this context, designing appropriate interventions to fostering sustainable consumption requires deeper knowledge about its underlying determinants. In this paper, we focus on some of the most important challenges that might drive future research within this area.

**Keywords:** sustainability; beverages; consumer

#### **1. Introduction**

The most accepted concept of sustainability defines it through the three overlapping principles of environmentally sound, economically feasible, and socially equitable production. Sustainable production comprises business practices that are sensitive to the environment (environmentally sound), responsive to the needs and interests of society at large (socially equitable), and economically feasible to implement and maintain (economically feasible) [1]. However, and despite its popularity, this term is frequently associated by producers and consumers with the environmental aspects of production only, neglecting other important issues, and every producer might have a different understanding of this term.

Within the beverage industry, sustainability can be considered as a strategy to differentiate firms or products in order to meet some market segment demands. In fact, many firms claim socially or environmentally friendly orientations when producing and marketing beverages, integrating sustainability into their communication strategy to reinforce their brand and market positioning. Presently, consumers not only make decisions based on how well products satisfy their needs but also how these products influence society at large. Consequently, many consumers have integrated sustainable and environmental considerations into their lifestyle choices. Through the adoption of sustainable practices, firms could obtain a competitive advantage and increase sales with clear product differentiation.

However, sustainability is also a needed strategy to guarantee the future development of the beverage sector. According to [2], most environmental consumption impacts are related to a few product categories. In fact, 70–80% of total impacts from domestic consumption relate to food and drink, housing (e.g., domestic energy use), and transport (e.g., leisure and holiday travel). Particularly, food and drink consumption have a significant eco-footprint due to, among other factors, the use of land, energy, water, and chemicals in production as well as pollution during the distribution system [3]. In this sense, the

**Citation:** Rodriguez-Sanchez, C.; Sellers-Rubio, R. Sustainability in the Beverage Industry: A Research Agenda from the Demand Side. *Sustainability* **2021**, *13*, 186. https:// dx.doi.org/10.3390/su13010186

Received: 25 November 2020 Accepted: 23 December 2020 Published: 28 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/).

challenge of sustainable development is to meet current demands without ruining future generations and the planet in the long term [4].

Accordingly, academic research has paid special attention to sustainable consumption in this industry, which may encompass a plethora of behaviors from consumer purchase of eco-friendly products to water use at home [5,6]. In the same line, [4,7] run systematic reviews aimed to analyze articles dealing with new trends in sustainable consumption practices. Most of the studies focus their attention on the food industry [8,9], while few studies analyze the beverage category, with the only exception of wine, which has received great interest among researchers [10]. Furthermore, as beer is the most consumed alcoholic beverage in the world [11], several authors have focused their attention on this beverage, identifying the rise of some trends (e.g., craft beer) regarding sustainability. For example, consumers' increasing attention to health-issues impulses breweries to expand their assortment to introduce non-alcoholic drinks in the market [11]. Moreover, craft beer consumption has been perceived by consumers as more sustainable [12]. In the same vein, some authors have focused their attention on organic food and beverages, given the increasing interest of consumers in this category [13]. Indeed, according to the forecasts from some international market research studies [14], the organic beverage market is estimated to reach \$181.78 billion by 2023 from \$99.76 billion in 2018.

Furthermore, designing appropriate interventions to fostering sustainable consumption requires deeper knowledge about the underlying determinants of consumer behavior [15]. The analysis of the drivers influencing sustainable consumption has been studied over the last decades from different disciplines such as environmental psychology [16] or (social) marketing [17]. Besides, the analysis of sustainable consumption from the demand side has been almost exclusively focused on the environmental dimension of the concept of sustainability. While environmental psychology has mainly focused on studying proenvironmental behaviors such as water/energy saving or recycling, it has been in the field of marketing where green purchase behavior has played a prominent role. Focusing on understanding the determinants of consumers' green purchase decisions, social marketers, public managers, and policymakers may be able to influence consumers' environmental mindset and behavior in the long term [18]. In fact, the beverage industry has incorporated several strategies to achieve a cost-effective transition to a circular economy and to combat climate-change-related risks to their operations. In this sense, [19] identified four types of managerial practices related to Circular Economy (CE) adoption at the product level that could be applied in the beverage industry: (i) Energy efficiency and usage of renewable sources of energy; (ii) Product and process optimization for resource efficiency; (iii) Product design for circularity; and (iv) Exploitation of waste as a resource.

In any case, sustainable beverage purchase behavior is still not well understood in academia, especially if we consider the concept of sustainable consumption as a whole [17]. In this paper, we focus on some of the challenges that may drive future research on this area. Particularly, we focus our attention on alcoholic (wine and beer) and nonalcoholic beverages (water, milk, and juices), as they represent a huge market share among sustainable beverages.

#### **2. From Green to Sustainable Consumption**

When an individual makes a purchase decision there is a possibility that it contributes to a more or less sustainable consumption pattern, since all purchasing actions have ethical, waste, resource, and community impact implications. Green purchase behavior refers to "purchasing and consuming products that are benign toward the environment" [20] (p. 190), which might contribute to a sustainable consumption pattern [21]. Green purchase behavior is seen as a complex form of consumption both intellectually and morally as well as in practice. As [22] points out, when individuals consider buying in an environmentally responsible way, they face an increasingly complex decision-making process. On the one hand, this type of behavior involves difficult motivational conflicts because there is usually a conflict between the self-interested benefits of consumers and environmental protection

related to collective goals. On the other hand, there are normally external barriers to green purchase, which arise from the political, economic, or cultural circumstances in the market and society [22]. For example, eco-friendly products are frequently seen as an expensive option and making sustainable consumer choices involves a fair amount of effort such as garnering information [23]. In addition, many consumers do not consider environmental aspects as a primary element when they make a purchase decision. In such situations, only when they perceive that the product achieves a certain level of product performance, "green" characteristics of the product may determine product choice [18]. In addition, research has shown that there is now a "typical" green consumer that is involved with environmental issues in all sorts of purchasing situations, especially when it comes to different contexts/environments [24] and/or (high/low) purchase product involvement [25]. Consequently, designing strategies to achieve a behavioral change toward sustainability becomes a challenging task.

However, as [17] highlights, sustainable consumption implies not only choosing "green" products but also reducing consumption and considering the full consumption cycle. While prior research in the beverage sector has mainly focused on understanding consumers' initial product choices, far less attention has been given to the reduction of consumption and the full consumption cycle (e.g., product usage or disposal), due presumably to the fact that reusing or recycling available products has lower environmental benefits than changing purchasing behavior [26].

Regarding the reduction of consumption from an environmental point of view, only a few studies about bottled water usage and intentions to reduce its consumption have been carried out so far [27,28]. Even though bottled water consumption may be considered in the beverage sector one of the products that are contributing the most to several environmental problems such as pollution, water wastage, and climate change [27]. In addition, regarding the full consumption cycle, it should be also taken into account that the beverage product life cycle is short. In this way, while for durable goods (e.g., household appliances, clothes ... ) consumers' post choice behaviors such as product usage (e.g., product maintenance) or product life extension (e.g., re-use) have an important impact on the sustainability of consumption, for commodity goods, disposal (e.g., recycling) is crucial. For example, [18] finds a high correlation between eco-friendly purchase and disposal decisions, being the environmental awareness of consumers a predominant predictor of disposal decisions. In any case, the available academic studies on beverage disposal are scarce to date and further research may shed light on this regard.

Given the imperative of sustainable consumption in the beverage sector, it is critical to understand what factors contribute to this type of behavior. Most of the studies focused on explaining sustainable consumption in this sector have analyzed these determinants independently on the beverage category (e.g., juice, wine, beer, milk, and so on) since the factors that influence purchasing decisions of each category may be different and therefore also the role of environmental issues in the purchase choice. For example, as [10] point out, consumer's wine choice is more complex than other products because it is one of the most differentiated products on the market. Wine consumers must deal with a large amount of information on wine labels (e.g., country of origin, region of products, grape variety, or brand) that are important drivers for purchase decisions. Besides, due to sustainable wine production (i.e., wine with sustainability characteristics), current certifications indicate new characteristics for consumers to be evaluated, making the purchasing decision even more complex. Furthermore, compared to other alcoholic beverages, wine is perceived as being more natural, reflecting a sense of agricultural seasonality and linked closely to a rural way of life [13], which are elements highly linked to the concept of sustainability. Hence, research to date has mainly focused on identifying how sustainability characteristics influence wine purchase decisions and other beverages have been virtually neglected. Nevertheless, it must be pointed out that studies on sustainable beer consumption have significantly increased over the last years [29,30]. In fact, beer is the most consumed

alcoholic beverage in the world [11], and sustainability might constitute an important trigger to foster beer consumption.

#### **3. Research Trends on Sustainability in the Beverage Industry**

Several studies have analyzed the factors influencing the purchase of sustainable beverages from different perspectives: (i) the willingness to pay a premium price for sustainable products; (ii) the packaging of sustainable products; (iii) the distribution of sustainable products and fair trade; and (iv) Factors influencing the purchase of sustainable beverages.

#### *3.1. Willingnes-to-Pay for Sustainable Beverages*

Most of the studies to date have mainly addressed sustainable beverage consumption analyzing intentions of willingness to pay (WTP) premiums for organic products. Accordingly, a recent review of the literature on this topic concludes that further work needs to be done in a real market context using, for example, household panel data [10]. This has relevance for studies on environmental issues since it is well known that there is usually an inconsistency or gap between intention to purchase eco-friendly products and actual purchase behavior [31]. In addition, regarding the characteristics of sustainable products, literature has focused primarily on production as a key element when consumers make their green purchase decisions. Thus, there has been a great deal of research lately on consumer preference and WTP for organic wine [13]; organic beer [29]; organic milk [32], or organic fruit juice. Importantly, it seems that there is still some confusion among consumers with the "organic" term and, in some cases, they consider organic products not to be environmentally friendly [23]. Most consumers associate "organic" only with being chemical-free and they are unfamiliar with organic farming standards and practices. Therefore, it comes as no surprise that personal health has been shown as a more important driver for organic food and beverage purchase by consumers than concerns for the environment [33].

