According to the Food and Agriculture Organization of the United Nations (FAO) research data [1
], at least one-third of the world’s food is wasted every year. The total amount is almost as much as three billion tons. Lebersorger and Schneider [2
] discovered in a study in Austria that food waste has increased since July 2014, reaching a peak in December 2014. Bilska et al. [3
] indicates that the causes of food waste at the supply chain include such factors as incorrect package labels, incorrect food product weight labels, damaged expiration date labels and packaging, among others. In addition, according to the Taiwan Food and Drug Administration [4
] report, Taiwan throws out 3.6 thousand tons of food a year. Nearly thirty percent of this amount is suboptimal food thrown directly in the trash. Hooge et al. [5
] show that even though suboptimal foods differ from regular products in appearance, expiration date or packaging, their quality and safety are similar. Consumers often reject edible foods with changes in visual, sensory quality, or which passed the expiration date [6
]. Therefore, the World Food Programme (WFP) is advocating the Food Recovery plan. The goal is to create a three-way benefit for society, economy and the environment through food waste reduction [7
]. However, there is not much related research focusing on suboptimal food, which is the reason for this study.
A number of models, designed to better predict and understand human behaviors, have been proposed by social psychologists over the past several decades. Among these models, the theory of planned behavior (TPB) is the one that is most frequently utilized, and it is widely regarded as being effective when used for predicting behavior in general [9
]. For this reason, a range of researchers have successfully utilized the TPB in order to better understand consumer decisions regarding food, a phenomenon which explains, in turn, why consumer food choices have become a topic of substantial interest among researchers conducting TPB-related studies [12
According to the TPB, a combination of three factors is responsible for the formation of behavioral intentions, with those factors being subjective norms, perceived behavioral control, and attitudes toward the behavior in question [20
]. Figure 1
depicts the theory in the form of a structural diagram. In this context, an attitude toward a behavior is defined as the ‘degree to which a person has a favorable or unfavorable evaluation or appraisal of the behavior in question’. Three main components in the attitude of a person, including affect, cognition and behavior. The first component is affect. This component relates to a person’s emotional response to product. The second component is cognition. This component is a person’s belief or knowledge about product. The latter component is behavior. This component relates to the tendency of a person to perform a certain action with regard to its attitude towards product. While a subjective norm refers to any ‘perceived social pressure to perform or not perform the behavior’. Finally, perceived behavioral control refers to ‘an individual’s perceived ease or difficulty of performing the particular behavior’. Accordingly, if an individual perceives himself or herself to have a relatively high level of behavioral control with respect to a given behavior, then that individual will have a greater likelihood of exhibiting a firm intention to engage in said behavior. The term behavioral intention itself, meanwhile, refers to the individual’s willingness to engage in the behavior in question, with the assumption being that such willingness must necessarily exist immediately prior to the actual behavior itself being carried out [20
According to Ajzen [20
], the TPB can be modified through the addition of new variables or by changing the path for existing variables. Relatedly, previous studies have found that, while the TPB fundamentally assumes that behavioral intentions are the result of subjective norms, attitudes, and perceived behavioral control, a number of domain-specific factors not incorporated in the model are also relevant [21
]. Moreover, recent studies in the field of psychology have generated noteworthy evidence supporting the inclusion of various predictor variables in the TPB. More specifically, the value of the TPB in terms of its predictive power regarding various domains has been shown to be enhanced when additional predictor variables are included [22
]. The study has also attempted to include new constructs (environmental concern and sensory appeal) in the TPB taking support from the extant literature.
When consumers purchase products, they do not base their decisions purely on their preferences for the product. Consumer behavioral changes that arose from the increase in environmental awareness include: valuing a low-carbon diet, encouraging support for local produce, supporting the purchase of seasonal, organic and fair-trade products, and attaching greater importance to product labeling [24
]. Aschemann-Witzel et al. [26
] explored consumer behavior towards price-reduced suboptimal foods in the supermarket and the relation to food waste in households. Moreover, Loebnitz and Grunert [27
] also discovered that food abnormality, consumers’ environmental concerns and social trust are significantly correlated with product purchase desire. Aschemann-Witzel [28
] explored the factors that influence acceptance of expiration date based pricing of suboptimal food by applying four sets of stickers writing “fight foodwaste”, “lower price”, “save more” and “fight foodwaste and lower price- save more” to the suboptimal food items in store. Therefore, this study will consider environmental concern as one of the research variables.