#### *3.2. Sustainable Packaging*

Packaging is of special concern to some type of beverages such as juice, water, or beer, as they are the main beverage contributions to the packaging fraction [34]. However, as far as we know, only very few studies have addressed packaging as an important element for green purchasing decisions in the beverage sector [18]. From a consumer perspective, the use of packaging in commodity products transcends beyond its functional role (e.g., conserving the quality and freshness). In fact, consumers also consider the design, the image, and ease of packaging to correctly identify the product [35]. In this sense, [36] examines the factors that influence consumers' product purchasing behavior and their recycling behavior with respect to sustainable packaging. Their results evidence that variables such as gender, environmental awareness, or attitude towards green purchasing are factors that differentiate consumers that consider environmentally friendly packaging important and consumers that consider packaging as unimportant when making product purchase decisions. Therefore, environmental strategies should intend to make the most of this product attribute, and research in the beverage sector should be focused on which features of sustainable packaging are the most preferred by consumers rather than how this packaging is disposed of.

#### *3.3. Distribution and Fair Trade*

Regarding the distribution of beverages, increased attention has been given recently to local products as local food and drinks shoppers are driven by environmental (short supply chains) as well as community (economic and social) concerns [37]. A paradox between organic vs. local products may arise at this point as [23] wonders: "what is better for the planet buying organic quinoa grown in Peru or a non-organic alternative produced locally? My guess, would be the latter". Although further research is needed to understand consumer's opinion in this regard, the current discussion suggests that when

organic food is not produced locally, it loses authenticity. Thus, for many consumers, food miles rather than organic labels are the representation of sustainability [38]. However, it is worth mentioning that local food production seems to be more related to the social (e.g., the creation of a consumer-farmer relationship based on trust) and the economic (e.g., economic benefits for regional economies) dimensions of sustainability than the environmental one [39]. Even some studies [40] raise doubts about local food to be better for the environment than non-local food. Either way, only a few studies have been carried out to date on consumer perception about locally produced products, with the exception of wine [41,42].

Besides, buying fairly traded products is another way to consume sustainably. Fair Trade is defined as: "an initiative for small farmers and wage workers in the South, who have been restrained in their economical and/or social development by the conditions of trade ('disadvantaged')" [43] (p. 2). So, this type of trade relations focuses on the social and economic dimensions of sustainability. Fairtrade beverage purchasing is growing in the last years. For example, according to [44] data, Fairtrade alcohol was the biggest growth of all the monitored sectors in the UK, with shoppers spending 34% more on this over 2017. However, academic studies focus on consumers' perceptions; preferences and willingness-to-pay for Fairtrade drinks are almost non-existent. Future research should therefore gain a deeper understanding of consumers' attitudes and their buying habits regarding this type of product.

#### *3.4. Determining Factors of Sustainable Consumption*

Several studies have shown that the purchase of sustainable beverages (or green purchase behavior) can be influenced by a plethora of factors. These factors can be categorized into the following underlying causes [16]: (i) personal capabilities, (ii) psychological determinants (e.g., attitudes, beliefs or norms), and (iii) contextual forces and habits. The following explanation, however, should be treated with caution since, as aforementioned, studies usually ask for purchase intentions (e.g., WTP) in unrealistic scenarios and results can be overestimated regarding actual behavior. Furthermore, wine has been the main subject of previous studies, so it cannot always be extrapolated to other beverage products.

#### 3.4.1. Personal Capabilities

In general, previous studies [42,45,46] have shown that the typical organic drink consumer is female, highly educated, upper-middle-class income, and lives in urban areas. In the case of purchasing local wine, it seems to be different. Ref. [47] found that men paid more for wine with a lower distance traveled than women in Germany, and [42] shows that people living in a rural area were more prone to buy local wine in the US. Regarding age, studies revealed mixed results. For example, while [48,49] find that younger consumers had higher attitudes toward sustainable wine, [46,50] find older people to have higher WTP and probability for buying wine with sustainability characteristics respectively. Furthermore, some other studies have not found any differences in terms of sociodemographic factors for intention to buy organic milk [51] or for WTP for organic wine [41]. In summary, it seems there are conflicting results regarding the impact of personal capabilities on sustainable beverages purchase. This highlights the need for further research.

Furthermore, it should be highlighted that sensory acceptance constitutes one of the main choice criteria for consumers [52], which is forcefully reliant on cultural backgrounds, as well as previous sensory exposure to a specific food product [53]. Besides, the market for functional, natural, and non-alcoholic beverages is steadily increasing all over the world [54], because of the increasing consumer awareness of the importance of healthy nutrition.

#### 3.4.2. Psychological Determinants

A well-established theoretical framework that has often been applied in food and drink studies to analyze the psychological determinants of sustainable consumption is the Theory of Planned Behavior [55]. According to this theory, the most immediate predictor of behavior is an intention to engage in the behavior (i.e., a motivation or plan) and intentions are, in turn, predicted by attitudes (i.e., mental disposition and feeling about the environment), perceived behavioral control (i.e., ability to perform the behavior) and subjective norms (i.e., social pressure). Focusing on attitudes, studies on wine with sustainability features and organic milk reveal a close relation between attitudes and purchase behavior [56,57]. However, food organic research has also shown that although individuals declare to have high positive attitudes toward the environment or organic products, it does not necessarily translate into actual purchase behavior [58]. In the environmental literature, this is generally known as the "attitude-behavior gap" and it has been found not only for green purchase behaviors but also for other pro-environmental behaviors [59]. Among the reasons that might explain this inconsistency, environmental researchers suggest the use of constellations of behaviors (i.e., index), the overestimation of attitudes in survey studies due to social desirability, the measurement of general environmental attitudes instead of specific attitudes to the behavior and the intensity of attitudes [11,49]. Future studies in the beverage sector should consider these aspects.

Another theory widely used in organic food and drink literature has been the Norm Activation Model (NAM) [60]. This model states that pro-environmental behavior (e.g., purchasing organic beverages) depends mainly on the activation of personal norms, which reflect feelings of moral obligation to behave in a certain manner. For example, [61] find that people with strong personal ecological norms used "organic production" and the "EU-BIO-Label" as additional criteria during their milk decision purchase. Moreover, as an extension of the NAM model, [16] proposed the Value-Belief-Norm theory of environmentalism (VBN), which states that pro-environmental personal norms are influenced by values (i.e., general goals that serve as guiding principles in life) and beliefs. As [10] review points out, environmental values and beliefs on environmental protection are important motivators for buying sustainable wine. Additionally, values reflecting the need for living a hedonistic life have been also found to be an important precursor of belief systems that influence the purchase of organic wines [62]. Apart from the variables included in these models, other psychological factors such as trust, knowledge, or self-identity have been used in past studies to explain organic beverage purchases. Trusts in the retailer selling [63] and in winery [64] are considered important factors influencing consumers' behavioral intentions to purchase organic wine. The same result was found for organic milk purchase behavior when trust in farmers is analyzed [53]. In this last study conducted in Italy, it was also found that self-identity as a green consumer influenced consumers' intentions to buy organic milk. Furthermore, studies reveal that knowledge can play a significant role in motivating the purchase of organic food [37]. However, while consumer environmental knowledge seems to influence the willingness to buy environmentally friendly wines [65], greater knowledge about wine culture seems to be related to a lower willingness to pay a premium value [66].

#### 3.4.3. Contextual Factors and Habits

Although the above psychological factors play an important role in understanding sustainable purchase decisions in the beverage sector, contextual factors and habits might constraint or incentive this relationship. Among the contextual factors studied in the past, the price has been observed to be one of the main characteristics of the product when a consumer buys organic wine or milk [13,49,67]. The influence of price on buying behavior may be positive when high prices signal quality and status for consumers, or negative when high prices mean that a sacrifice must be made [68]. In this sense, research has shown that price is perceived as a barrier toward organic milk purchase, especially for consumers with budget constraints, and consumers often overestimate the price level of this type of milk when there is a lack of knowledge about it [68]. In the case of sustainable wine, some studies reveal that the higher the increase in price, the lower the probability of purchase [41]. Although, other studies [69] show that medium-high priced wines with

sustainability characteristics are preferred over lower-priced wines when information is given to the consumer through organic certification labels. In addition, the origin of wine and availability (close to consumers' homes) are other contextual factors moderating the transformation of attitudes into purchase behavior [10]. The taste of wine or milk is also considered an important motivator during the decision-making process [61,70]. In the case of milk, interestingly, those consumers that think organic products are intensively tasted are less willing to pay a premium for organic milk [69]. Finally, being stuck in a routine (habits) of purchasing behavior seems to also play a major role in sustainable beverage purchasing decisions for both wine and milk [50,71].

As can be seen above, previous work has been limited to analyzing cognitive and contextual factors and there is virtually no research focused on the affective component of attitude and emotions related to sustainable beverage consumption. Notwithstanding the existence of authors suggesting that these factors may play an important role as drivers of involvement in organic food purchase [72]. Future studies on the emotion topic are therefore required in the sustainable beverage sector.

#### **4. Conclusions**

The goal of this paper has been to review some of the papers that address sustainability in the beverage sector from the demand side. A deeper understanding of the drivers that determine sustainable consumption is needed to foster sustainable practices in the beverage industry. Although several managerial practices have been identified in the beverage industry in the context of the circular economy [19], it is mandatory to identify the responsibility for sustainable practices among consumers and firms (e.g., packaging recycling).

Particularly, future research should gain a deeper understanding of consumers' attitudes and their buying habits regarding sustainable beverages. From personal characteristics to the effect of emotions, attitudes, and lifestyles, more research is needed to understand the drivers that motivate consumers to buy sustainable beverages. Furthermore, it would be very interesting to know not only how these variables drive consumers' behavior but also whether there are differences among different beverages within the industry, as the consumer might act differently depending on the product. Finally, previous work has been almost limited to the environmental dimension of sustainability to explain sustainable consumption in the beverage sector. Future research therefore should be more focused on the social and economic dimensions of sustainability for improving regional economies.

**Author Contributions:** C.R.-S. and R.S.-R. have contributed equally to the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Determinants of Consumers' Retention and Subjective Well-Being: A Sustainable Farmers' Market Perspective**

## **Bi-Kun Tsai**

Graduate Institute of Bio-Industry Management, National Chung Hsing University, Taichung City 40227, Taiwan; pktsai@dragon.nchu.edu.tw; Tel.: +886-4-2284-0849 (ext. 622)

Received: 14 October 2019; Accepted: 12 November 2019; Published: 14 November 2019

**Abstract:** Farmers' markets have received much attention in many countries, and the amount of research on farmers' markets is gradually increasing. The consumption process of consumers at farmers' markets include both economic and social aspects, but most past studies have only focused on a single aspect. The economic perspective mainly focuses on transaction issues such as purchase motives, quality, satisfaction, purchase behavior, and post-purchase behavior, whereas the social perspective focuses on the social relations and psychological feelings created when consumers go to markets. This study aimed to integrate the economic and social perspectives and analyze the relationships among product performance evaluation, relational capital, repurchase intention, and subjective well-being of consumers at farmers' markets after their purchase experiences. I chose three recurrent farmers' markets in Taiwan, obtained 358 valid samples, and performed structural equation modelling analysis. The results indicated that the economic product performance exerted a significant and positive influence on repurchase intention, but its influence on subjective well-being was not significant. In contrast, the social relational capital was found to be a positive and significant factor of both repurchase intention and subjective well-being. On the whole, relational capital is more important than product performance. The suggestions for practice were as follows. First, farmers' markets have economic and social value and are thus worth being promoted by government agencies. Second, the managers of farmers' markets should implement a set of management mechanisms to ensure product performance and also create a market atmosphere that facilitates social interactions between farmers and consumers.

**Keywords:** farmers' market; product performance; relational capital; repurchase intention; subjective well-being

#### **1. Introduction**

Under the influence of global agri-food systems, local production for local consumption has become an alternative route of agricultural development aside from mainstream agriculture in many countries [1], its methods including community-supported agriculture, farm-to-school programs, culinary tourism, and farmers' markets. Local production for local consumption aims to lower transportation costs. Farmers produce and sell their products themselves, reducing food miles and carbon footprints, thus being friendly to the environment. In this way, consumers can obtain fresh and safe local agricultural products. Shortening the distance between farmers and consumers also enables them to establish face-to-face channels of interaction and build mutual trust, which can consolidate interpersonal interactions and institutional norms and strengthen ties between consumers and local agriculture. Farmers' markets, which have advantages in site conditions and local products and possess community relations from the place of production to the dining table, are said to be crucial platforms in implementing local production for local consumption and have the ability to

create agricultural production-sales models that take both environmental protection and friendly relationships into account.

As an alternative method of agricultural development, farmers' markets are based on biodiversity, environmental protection, food system localization [2], reduction of food miles, and the concept of smallholding. Their focus is to provide farmers and consumers with the opportunity to communicate face to face and to offer fresh agricultural products containing local culture. Although researchers have yet to reach a consensus on whether localized food systems are more environmentally friendly than globalized or modernized food systems [3,4], studies have demonstrated that via re-spatialization and re-socialization [5,6], farmers' markets can reduce food miles and carbon footprints and thus contribute to environmental sustainability [7–9].

With the rising trends of alternative agriculture and the "real food" revolution [10], farmers' markets have proliferated in Northern America, Western Europe, and Asia, and they are aspiring to set themselves apart from highly professionalized, resource-intensive global agri-food systems, which modern supermarkets are representative of. Farmers' markets have actually been around for a long time. They have often formed at locations in cities or towns in the form of retail stores or stands and enabled farmers and consumers to make face-to-face transactions and get to know each other. The increase of modern, transnational chain supermarkets around the world have caused these traditional farmers' markets to gradually disappear. Gradually, consumers, who are at the end of the food supply chain, no longer knew where their food came from or what value concepts and local culture were held within the production processes of the food. Consequently, the social ties and mutual trust that were originally embedded within the production and consumption of local food faded away. The recent regeneration of farmers' markets represents a shift in consumer focus in three aspects: from the price and quantity of agricultural products to their value and quality, from synthetic food to authentic food, and from conspicuous consumption to conscious consumption [11].

Cook [12] asserted that consumption is closely connected to leisure. Although the farmers' markets in different countries may operate differently, most of them are not limited to face-to-face buyer-seller relationships in which the farmers sell directly to the consumers. Rather, they attach more value to the interactions and interpersonal connections between the farmers and the consumers. In these interaction processes, the farmers can personally explain the unique qualities of their agricultural products, the environment they are produced in, and the ways they can be eaten. Thus, consumer participation in farmers' markets include consumption-oriented economic interactions and leisure-oriented social interactions [13]. The former involves the economic perspective of the consumer purchase behavior at farmers' markets, which mostly focuses on three aspects. The first is purchase motives and preferences, such as freshness [12,14], quality [15,16], taste [17], price, and food safety [18,19]; the second is purchase behaviors and repurchase behaviors [20,21], and the spillover effects of consumption at the markets on local economy [22–24]. The latter meaning "leisure-oriented social interaction" refers to the social perspective of the purchase consumer behavior at farmers' markets. Less research has been done on this topic. Some studies have investigated issues associated with the social interactions between consumers and farmers or other consumers at farmers' markets, such as trust and social capital [11,25,26]; another study examined the influence of consumer participation in the market on family relationships, quality of life, and subjective well-being [27].

In short, the majority of past studies on the purchase consumer behavior in farmers' markets have focused on the economic perspective and discussed market transaction elements such as purchase motives, quality, satisfaction, purchase behavior, and post-purchase behavior. Less attention has been given to the social perspective, which includes the social relations and psychological feelings created when consumers go to markets and communicate face-to-face with farmers or encounter other consumers. Social capital can be divided into two forms: relational capital, which is the private good that individual members within an organization can invest in, accumulate, and use at their own discretion; and system capital, which is the collective good that the members within an organization invest in and accumulate together [28]. This study aims to integrate the two aforementioned perspectives ("consumption-oriented economic interactions" and "leisure-oriented social interactions") and combine two economic constructs, namely, the market performance evaluation and repurchase intentions of consumers, and two social constructs, namely, the relational capital formed by consumers at farmers' markets and their subjective well-being, into a single research framework. In conclusion, the consumption processes at farmers' markets contain the characteristics of both purchase experiences and social experiences. In practice, more attention is often paid to purchase experiences, whereas social experiences are overlooked. Past studies involving farmers' markets have mostly focused on purchase motives, purchase behavior, market segmentation, and marketing strategies as well as on the prepurchase and post-purchase stages. The contribution of this study is that it extends research on farmers' markets to post-purchase behavior and psychological feelings, integrates economic and social perspectives, compiles relevant literature, and structurally analyzes the relationships among four crucial concepts: product performance evaluation, relational capital, repurchase intention, and subjective well-being of consumers. Specifically, the objectives included the following:


This study makes the following contributions to research on consumers at farmers' markets. First, most of the past studies on purchase behavior at farmers' markets have focused on market segmentation or the purchase motives, preferences, and purchase intentions of consumers [16]. This study adopted product performance, a construct associated with the evaluation of product quality in farmers' markets, as the antecedent variable of consumer repurchase intention and subjective well-being to further academic understanding in this aspect. Next, despite the many studies regarding the influence of leisure activities on subjective well-being [29–31], research has yet to be done on the influence of consumption activities at farmers' markets on the subjective well-being of consumers. Finally, this study integrated the economic and social perspectives, examined the correlation between product performance and relational capital at farmers' markets, and investigated the influences of these two constructs on repurchase intention and subjective well-being. This paper is a pioneer in this respect.

#### **2. Literature Review and Hypothesis Development**

A farmers' market can be described as a recurrent market at a fixed location where farm products are sold by farmers themselves directly to the public [25,32]. Most of the vendors at farmers' markets are smallholders themselves and sell their products at farmers' markets to supplement their income, and the markets will often lay down criteria for farmers hoping to participate. The promotion of farmers' markets brings several substantial benefits, such as (1) reducing the distance between farmers and consumers so that they can build trust as well as steady production and supply; (2) decreasing food miles, marketing margins, and quality loss; (3) re-familiarizing the public with local agricultural products and boosting the local economy; and (4) increasing local food supply and the overall domestic food self-sufficiency rate. In addition to the benefits above, farmers' markets have various functions. One is serving as cultural attractions for agrileisure [27], and another is serving as incubators to provide training for agricultural entrepreneurs and develop networks and social ties that help farmers participating in the markets to improve product quality, marketing, and management effectiveness. Another function of farmers' markets is that they promote social well-being. They enable direct, face-to-face interactions, which enhances the social embeddedness between consumers and farmers. This social embeddedness improves local identity, community solidarity, family life satisfaction, and personal subjective well-being [13,33–35]. The purpose of this framework is to determine the

influence of the performance evaluation given by consumers to products at farmers' markets and the relational capital that they accumulate during social interactions on their repurchase intention and subjective well-being.

#### *2.1. Product Performance*

Consumption at a farmers' market can be divided into three stages: prepurchase, encounter, and post-purchase [36]. The prepurchase stage includes the following steps: awareness of need, information search, evaluation of alternatives, and purchase decision. In the encounter stage, where the consumer actually goes to the market to make purchases, the consumer can communicate and interact face to face with the farmers and other consumers. During this process, the consumer can get to know the production philosophies and agricultural products of the farmers and accumulate social capital via the social interactions in the process at the same time. In the post-purchase stage, the consumer evaluates service/product performance, which in turn impacts their future intentions/expectations, including word-of-mouth, repurchase intention, and subjective well-being [37,38]. The evaluation of product performance begins in the prepurchase stage, when the consumer begins to form expectations of the quality, safety, and superiority of agricultural products at the farmers' market. Later, the consumer then compares the expectations with the actual feelings of using/eating the agricultural products. Thus, in meaning, the evaluation of product performance is similar to product satisfaction. Product performance is therefore a crucial factor of repurchase intention and even subjective well-being.

#### *2.2. Relational Capital*

Based on the social embeddedness theory, Hinrichs [39] indicated that economic interactions and social interactions exist simultaneously within farmers' markets. The process of consumer transactions socially embedded with the interactions of farmers and physically embedded with the contexts of farmers' markets and local communities. Consumers generally go to farmers' markets with their friends and family [40]. At the same time, they have social interactions with farmers or other consumers while they make purchases at the market. Thus, relational capital forms during the purchase process. In this study, I define relational capital as "the valued number of resources an actor can employ and use through direct or indirect personal relations with other actors who control those resources and in which the actor is intentionally investing and which should eventually pay off" [28]. During the purchase encounters at farmers' markets, consumers engage in purchase behavior and economic interactions based on purchase motives, product information, and their experiences with certain sellers in the past.