Baker et al. [29
] indicates that a product’s sensory appeal will affect consumers’ product choice preference and purchase desire. Sensory appeal refers to the appeal of the product’s taste, appearance, texture and smell to consumers [15
]. Loebnitz and Grunert [27
] shows that consumers will choose fruits and vegetables with a perfect appearance. Thus, sensory appeal and purchase desire are inseparable. Symmank et al. [30
] demonstrate a positive relationship between sensory perception, overall liking, and purchase intention for visually suboptimal bananas, yet overall liking and purchase intention decreases when the product exceeds a certain ripening status. Loebnitz and Grunert [31
] indicated that consumers perceive abnormally-shaped vegetables as more risky, and paradoxically, they associate natural vegetable shape-abnormalities with GM, despite having no other information available. Hence, this study will also consider sensory appeal as one of the research variables.
Above all, this study utilizes the TPB model and considers the two research dimensions of environmental concern and sensory appeal. We aim to develop an extended Theory of Planned Behavior (TPB) model, which includes environmental concern and sensory appeal to predict consumers’ purchase intention to suboptimal foods. We hope that the findings will provide important insights for suboptimal food and useful recommendations for marketing channels, suggesting promotion of suboptimal food may be the key of potential business.
A summary of the verification of the hypotheses made in this study is shown in Table 6
. In this study, the clear identification of the determining factors of consumer attitudes and intentions regarding the purchase of suboptimal foods was the main research objective. To this end, the TPB was utilized as the theoretical framework for the study, while the study also sought to expand upon the TPB via the incorporation of additional constructs within it. At first, H1 and H3 were supported. The results of the SEM showed that it was important whether purchasing suboptimal foods would be desirable (attitude) or easy to purchase (perceived behavioral control). The results of the study indicated that consumers’ intentions toward the purchase of suboptimal foods are significantly affected by the attitudes of those consumers toward the suboptimal food in question, as well as by the consumers’ perceived behavioral control, which is similar to conclusions drawn in Dean et al. [33
] and Zhou et al. [35
]. On the other hand, subjective norms of suboptimal foods did not directly affect purchase intention. Therefore, we concluded that H2 was not supported, whereas the subjective norms of the consumers were not found to exhibit any significant effects on said purchase intentions.
More specifically, the results indicated that, among the key variables of the TPB, perceived behavioral control is the factor most responsible for determining the intentions of consumers regarding the purchase of suboptimal food, a finding which in turn suggests that consumers deal in a relatively effective manner with factors which could be regarded as disabling factors. That said, it should be reiterated that the purchase intentions of consumers regarding suboptimal food were found to also be affected to a significant extent by their attitudes towards the suboptimal food in question. This finding underscores the substantial relevance of consumers having positive attitudes regarding a given suboptimal food when they are engaged in the process of considering whether or not to buy it. Meanwhile, because the subjective norms of the consumers in this study were not found to exhibit any significant influence on the intentions of those consumers toward the purchase of suboptimal foods, it can be concluded that that the purchasing of suboptimal foods has not yet become a social norm among people in Taiwan.
The additional constructs incorporated in the TPB model in this study were environmental concern and sensory appeal. The research hypothesis (H4) states that the more consumers are concerned about the environment, the more likely they are to buy suboptimal foods. The results indicated that environmental concern had significant effect on the intentions of the consumers in this study to purchase suboptimal foods. The results of this study are consistent with research by Pagiaslis and Krontalis [49
], and Smith and Paladino [50
]. It proves that consumers in Taiwan now do care much about environmental issues.