Lyson et al. [41] indicated that from a neoclassical perspective, the economic meaning of farmers' markets may not be very important. However, from the community perspective, farmers' markets are venues where consumers can obtain high-quality agricultural products while establishing relational capital. They stated that " ... [farmers' markets] can nurture local economic development, maintain diversity and quality in products, and provide opportunities for producers and consumers to come together to solidify bonds". Offer [42] pointed out that the reciprocal transactions between buyers and sellers at farmers' markets are "preferred when trade involves a personal interaction, and when goods and services are unique, expensive or have many dimensions of quality". Hunt [15] advocated that a correlation exists between the social interactions that consumers engage in at farmers' markets and the quality of the agricultural products that they purchase. He pointed out that the social interaction and information-sharing functions of farmers' markets enable consumers and producers to influence each other. Producers can get to know what quality requirements consumers have of agricultural products and improve their production methods accordingly, while consumers can acquire more information and purchase agricultural products of higher quality. Based on the above, consumers then form evaluations of product performance in the post-purchase stage. At the same time, consumers also have discussions with the friends and family accompanying them during the

purchase process as well as form and accumulate relational capital at the farmers' market from their social interactions with farmers [43]. Accordingly, the first hypothesis of this study is as follows:

**H1.** *A positive and significant correlation exists between the product performance evaluation and relational capital of consumers at farmers' markets.*

#### *2.3. Repurchase Intention*

I view repurchase intention as the probability that a consumer will revisit a seller or repurchase a product based on previous purchase behavior or experience in a farmers' market and after considering various factors [44,45]. Several marketing studies have demonstrated that the quality evaluation and satisfaction with regard to purchased products are antecedents of many loyal customer behaviors, such as repeat purchase, positive word-of-mouth, and the propensity to buy more. Thus, the evaluations that consumers make of product quality at farmers' markets are an important factor influencing their repurchase intention. Accordingly, I put forward the following hypothesis:

### **H2a.** *The product performance evaluation of consumers at farmers' markets exerts a positive impact on their repurchase intention.*

In a study on customers at a farmers' market in Italy, Cassia et al. [46] discovered that the factors influencing customer satisfaction include tangible factors, such as product quality, comparative prices, and convenience, and the social capital factors that exist among customers, farmers, and territory. Research [47] has also found that the perceived social embeddedness of customers at an organic farmers' market exerted a positive influence on their repurchase intention. Conceptually speaking, both social capital and social embeddedness are closely associated with relational capital. Accordingly, I put forward the following hypothesis:

**H2b.** *The relational capital of consumers at farmers' markets exerts a positive impact on their repurchase intention.*

## *2.4. Subjective Well-Being*

The experience, feeling, and understanding of well-being varies from person to person. Subjective well-being can be defined as a type of positive mental experience that an individual forms towards his or her own existence and progress and that is generated from the collective effects of factors such as objective conditions and satisfaction of needs [48]; it includes cognitive and affective aspects. The cognitive aspect equates subjective well-being with life satisfaction, and it is an overall evaluation based on the standards that an individual holds towards his or her quality of life. The affective aspect regards subjective well-being as the positive and negative evaluations that an individual has towards his or her current happiness and sense of value. As can be seen, well-being is a comprehensive concept combining an individual's life satisfaction, happiness, and sense of value. Similar concepts include satisfaction with life, quality of life, and happiness [49].

Numerous factors can influence subjective well-being, such as personality, self-efficacy, social relations, socioeconomic status, work, consumption, and leisure [50]. The purchase experiences at farmers' markets, which contain both consumption and leisure, are naturally a factor as well. Marketing research has shown that the primary task of salespeople is to convey superior value to customers so as to maintain and improve the well-being of customers and society [51]. Based on this definition, consumers should form favorable evaluations of product performance from the marketing activities of farmers at markets, which in turn influences their subjective well-being. However, health research [52] has indicated that personality, health, interpersonal network, and various domain satisfactions can explain 70–80% of subjective well-being. Still, literature regarding tourism has pointed out that the overall customer satisfaction of tourists (similar to the evaluation of tourism product performance) wields a positive influence on subjective well-being [53]. Based on the arguments above, I believe that the product performance perceived by consumers after making their purchases

at a farmers' market exerts a positive impact on their subjective well-being. I thus formulated the following hypothesis:

**H3a.** *The product performance evaluation of consumers at farmers' markets exerts a positive impact on their subjective well-being.*

Relatively more research exists on the influence of relational capital on subjective well-being. Studies show that some concepts similar to relational capital in meaning, such as social factors [46], social capital [54,55], and interpersonal trust [56], have an impact on subjective well-being. More specifically, social capital such as marriage, family, friends, and neighbors can offer individuals with timely and important social support, thereby bringing them joy, a sense of belonging, and self-esteem, which in turn enhances subjective well-being. Similarly, when consumers go to farmers' markets, the relational capital formed from their interactions with the friends and family accompanying them and their face-to-face interactions with the farmers will increase their subjective well-being. Accordingly, I put forward the following hypothesis:

**H3b.** *The relational capital of consumers at farmers' markets exerts a positive impact on their subjective well-being*.

Figure 1 exhibits the conceptual model of this study, presenting the relationships among the four constructs, namely, product performance, relational capital, repurchase intention, and subjective well-being, and the hypotheses that will be tested.

**Figure 1.** Proposed conceptual model.

#### **3. Methodology**

#### *3.1. Construct Measurement*

A structured questionnaire was developed to collect data, including four demographic variables, namely, participant gender, age, educational background, and average monthly household income and the four potential constructs, namely, product performance, relational capital, repurchase intention, and subjective well-being. Gender included male and female. Age was originally divided into six groups: under 20 years old, 21~30 years old, 31~40 years old, 41~50 years old, 51~60 years old, and over 61 years old. However, after compiling the statistics, I rearranged them into three groups: under 40 years old, 41~60 years old, and over 61 years old. Educational background was originally divided into four groups: junior high school or below, senior/vocational high school, university/junior college, and graduate school or above. However, after compiling the statistics, I rearranged them into three groups: senior/vocational high school or below, university/junior college, and graduate school or above. Average monthly household income was originally divided into nine groups: NTD 20,000 or lower, NTD 20,001~40,000, NTD 40,001~60,000, NTD 60,001~80,000, NTD 80,001~100,000, NTD 100,001~120,000, NTD 120,001~140,000, NTD 140,001~160,000, and NTD 160,001 or higher. However, after compiling the statistics, I rearranged them into four groups: NTD 40,000 or lower, NTD 40,001~80,000, NTD 80,001~120,000, and NTD 120,001 or higher.

In addition to the demographic variables of the participants such as gender, age, educational background, and average monthly household income, the potential constructs of this study all contained several indicators, each calculated using a single question item measured on a five-point Likert scale. The responses ranging from strong disagreement to strong agreement were given 1 to 5 points, respectively. After referring to the literature [57,58], I divided product performance into three variables, namely, product quality, product safety, and product superiority. Relational capital was divided into four variables, namely, face-to-face interaction, communication and understanding, learning experience, and development of friendship [59,60]. Repurchase intention was divided into three variables: repurchasing frequency, repurchase quantity, and repurchase priority [21,43]. Subjective well-being was divided into four variables: life satisfaction, happiness, fulfillment, and self-reported health [56,61]. The question items regarding product performance were worded in the negative direction and reverse scored. All of the question items in the questionnaire were reviewed and revised by three experts familiar with farmers' markets to ensure content validity. Table 1 presents the potential constructs, indicators, and their question items:


**Table 1.** Potential constructs, indicators, and question items.

#### *3.2. Data Collection*

I first prepared a structured questionnaire and then chose three recurring farmers' markets in Taiwan that had been around for a long time. Two sold environmentally friendly agricultural products, whereas the other sold organic agricultural products. All three markets are managed by a market committee consisting of member vendors who ensure that all agricultural products sold at the market are produced, processed, and marketed by the farmers themselves and that all the agricultural products pass safety inspections and organic produce verification. I adopted convenience sampling of consumers with purchase experiences at these three markets. As convenience sampling was employed, the analysis results may not be representative of the entire population [62]. The interviewer confirmed that the participants were return customers who had purchased products at the market in the past before administering the questionnaire. The survey was conducted from February to April in 2016. A total of 394 questionnaires were distributed. After eliminating incomplete questionnaires, I obtained a total of 358 valid questionnaires, representing a valid recovery rate of 89.5%. Most of the valid samples were from women (65.9%), and over half of the participants were between the ages of 41 and 60 (54.2%). The highest level of education attained by most of the participants was university or junior college (60.3%), and in monthly household income, the largest group was between NTD 40,000 and NTD 80,000 (37.7%). Table 2 shows the sample characteristics in detail.


**Table 2.** Sample characteristics.

#### **4. Results**

#### *4.1. Analysis of Descriptive Statistics*

Table 3 presents the descriptive statistics of different participant groups with regard to the four potential constructs. On the whole, relational capital received the highest score (mean = 4.23, SD = 0.62) and was followed by repurchase intention (mean = 4.12, SD = 0.71), subjective well-being (mean = 3.91, SD = 0.59), and product performance (mean = 2.75, SD = 0.74). Looking at the individual latent variables, I found that female participants expressed higher product performance evaluations, relational capital, and repurchase intention than male participants after their purchase experiences, and they also perceived greater subjective well-being than male participants on the whole. Older participants displayed both higher product performance evaluations and repurchase intentions than younger participants. Participants between the ages of 41 and 60 presented the highest relational capital, and those under the age of 40 showed poorer subjective well-being. Participants with higher educational backgrounds expressed poorer product performance evaluations as well as higher relational capital, repurchase intention, and subjective well-being. Participants with higher monthly household income presented higher product performance evaluations, relational capital, and subjective well-being, whereas those with slightly lower monthly household income (NTD 40,001~80,000) displayed the highest repurchase intention.


**Table 3.** Descriptive statistics of different population groups in potential constructs.

#### *4.2. Measurement Model Test*

Prior to structural equation modeling (SEM) analysis, I performed an outlier test, a normality test, and multicollinearity test to ensure that the data fulfilled basic assumptions [63,64]. The results of the

Q-Q plot indicated no problematic outliers in the measurement results of the individual constructs. The highest absolute value of the skewness coefficients of the four constructs was 1.201, and the highest absolute value of the kurtosis coefficients was 4.890. For a normal distribution, the absolute values of the skewness and kurtosis coefficients must be less than 3 and 10, respectively [64]. Thus, all four constructs can be regarded as following normal distributions. The most reliable index for multicollinearity testing is the variance inflation factor (VIF). The VIF of the two exogenous latent variables in this study, product performance and relational capital, was 1.022, thereby indicating lack of a high correlation between the two [65]. The measurement model was first tested, then structural equation modeling (SEM) analysis was conducted using AMOS 20. The measurement model test results indicated that the overall fitness of the measurement model was acceptable (see Table 4) with *<sup>X</sup>*<sup>2</sup> *d f* = 2.126, RMSEA = 0.056, GFI = 0.938, and AGFI = 0.908. Table 4 shows that the composite reliability of the potential constructs ranged from 0.724 to 0.905, all exceeding 0.70 and thus indicating acceptable reliability [66]. According to Anderson and Gerbing [67], the requirements for convergent validity include standardized factor loading greater than 0.400 and reaching the 0.001 level of significance as well as average variance extracted (AVE) greater than 0.500. As shown in Table 3, the standardized factor loadings of the potential constructs ranged from 0.531 to 0.919 and all reached statistical significance (*p* < 0.001). Except for the AVE value of product performance being 0.469, the AVE values of all the other potential constructs exceeded 0.500. Thus, the convergent validity of the potential constructs was acceptable.


**Table 4.** Measurement model test.

Goodness-of-fit: χ2/df = 2.126, RMSEA = 0.056, GFI = 0.938, AGFI = 0.908, \*\*\* *p* < 0.001.

Chin [68] pointed out that AVE values greater than 0.500 and coefficients of the correlation between constructs being lower than the square roots of the AVE values are required for discriminant validity. The AVE values of the constructs in the model ranged from 0.469 (close to 0.500) to 0.744, and their square roots fall between 0.685 and 0.863 (see Table 5). Thus, the measurement model has suitable discriminant validity.


**Table 5.** Correlation matrix of constructs and square roots of their AVE values.

Note: The bold figures on the diagonal are the root values of potential constructs.

#### *4.3. Structural Equation Modeling (SEM) Analysis Results*

After confirming the suitability of the measurement model, I conducted SEM analysis to test model fitness and the hypotheses. The important fitness indices of the overall theoretical model were as follows: χ2/df = 2.233, GFI = 0.935, AGFI = 0.905, and RMSEA = 0.059. Except for RMSEA being slightly higher than 0.05, the remaining indices all reached the ideal criteria.

The analysis results of the overall theoretical model (Figure 2) indicate a significant and positive correlation between product performance and relational capital (Ψ = 0.20, *p* < 0.01). Thus, H1 is supported. The influence of product performance on repurchase intention was positive and significant (β = 0.30, *p* < 0.001), thereby supporting H2a. The influence of relational capital on repurchase intention was positive and significant (β = 0.32, *p* < 0.001). Thus, H2b is supported. The standardized coefficients of the influences of product performance and relational capital on repurchase intention were 0.30 and 0.32, respectively. I can thus say that the effect sizes of the two on repurchase intention are roughly the same. The influence of product performance on subjective well-being was positive but did not reach the level of significance (β = 0.08, *p* < 0.22). Thus, H3a is not supported. The influence of relational capital on repurchase intention was positive and significant (β = 0.24, *p* < 0.001), thereby supporting H3b.

**Figure 2.** Structural equation modelling (SEM) analysis results. Note: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001. The dashed line indicates that the influence was not statistically significant.

#### **5. Discussion and Conclusion**

Farmers' markets have many functions, such as directly connecting farmers and consumers and promoting local production for local consumption. Furthermore, they can serve as cultural attractions, which can provide opportunities for rural tourism. They offer opportunities for training in production and marketing so that farmers can improve their effectiveness in marketing and management. Finally, they provide venues for social interactions and social embeddedness between consumers and farmers and improve their social well-being. This study combined the economic and social perspectives of consumer participation in farmers' markets and examined the relationships among important concepts of the post-purchase stage of consumer consumption at farmers' markets, namely, product performance, relational capital, repurchase intention, and subjective well-being.

The first finding was that a significant and positive correlation exists between product performance and relational capital, which indicates that farmers' markets can provide consumers with quality, safe, and superior agricultural products as well as the opportunity to form and accumulate diverse relational capital such as face-to-face interactions, communication and understanding, learning, and development of friendship. Consumers who perceived greater product performance at farmers' markets accumulated more relational capital, so the two have a mutually complementing effect. In other words, farmers' markets with greater product performance can better enable consumers to form and accumulate relational capital during the consumption process. This result confirms the views expressed in previous studies [10,41,42], in which the social interactions such as conversations and information-sharing between buyers and sellers at farmers' markets enable consumers to establish social capital. At the same time, they promote consumers' understanding of market product characteristics and their ability to choose products of good quality, thereby leading to good product performance evaluation.

The second finding was that product performance and relational capital are both important factors of the repurchase intention of consumers, and the two are roughly equal in importance (β values equaling 0.30 and 0.32, respectively). The analysis results indicate that the effects of product performance and relational capital on repurchase intention are almost the same. In other words, whether consumers can experience high-quality interpersonal relationships and social interactions during consumption processes is just as important to customer loyalty to farmers' markets as whether consumers can purchase safe and high-quality agricultural products. Thus, existing customers can be encouraged to return to the market using an economic approach: for example, improving the quality of the agricultural products on sale to increase consumer ratings and satisfaction regarding the products. A social approach, which involves improving the quality of interactions between producers and consumers and providing consumers with deep psychological attachment and social support, could also prove effective. The analysis results indicate that these approaches are of equal significance.

The third finding was that the economic product performance does not have a significant impact on the subjective well-being of consumers at farmers' markets. This is consistent with the results derived by Cassia et al. [46] and Chen and Scott [47]. In contrast, the social relational capital does have a significant impact on the subjective well-being of consumers. This means that the economic perspective of whether consumers can buy safe and high-quality agricultural products at farmers' markets does not affect their subjective well-being. However, if a consumer has a good consumption experience with good interpersonal relations and social interactions at a farmers' market, then he or she will be more likely to go again. On the whole, the economic product performance and the social relational capital are of equal importance to repurchase intention. However, only the social relational capital is significant to subjective well-being. The question of whether product performance or relational capital is more important in this study is thus answered. The empirical data of this study indicates that the social relational capital is more important the economic product performance.

The managerial implications of the literature review and research findings are as follows. (1) Consumers can engage in more in-depth social interactions or establish close relational capital at markets, and at the same time, information exchanges enable them to acquire more product information and purchase agricultural products of greater quality. Thus, farmers' market planners should aim to provide high-quality agricultural products and maintain a friendly, social atmosphere that is conducive to communication. (2) To enhance consumer loyalty and repurchase intention, measures to improve the quality of economic and tangible products are necessary; however, measures to improve the social and intangible qualities of the environment cannot be ignored. (3) Aside from contributing to the local economy and serving as a form of leisure consumption, shopping at farmers' markets can enhance the subjective well-being of the consumers. In policy planning, administrative departments should encourage the establishment and promotion of farmers' markets.

Finally, I put forward five suggestions for future research. First, research applying both the economic perspective and social perspective to farmers' markets is still new; I suggest that future studies develop and test different constructs and frameworks to enrich research in this respect. Second, there are still many other factors that impact the product performance evaluation, relational capital, repurchase intention, and the well-being of consumers at farmers' markets, such as

demographic variables, personality, and socio-economic status, which I did not include in our analysis model. Future studies can use control variables or interference variables to encompass these factors and thereby enhance the rigor of research in this respect. Third, the operations of farmers' markets involve various stakeholders, including the government, communities, managers, farmers, and consumers. Future studies should include all of the aforementioned stakeholders as participants when investigating a topic. Fourth, including different demographic variables such as gender, age, educational background, and income in the SEM for multi-group path analysis to compare the moderating effects of different population groups would be interesting and meaningful. There were only 358 valid samples in this study, which was not enough for said analysis. It is recommended that future studies collect more samples for multi-group path analysis. Fifth, because of using convenience sampling, this research may have compromised the representativeness of the sample. Future studies can use random sampling method to enlarge sample's representativeness.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Culinary Tourism Experiences in Agri-Tourism Destinations and Sustainable Consumption—Understanding Italian Tourists' Motivations**

## **Riccardo Testa, Antonino Galati, Giorgio Schifani, Anna Maria Di Trapani and Giuseppina Migliore \***

Department of Agricultural, Food and Forestry Sciences, University of Palermo, Viale delle Scienze13 - Edificio 4-, 90128 Palermo, Italy

**\*** Correspondence: giuseppina.migliore@unipa.it; Tel.: +39-091-23896618

Received: 1 August 2019; Accepted: 21 August 2019; Published: 23 August 2019

**Abstract:** Culinary tourism represents an emerging component of the tourism industry and encompasses all the traditional values associated with the new trends in tourism: respect for culture and tradition, authenticity and sustainability. Italy is known worldwide for the richness and variety of its gastronomy, and agri-tourism represents one of the most important places where culinary tourists can experience local food and beverages. By using a modified version of Kim and Eves' motivational scale, the present study aims to investigate which motivational factors affect the frequency of culinary tourists to experience local food and beverages in agri-tourism destinations. The findings of the present study reveal that the social and environmental sustainability, among the other motivations, has shown to play a crucial role in influencing Italian tourists' frequency to experience local food and beverage in agri-tourism destinations.

**Keywords:** gastronomy; local food; cultural experience; social and environmental sustainability; rural development; food consumption

#### **1. Introduction**

The local food movement is part of the contemporary social movements aiming to change the global agricultural landscape by altering the way we understand and interact with the food system [1,2]. Its main goal is focused on shortening the distance between producer and consumer, in order to increase the social, economic and environmental sustainability of the food system, and to strengthen the cultural identity of the territories [3]. According to Feenstra [4], the local food movement is founded on social and cultural interests, by including support for producers, local economies and environmental protection, through the production, processing, distribution and consumption of local foods. An emerging component of this movement is culinary tourism [5], which emphasizes unique foods and dishes from the culture of a specific region [6]. Hall and Sharples [7] considered culinary tourism as the leisure pursuit of a memorable eating and drinking experience, made to places where good foods are prepared for the purpose of fun or entertainment, which incorporates visits to local producers, food fairs, farmers' markets, cooking demonstrations and any food-related tourism activities. Among EU countries, Italy has shown a considerable growth of culinary tourism over the last years, becoming one of the most dynamic and creative segments of tourism [8]. According to the latest available data [9], culinary tourists in Italy have exceeded 110 million in 2017 (of which 43% domestic tourists), with an economic impact of 12 billion euros (15.1% of the Italian tourism sector). Italy, in fact, represents a tourist destination whose brand image is connected, with varying levels of intensity, to gastronomic

values, thanks to the fact that it represents the first EU country for protected designation of origin (PDO), protected geographical indication (PGI) and traditional specialty guaranteed (TSG). According to EU regulations, the PDO label identifies a product in which all the phases of the production process (production, processing and preparation process) must take place in the specific region. The IGP differs from PDO, as at least one of the stages of production, processing, or preparation takes place in the region. TSG refers to all those food products whose production processing is recognized and used throughout the area concerned, according to traditional rules agri-food labels (825 agri-food products with these labels) and traditional agri-food products (5056 labels).