At last, H5 was supported. the study results indicated that the respondents’ purchase intentions were significantly and positively affected by their perceptions of the sensory appeal of suboptimal food. The results of this study are consistent with research by Fotopoulos et al. [54
], and Padel and Foster [55
]. In other words, in the event that consumers have the perception that buying a suboptimal food is exciting, fun, or otherwise pleasurable in some regard, the sensory experience conveyed by the food will be a relevant factor in determining the consumers’ purchase intentions toward the food.
The results of the study indicated that both the original TPB and the proposed theoretical framework fit the data well. However, the model fit of the TPB in terms of the explained variance (that is, the adjusted R2) was substantially enhanced by the incorporation of the additional constructs of environmental concern and sensory appeal within the TPB as part of the proposed framework, with the explained variance increasing from 40.2% for the original TPB to 58.6% for the proposed framework. This boost to the model fit validates the incorporation of these two constructs within the TPB, at least when utilizing the TPB for the purpose of predicting the intentions of consumers in a developed nation regarding the purchase of suboptimal foods.
According to the questionnaire results, any lack of access to suboptimal foods poses a substantial barrier to consumers. More specifically, the results indicated that consumers take a practical approach toward the purchase of food and do not want to make visits to several stores in order to find the products they want. As such, when developing a strategy for the sale or marketing of suboptimal foods, the availability of the foods must be regarded as a particularly important parameter to consider. In general, the study results indicated that consumers exhibit high intentions to purchase suboptimal foods (mean of PI = 6.12/7). As such, the interested parties (such as farmers, wholesalers, etc.) should strive to enhance the distribution channels for suboptimal foods so that the foods will be available at markets and stores that are convenient for their intended customers, thus making it easy for those consumers to purchase the suboptimal foods.
The present study had two primary goals. First, the study sought to explore the utilization and value of the TPB when applied to investigating the intentions of consumers toward the purchase of suboptimal foods. Second, the study sought to enhance the predictive power of the TPB through the incorporation of two additional constructs within it, namely, sensory appeal and environmental concern.
The results of the study indicated three key findings. The first is confirmation that, with respect to consumer intentions regarding the purchase of suboptimal foods, the TPB can serve effectively as a framework for predicting said intentions. The second, however, is partial support for the notion that the utility of the TPB can be increased further still through the incorporation of metrics of sensory appeal and environmental concern. In other words, both sensory appeal and environmental concern appear to be useful constructs in furthering the understanding and predictability of consumer intentions regarding the purchase of suboptimal foods. Relatedly, the results of this study provide support for existing evidence regarding the relevance of these two constructs in the prediction of intentions with an environmental orientation. Finally, the third key finding of the present study is that among consumers in Taiwan specifically, personal attitudes, sensory appeal, and environmental concern are all of critical importance in any effort to predict consumer intentions regarding the purchase of suboptimal foods. At the same time, further investigations will be necessary in order to further illuminate the precise mechanisms through which sensory appeal and environmental concern exert their effects on purchase intentions.
5.2. Limitations of the Research and Future Research
This study had a number of limitations. First, rather than being asked for their feelings regarding a specific category of products, the study respondents were queried regarding their broad perceptions of various attributes and their behavioral intentions toward suboptimal food in general. Relatedly, because the behaviors and expectations of consumers may vary with respect to different categories of suboptimal food products (such as dairy foods vs. non-dairy foods), it is suggested that future studies regarding the purchase of suboptimal foods could provide additional insights by investigating more domain-specific attitudes and purchase intentions, that is, by investigating such attitudes and intentions toward specific suboptimal food items or categories. Another limitation of the present study was that it did not attempt to investigate the influences of any moderating effects. As such, future investigations could build upon the findings of the present study by looking at the impacts of such moderating effects, such as, for example, the socio-demographic attributes of consumers and their levels of trust in various actors within the food sector (such as farmers, producers, and vendors). Furthermore, it is suggested that future studies could conduct comparisons of those consumers who prefer and do not prefer suboptimal foods in order to ascertain the differing characteristics of the two groups, if any, as well as exactly how their behaviors and perceptions differ with respect to both conventionally-produced and suboptimal foods.