Culinary tourism includes several areas such as wine tourism, beer tourism, gourmet tourism and gastronomic tourism, and tourists can experience it through local and unique restaurants, breweries, wineries, culinary events, farmers' markets and agri-tourisms [10,11]. In Italy, agri-tourism represents one of the most important places where culinary tourists can experience local food and beverages. As indicated by McGehee and Kim [12], agri-tourism is a farm that combines agricultural production with a component of rural tourism. In Italy, agri-tourisms amounted to 23,406 in 2017, showing an increase of about 27% over the last 10 years [13]. Agri-tourisms are seen as effective means of supporting local economies as they represent an important source of income diversification for farmers [14], and a way to contributing to the preservation of landscapes and cultural heritage in the rural areas [15]. According to the latest available data, in Italy the number of tourists in the agri-tourism sector exceeded 12 million, with an economic impact of over 1.4 billion euros [16]. An estimate, the latter, that seems destined to grow, is also in consideration of the fact that tourists are increasingly interested in consuming food products and dishes that are characteristic of specific territories [17]. Accordingly, the ability of agri-tourisms and territories to attract culinary tourists could assume a winning role for the development of the whole economy in rural areas and to contribute to enhancing the value of the local farm's products through its association with the social and cultural context [18].

However, to date few efforts have been made to understand culinary tourists' motivations to experience local foods and beverages [19,20]. With reference to agri-tourism literature, the research has been mainly focused on the supply side of the market, with most studies concerning entrepreneurial motivations [21,22]. Few studies have investigated the tourists' motivations to visit agri-tourism destinations [23], some of them mainly focused on the recreational component of agri-tourism, including being with family, and enjoying natural landscapes and the smells and sounds of nature [24]. Very few efforts have been done on the link between the tourists' motivations of experiencing local food and beverage and the interest in visiting agri-tourisms [25]. To the best of our knowledge, no study has explored the role of motivations on the consumption of local food and beverages in agri-tourism destinations in Italy. Furthermore, it is not clear which motivations play a key role in influencing the choice to consume local food in agro-tourism destinations. The current study aims to reduce this gap by contributing to understanding which motivational factors affect the choice of Italian culinary tourists to experience local food and beverages in agri-tourism destinations. Considering the importance of the role played by food in the choice of a tourist destination, findings of the present study could contribute both to enrich the literature on culinary tourism and to drive agri-tourist operators who want to shape their business model to satisfy the costumers' expectations.

Using a behavioural approach, we conducted a survey in Italy through an online questionnaire with modified items from the Kim and Eves' scale measurement of tourists' motivations to consume local food [26]. We chose this scale as it seems the most comprehensive tool for inferring how tourists perceive local food and beverages in agri-tourism destinations. Furthermore, to understand whether motivations linked to sustainability aspects of food consumption affect the choice to experience local food and beverage in agri-tourism destinations, we implemented in Kim and Eves' scale some motivational items deriving from local food consumption literature.

#### **2. Culinary Tourism and Local Food Consumption: A Focus on the Existing Relevant Literature**

Culinary tourism is an increasingly expanding sector of tourism in which tourists experience local food and beverages of other destinations and cultures [27]. Over the years, in fact, culinary tourism is becoming an emergent alternative to mass tourism, inasmuch culinary tourists increasingly try to gain new experiences in an active, differentiated and unique manner than the choice of reaching standardized touristic destinations [28]. What emerges is that many holiday destinations worldwide are very sought-after for their traditional food and beverages [29,30].

In scientific literature, culinary tourists are also called foodies as they, by means of the local food and beverages, are looking for a genuine and memorable experience [31]. Several authors claimed that the most important factor that pushes culinary tourists towards specific destinations is the desire to taste local gastronomy [32–34]. Similarly, Dann and Jacobsen [35] highlighted that an important aspect of culinary tourism is to practice variegated sensory experiences, as the taste of local gastronomy is different to the taste of same food in own country or region.

Green and Dougherty [5] conceptualized culinary tourism as a subset of cultural tourism, by asserting that food and beverages are expressions of specific cultures. They perceived local food and beverages as guarantees of authenticity, since they emphasized unique regional dishes, telling a collective memory made of knowledge, flavours and peasant rituals. This was also supported by the UNWTO that recognized food as a key element of all cultures and a major component of global intangible heritage. Culinary tourism, in fact, incorporates moral and economical qualities dependent on the territory, landscape, local culture, local food items and authenticity [36]. In the context of tourism research, motivations are important constructs to understand tourists' behaviour, as they are seen to affect the choice of a touristic destination [37], or participation to a particular touristic activity [19,38]. In the literature, different motivations have been recognized to affect the choice of tourists to experience local food. Culture seems, in fact, an important motivator that affects culinary tourism. According to Fields [20], tourists' desire to experience local food and beverages in a tourist destination is strictly linked to cultural motivations, as experiencing new foods and dishes means also experiencing new cultures. In addition, Kim and colleagues [39] showed that healthy eating is another important motivational factor affecting tourists' interest in local food. They affirmed that tasting local food and beverages in their place of origin is perceived by tourists as a means to improve psycho-physical health, as they are perceived to be fresher and more nutritious.

Culinary tourism is perceived by tourists also as a change of everyday routine and eating habits in order to try new food experiences and to obtain a certain prestige with their family and friends [20,39]. This is in line with Schultz [40] who denoted that nowadays more and more tourists are in search of authentic travel and food experiences, thanks also to the role of the media that have positively influenced tourists' perspectives about the link between tourism and gastronomy [41].

Moreover, previous studies highlighted as culinary tourism could be an opportunity to socialize and be together with family and other people, as participating in festivals and events based on local food are able to build social relations, by contributing to make the tourist experience more pleasant [42,43]. This is also supported by Sotomayor and colleagues [24], related to agri-tourism experience in Missouri (U.S.), who found that being with family and enjoying natural landscape were important motivators for visiting agri-tourism destinations.

Moreover, Barbieri and colleagues [44] found that experiencing the farm lifestyle and learning about farming are crucial motivations for visiting agricultural environment for recreation. The same is true for the tourists' perception that agricultural environment is associated with the authenticity of experiences [45,46]. However, little effort has been made to understand how the motivations to visit the agri-tourism destinations are linked to the choice to consume local food and beverages. For example, Kline and colleagues [25] found a positive association between the consumers' concerns over the humane treatment of animals and environmental impacts of the mainstream supply chain and the interest of tourists to experience local food, in particular, meat, in agri-tourism destinations. The line of research on consumer preference for local food highlights further motivational dimensions affecting consumer choice, including environmental and social sustainability motives [47].

These motivations, which highlight an ethical aspect linked to the consumption of local food, have been little explored in previous research on tourists' motivation to consume local food [25]. In particular, the literature emphasizes that environmental protection play a crucial role in affecting local food consumption. Indeed, according to Migliore and colleagues [48] consumers are often driven to consume local food because they are motivated to reduce the environmental costs of food production and distribution. Moreover, some studies reported that consumers perceive local food to be better for the environment and also for society than organic food [49].The preference for local food seems linked to the perception that local farms, more in particular, the small-size ones, adopt agricultural practices with less dependence on synthetic pesticides and fertilizers [49].

Several studies highlighted that the growing interest of consumers is often associated with the perception that local food is more nutritious, healthier and of higher quality than that sold in the mainstream supply chain [50,51], also because products have travelled a short distance to reach the consumer's table [47]. However, the most frequently named motivations for expressed preferences of local food among consumers seem related to environmental protection and support of the local economy [47].

Local food consumption has been recognized to increase social and economic justice in rural communities, as consumers desire to support local farms that have difficulty in entering the traditional commercial channels [52]. In particular, Sage [53] explained that direct interactions between farmers and consumers generate solidarity, promoting the recovery of a sense of morality within the agri-food sector. Indeed, besides supporting farms from an economic point of view, consumers try to create direct relationships based on solidarity and trust.

In scientific literature, several studies have analysed how socio-demographic characteristics of culinary tourists affect their choices and destinations. According to some authors [54–56], there is a specific profile of culinary tourists, inasmuch they usually have both a medium-high level of income and education and an age range between 35 and 45 years. In particular, Pérez-Priego et al. [54] showed that the majority of culinary tourists are women, hold a university degree and have a household monthly net income higher than 2000 euro.

By integrating both tourists' motivation literature and local food research, this study contributes to enriching understanding about motivational factors, and socio-demographic characteristics, influencing the consumption of local food and beverage in Italian agri-tourism destinations.

#### **3. Methodology**

#### *3.1. The Modified Kim and Eves' Scale Measurement of Tourists' Motivations to Consume Local Food*

In the literature on consumer decision-making, motivations refer to a set of psychological constructs that explains why people behave in certain ways. They can be thought of as the antecedent condition that compels human behaviour to occur [19].

In order to explain tourists' behaviour to taste local food and beverages, Kim and Eves [26] developed a motivational scale composed of five motivational dimensions, generated by 26 items. The five motivational dimensions were cultural experience, excitement, interpersonal relation, sensory appeal and health concern (Figure 1).

In particular, 'cultural experience' is associated to the tourists' desire to experience different cultures, since experiencing new foods and dishes means also experiencing new cultures. 'Excitement' dimension is related to the need to practise exciting experiences during holiday, also associated with the need to escape from routine.

**Figure 1.** The five motivational factors of Kim and Eves' motivational scale.

The third dimension identified by Kim and Eves was 'interpersonal relation', which is seen as a desire to meet new people, spend time with friends and family and get away from routine relationship.

Culinary tourism is also seen as a sensory experience. 'Sensory appeal' is, in fact, the fourth dimension found by Kim and Eves and it is related to the sensory characteristics of food that can play an important role in culinary tourist' choices. 'Health concerns' is the fifth motivational dimension affecting local food and beverages consumption in touristic destinations identified by Kim and Eves.

Moreover, several previous studies have highlighted that prestige is another factor affecting tourist choice to experience local food and beverage in tourist destinations. In fact, Kim and colleagues [39], Crompton and McKay [19] and Botha and colleagues [57] pointed out that prestige, in the context of local food experience, is an expression of self-esteem and it derives from the need to create a favourable impression on other people. Therefore, considering the above-mentioned motivational factors, a modified version of Kim and Eves' motivational scale was proposed in this study, including new items deriving from local food literature. More in detail, in Table 1, are reported the 26 original items from Kim an Eves' scale, 4 items from previous studies on tourists' prestige, and 10 items recognized on local food literature, emphasizing the social and environmental sustainability, as well as the healthy eating associated with local food.




#### *3.2. Data Collection and Methods*

In order to detect and measure the effects of tourists' motivations on the frequency of consumption on local food and beverages in agri-tourisms, an online survey was performed, involving 412 Italian tourists who visited agri-tourism destinations. Data were collected between the spring and winter 2018, and participants were recruited through invitations to participate in the online survey. In this study, a snowball sampling recruitment technique was adopted, as it allowed to reach a large number of participants by means of their interpersonal relations and their social connections. It is worth noting that, despite this technique did not provide a fully representative sample, it allowed collecting a wide variety of information in a short time [58,59]. For the survey, a questionnaire was adopted, which was organized in three sections. In the first section, information was collected regarding the frequency of local food consumption in agri-tourism destinations, the main occasions affecting this frequency (e.g., food festivals, cooking school, or just to experiencing local food in its area of origin), and the preferred place for agri-tourism destination (rural area far from the cities or near tourist attractions). The second section of the questionnaire was reserved to gather information about the main motivations affecting tourist to taste local food and beverages in agri-tourism destinations. For each motivational items, participants were asked to rate the level of importance for these motivational items using a Likert scale ranging from 1 to 7 (where 1 was not important and 7 was very important). The third section of the questionnaire included participants' socio-demographic indicators, such as age, gender, education (in four categories: primary school, lower secondary school, upper secondary school and a university degree or higher) and household monthly net income measured in euros.

Of the 412 questionnaires collected, 29 were deleted because they were incomplete, while 383 were retained for the analysis. To identify the main motivational factors of Italian agri-tourists, data were analysed using a Principal Component Analysis (PCA). It allowed the analytical transformation of a set of correlated variables into a smaller number of independent variables or constructs, reducing the number of variables while minimizing the loss of information. To achieve a more meaningful and interpretable solution, during the extraction process, items with factor loadings higher than 0.5 were retained, while those with a value lower that 0.5 and those cross-loaded on more than one factor were eliminated (more in particular, they are the items No. 3, 4, 5, 6, 7, 19, 22, and 23, shown in Table 1). Cronbach's alpha was also calculated to measure the internal consistency of items in each factor or principal component. After that, factor scores were also calculated for each principal component, expressing the contribution of each observation to the composition of factors. The factor scores were used for the subsequent ordered logit econometric model, allowing ordered categories of a dependent variable to be modelled as a sequence of latent variables, y\*, through increasing threshold levels [60].

The dependent variable was constructed as the annual frequency of local food and beverages consumption in agri-tourism destinations. This was subdivided into three categories in increasing size by level of consumption: 'low', frequency of local food consumption in agri-tourism destinations less than 2 times a year; 'medium', frequency of local food consumption between 2 and 4 times a year; and 'high', frequency of local food consumption in agri-tourism destination higher than 4 times a year (Table 2).

**Table 2.** Distribution of the three categories of the dependent variable.


The independent variables were the factor scores obtained from PCA and some other agri-tourists' socio-demographic characteristics. In order to measure the effects of each motivational factor on local

food and beverages consumption, the odds ratios (ORs) were also determined. An OR quantifies the changes in the probability of the dependent variable after a unit change in the independent variable. This means that when the OR is equal to 1, the effect of the unit variation of the independent variable on the dependent variable is null, thereby maintaining the values of the other explanatory variables constant; the larger the deviation from the unit value, the greater the effect of the independent variable on the dependent variable [60].

#### **4. Results**

Of 383 participants in the survey, 209 were females (equal to about 55% of the respondents). Data processing showed that the average age of agri-tourists participating in the survey was 34 years ranging from 25 to 72 years, and 46% had a university degree and almost 50% had an upper secondary school. Over 53% of them lived in a household of three or four members. In total, 45% of respondents declared a household monthly net income between 2160 euros and 3240 euros, 18.5% of them between 3241 euros and 4350 euros, while only 6.3% declared a household monthly net income of over 5400 euros. Moreover, 31% of respondents visit agri-tourism destinations during food festivals, 42% just to experience local food in its area of origin and about 27% for cooking schools or other events organized in agri-tourism. The application of factor analysis allowed reducing the initial number of variables (31) into 6 factors, which accounted for 63.4% of the total variance. The KMO value at 0.93 exceeded the acceptable minimum value of 0.6 [61], and Bartlett's test of sphericity was observed to be significant (*p* < 0.000). The Cronbach's alpha scores were higher than 0.7, ranging from 0.78 to 0.89, indicating that the variables exhibited correlation with their factors and can be considered as internally consistent [61].

The Factor Analysis allowed identifying the main motivational factors characterizing tourists' choice to consume local food in agri-tourism destinations (Table 3). The first factor extracted, which represented one of the motivational dimensions characterizing tourists' choice, accounted for 16.8% of the explained variance and was named 'social and environmental sustainability'. It is characterized by items associated with the motivations of being in solidarity with local farmers, of contributing to the local economy, of maintaining the agricultural landscape, of eating food that has not travelled for long distance, of conserving the local environment and its natural resources, of perceiving local food more environmentally-friendly, and with the motivation associated with the importance that local food ate in agri-tourism is organically certified.

The second motivational factor extracted was named 'health concern' and what's more, it represented 13.8% of the explained variance. This factor is characterized by the items emphasizing the health aspects of local food, that is, local food is good for health, it is more nutritious, it is perceived free of synthetic chemicals harmful to health, it contains a lot of fresh ingredients produced in the local area and knowing the producer is guarantee of the wholesomeness of local food. The third motivational factor extracted accounted for 11.8% of the explained variance. It was named 'cultural experience' as it is characterized by items associated with the opportunity that eating local food increases the knowledge about different local cultures and increases understanding about local cultures, and with the perception that experiencing local food is an authentic experience and creates excitement. Compared to Kim and Eves' scale [26], in this study the excitement dimension has been incorporated into the cultural experience dimension, almost to emphasize that cultural motivations are associated with the desire to have an exciting experience.



The fourth motivational factor was named 'prestige', as a holiday destination, such as agri-tourism, through its local food and beverage can communicate something about tourists' status. This factor accounted for 10.9% of the explained variance and is characterized by items associated to the desire to take a picture of local food to show friends, to give advice about local food experiences to people who want to travel, to talk about local food experience and to enrich intellectually.

The fifth motivational factor accounted for 5.4% of the explained variance and was explained by two variables. It was named 'sensory appeal', as it is characterized by the importance that the local food ate in agri-tourism tastes good and smells nice. The sixth motivational factor was called 'interpersonal relation'. It is characterized only by two items that emphasize the role of local food associated to the desire to spend time with friends and family and a means to meet new people with similar interest.

As mentioned in the methodology section, the factors score of motivations identified during factor analysis process were analysed, together to the socio-demographic variables of respondents, with an ordered logistic econometric model. The explanatory variables implemented in the model approximate the main motivational factors and socio-demographic characteristics, affecting the choice of Italian culinary tourists to consume local food and beverages in agri-tourism destinations. Coefficients with a positive sign indicating that as an explanatory variable increase, so do the probability of falling in the category with the highest frequency of local food consumption in agri-tourist destinations. Some of the signs of the estimated coefficients were highly significant and consistent with the expected signs (Table 4).


**Table 4.** Results of the econometric model (Ordered Logistic Regression).

<sup>ˆ</sup> *p* > |z|: \* 10%, \*\* 5%, \*\*\* 1%.

The latent variable, defined as the frequency of local food consumption in agri-tourist destinations, increased with the rise in all the explanatory variables apart from 'health concern' and 'sensory appeal' motivational factors which were statistically not significant in explaining the high frequency of local food consumption in agri-tourist destinations. However, it is important to note that the non-significance of these two motivational factors could be due to the high mean values that the respondents gave to the related items, regardless of the agri-tourists' local food consumption levels.

The results obtained described the influence of the motivational factors on tourists' choice to experience local food and beverages at agri-tourist destinations. In particular, the frequency of local food consumption among participants increases with the growing importance attributed to particular factors. It was found that four of the six motivational factors examined (cultural experience, prestige, social and environmental sustainability, and interpersonal relation) were important in influencing tourists' local food consumption.

Among the socio-demographic variables, 'age' 'gender' and 'income' were found to increase the probability of consuming local food in agri-tourist destinations. The negative sign of coefficient related to the variable 'gender' showed that being female increases the probability to experience local food and beverages in agri-tourist destinations. Among the socio-demographic variables, only 'education' was not statistically significant. The not-statistically-significant nature of this variable could be due to the high level of education of participants; in fact, as previously mentioned almost 96% had a upper secondary school and a university degree.

The calculation of the odds ratios (ORs) revealed the ratio between the probability of high frequency of local food consumption in agri-tourist destinations and the cultural experiences, prestige, social and environmental sustainability, and togetherness motivational factors. Meanwhile, among the factors identified in the factor analysis, the 'social and environmental sustainability' motivational factor showed the biggest influence on the probability to have a high frequency of local food consumption in

agri-tourism destinations (1.44 times); followed in order by the 'cultural experience' motivational factor (about 1.39 times), 'prestige' and 'interpersonal relation' motivational factors, which showed almost the same effect (about 1.32 and 1.30 times, respectively). Among the socio-demographic variables, only income showed a relatively biggest influence on the probability to have a high frequency of local food consumption in agri-tourism destinations (1.10 times).

#### **5. Discussion**

By integrating existing studies investigating motivations to consume local food and the stream of research on tourists' motivations, the present study set out to investigate the main motivational factors affecting the choice of culinary tourists to taste local food and beverages in Italian agri-tourism destinations. Gastronomy, in fact, is one of the most important elements affecting the authenticity of a tourist destination. Italy is known worldwide for the richness and variety of its gastronomy [17], and agri-tourism represents one of the most important places where culinary tourists can experience local food and beverages [62]. This is one of the first study investigating tourists' motivation to experience local food and beverage in agri-tourism destinations, revealing that 'cultural experience', 'prestige', 'interpersonal relation' and 'social and environmental sustainability' play a crucial role in influencing Italian tourists' frequency to consume local food and beverage in agri-tourism destinations. Relative to 'cultural' motivator, Kim and colleagues [39] highlighted that local food is an important means through which tourists experience the culture of a destination. Similarly, Fields [20] suggested that cultural motivation allows tourists to experience the culture of a particular destination, making them closer to the place. Also, Kim and Eves [26] found that culture is an important aspect of local food consumption since experiencing local food allows tourists to increase knowledge about different local cultures. This is also consistent with Ruiz Guerra and colleagues [46], who emphasized that culinary tourism, in particular oleotourism, seeks to combine environment, culture, tradition and gastronomy that create a new model of sustainability in rural environments. 'Prestige' is another important motivator influencing the consumption of local food and beverage in agri-tourism destinations. This is reliable with past research, uncovering that local food experience has a role in conscience improvement or smugness [39]. McIntosh et al. [38] described status and prestige motivations as closely related to the tourists' wish of attracting attention from others. This was also discussed by Hall and Winchester [63] who observed that the tourists' desire to learn about traditional food or wine contributes to creating a favourable impression on others. 'Interpersonal relation' is also recognized to play a crucial role in tourists' behaviour to have a high frequency of consumption in agri-tourism destinations. This is also consistent with Kim and Eves [26] who pointed out that socializing with new people and being together with family is recognized to be an important factor in tourist motivation to experience local food and beverage in tourist destinations. Social and environmental sustainability is a new motivational factor that the econometric analysis showed to play an important role in explaining tourists' behaviour to consume high frequency of local food in agri-tourism destinations. This is in line with Kline and colleagues [25] who found that animal welfare and environmental sustainability are important motivators for eating local meat in agri-tourism destinations. This is also supported by the literature on local food consumption, revealing that the desire to protect the environment play a crucial role in consumers interest towards local food [64,65]. Migliore and colleagues [47] highlighted that consumers' concerns to reduce the environmental costs of food production and distribution plays a crucial role in influencing consumers' interest in local food. Seyfang [66] emphasized that the 're-localization' of food is associated to consumers' motivation of reducing the impacts of 'food miles', understood as the distance food travels between being produced and being consumed. The environmental and social sustainability dimension is also linked to the importance tourists attribute to the organic certification of local food products. This is consistent with Basha and Lal [67] who highlighted that environmental concern and the desire to help local producers assisted the economy, developed their society and significantly affected the purchasing intentions for organically produced foods.

However, compared to Kline and colleagues [25], the present study highlights that social sustainability is another important motivation that together with environmental sustainability plays a crucial role in affecting the choice to consume local food in agri-tourism destinations. This is consistent with Zepeda and Leviten-Reid [68] who emphasised that consumers are motivated to purchase local food, as they wish to sustain the social and economic conditions of a local rural community, by recirculating financial capital and encouraging new forms of entrepreneurship. This contributes to the resilience of rural communities, where local farms are often strongly integrated, playing a positive role in strengthening and supporting the social and economic conditions of the local community [69,70]. This is particularly true in a rural tourism context since tourism could represent one of the most important economic development strategies in these areas [71]. Ferrari et al. [72], in fact, showed that an increase in rural tourists results in an overall positive increase in demand both for local food productions and handicraft products, generating an increase in regional value-add. This result strongly supports the relevance of sustainability as a crucial determinant of the competitiveness of agri-tourism destinations. Research seems to agree that the competitive destination has to deliver an experience that is more satisfying compared to similar destinations, and it is associated with the ability to preserve natural and cultural resources, which, in turn, increases long-term well-being of its residents [17,73,74]. This has important implications for rural development, as at farm level, tourism contributes to enhancing the value of the farm's products through its association with the social and cultural context [18]. Thus, culinary tourism plays a crucial role in the sustainable development of territories, as it allows to improve its economy and also to strengthen the cultural and social identity of residents [75]. The strong linkage between local food and tourism could stimulate the creation of entrepreneurial networks in a territory, by strengthening the whole economy and increasing the quality of life of residents [76]. In this context, the culinary tourism can represent a way to reduce the growing problem of sustainability in tourism, by ensuring socio-economic development of entire communities [77]. Therefore, agri-tourism destinations could represent an alternative to the unsustainable mass tourism practices, which have caused a detrimental use of urban and coastal spaces for tourism purposes [78].

However, in the analysis, not all these motivational factors identified during the factor analysis procedure are able to explain the high frequency of consumption of local food in agri-tourism destinations. Although 'health concern', as shown in the econometric results, is not a motivator influencing the frequency of consumption of local food in agri-tourism destinations, it was found to be an important characteristic of local food in holiday destinations, regardless of how often local food is consumed. The same is true for 'sensory appeal' motivational factors. The high scores on the items related to the perception that local food is healthy are rather in line with the previous literature, where local food is perceived to be fresher and more nutritious than the food that has travelled for long distances [51] and free of synthetic chemicals [47]. Moreover, knowing the producer is perceived by consumers as a guarantee of the wholesomeness of local food [48].

Finally, relative to socio-demographic characteristics of respondents, similar to Baderas-Cejudo et al. [79] and Nicoletti et al. [80], findings reveal that being older and with a higher level of household monthly net income significantly increase the probability of tourists to experience local food and beverages. In particular, older tourists seem to feed an important niche market, as they have both the time and purchasing power to try and experience local food and beverages, especially in rural areas. On the contrary, other studies showed that culinary tourists usually are aged between 35 and 45 years, and women are more attracted towards gastronomy destinations than men [81,82].

Although the results of the present study show that the variable educational level is not statistically significant to explain the frequency of local food consumption at agri-tourism destinations, the educational levels of participants is rather high as emphasized in other studies [83,84].

#### **6. Conclusions, Limitations, Implications and Future Research**

In an era of globalisation, there is a particular desire to enjoy varied, rather than mono-cultural ambience and experience. In this context, over the last years, in order to increase the social, economic

and environmental sustainability of the food system, and to strengthen the cultural identity of the territories, local food movements are spreading worldwide. Culinary tourism represents an emerging component of the tourism industry and encompasses all the traditional values associated with the new trends: respect for culture and tradition, a healthy lifestyle, authenticity, and sustainability. Accordingly, promoting culinary tourism in agri-tourisms represents a winning strategy for the development of the whole economy of rural areas.

At the local level, in fact, food and beverages experienced in agri-tourism can contribute to rural socio-economic development by creating new job opportunities and new added value, becoming a resource especially for many small-sized farms that cannot compete in an increasingly globalized market. Similarly, rural areas can play a crucial role in differentiating the tourism offer, giving the opportunity to experience local cultures and traditional dishes and beverages.

The results seemed to show that different motivations affect tourist in agri-tourism destinations; however, the main motivational factor that seems to explain the consumption of local food and beverages in agri-tourism destinations is 'social and environmental sustainability', a new motivational factor deriving from the measurement-scale proposed. This highlights that sustainability could play a crucial role in the competitiveness of agri-tourism destinations.

Understanding which motivational factors affect the tourists' choice to consume local food and beverages in agri-tourism could contribute to defining competitive marketing strategies of tourist destinations, in order to better align them to tourists' preferences. Marketing is a useful tool for agri-tourism's competitive strategy, as it provides agri-tourism operators with the ability to differentiate the products or experiences they offer from those of their competitors. For example, agri-tourism operators could focus their strategies by communicating the link between culinary tourism and the environmental and social sustainability of agri-tourism destinations. Therefore, it could be important for agri-tourisms to adopt CSR initiatives and business strategies focused on the social and environmental components of sustainability. In line with this, the adoption of certified environmental management systems, such as organic certification for local foods, could contribute to satisfying the needs of tourists in the environmental field. From this point of view, it is essential for the agri-tourisms adopting effective communication strategies, for example, the CSR reporting, in order to communicate to the tourists and local communities the engagement in the environmental and social field. At the same time, agri-tourism operators should commit themselves to better communicate the link between the local food and territories and the cultural content of dishes offered.

The findings of the present study could also enrich the extant literature on culinary tourism and agri-tourism demand, and reinforce business literature which supports that consumers have a positive attitude towards sustainable food products and tourism.

However, the present study faces some limitations. They are inherent to its very methodological nature and to the convenient sample used in the study, based on the voluntary participation of respondents. Therefore, the study does not intend to provide conclusive evidence but helps readers to have a better understanding of the trend of culinary tourism in agri-tourism destinations. Moreover, this study only focused on motivations, which are a part of psychological factors known to influence behaviour. Excluded were other individual factors such as attitudes, consumers' awareness and personal values, as well as cultural and social factors.

Therefore, further advancement in culinary tourism research should take into account a larger sample, as well as extending the study to foreign tourists' preferences, and other social and cultural contexts in order to validate the effort proposed in this study. Finally, future research should incorporate different theories to better understand the complex issue of individual behaviour. Furthermore, in the light of the increase in the agri-tourism demand, further research should take into account the Dialogical Self Theory or the Social Capital Theory [85] in order to explain how local communities shape their perceptions in light of changes in the environment due to a growing tourists' presence in the rural area.

**Author Contributions:** This study is a result of the full collaboration of all the authors. However, R.T. conceived and designed the study, wrote the 'Culinary Tourism and Local Food Consumption: A focus on the Existing Relevant Literature' section, and 'The Modified Kim and Eves' Scale Measurement of Tourists' Motivations to Consume Local Food' sub-section, G.S. and A.M.D.T. collaborated to write the 'Data Collection and Methods' sub-section and 'Conclusions' section, and A.G. wrote the 'Discussion' section, while G.M. wrote 'Introduction and Results' sections, and coordinated the study.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Sustainability* Editorial Office E-mail: sustainability@mdpi.com www.mdpi.com/journal/sustainability

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-1554-0