Consumer acceptance and willingness to pay for direct and indirect entomophagy

Abstract

Over the last few years, the interest on alternative protein sources, such as edible insects, has been growing rapidly. However, Western consumers’ acceptance of insects as a food source is very low, mainly due to unfamiliarity with insect-based food. We investigate consumers’ attitude and behavior and estimate their willingness to pay (WTP) a premium for three products that vary on a between-subjects basis, direct or indirect consumption of insects. The data were collected through an online questionnaire of 451 consumers in Greece and WTP was elicited using the contingent valuation method (CVM). Our results show that the majority of Greek consumers are not willing to pay a premium for an insect-based energy bar and cookie (direct entomophagy) and they would require a discount to acquire such products. On the other hand, consumer acceptance is higher for a gilt-head bream that is fed with insect-based feed (indirect entomophagy). Results show that younger adults, indirect consumption of insects, looking for new sources of food and evaluating certification and trust as important for innovative products, are all associated with a higher WTP. Consumers with positive WTP are on average willing to pay a premium of 15.8, 17, and 31.8 per cent for the energy bar, cookie, and gilt-head bream, respectively, while consumers that are not WTP a premium would require discounts of 43.8, 42.4, and 30.7 per cent, respectively.

1 Introduction

The global increase in demand for meat, coupled with finite inputs for production, has led to a larger intensification in the exploration of alternative protein sources (Van Huis 2016). Consequently, over the last few years, insect-based food products have drawn a lot of attention in both developed and developing countries (Imathiu 2020). The use of insects as food is widespread in developing countries and it is generally part of the diet of at least two billion people, mostly located in tropical countries. According to Jongema (2017), 2,111 recorded species of edible insects are consumed by more than 3,000 ethnic groups of humans (Van Huis et al. 2013; Van Huis 2013), and these foods are consumed in 130 countries with the African, Asians, and American continents being the most entomophagous (Ramos-Elorduy 2009; Gahukar 2011; Van Huis et al. 2013).¹

However, in most Western countries the interest in consumption of food based on edible insects as a substitute of meat, remains very low (Hartmann and Siegrist 2016; House 2016). The main reason of the low acceptability is, not surprisingly, the feeling of disgust that insects cause to people that are not familiar to insects as part of the human diet (Rozin and Fallon 1987; Ruby et al. 2015). Lack of information on the safety of insect-based products as well as lack of a clear legal framework and regulations, did not help in the past in wider consideration of insects as a useful primary food source by consumers or as feed input in the business sector (Halloran et al. 2014). However, the more recent adoption of the novel food regulation in EU (2015/2283) as applied since January 2018 (Lähteenmäki-Uutela et al. 2018), paves the way for wider adoption of insect-based foods. Edible insects fall under the definition of novel foods and the regulation sets a clear procedure on how insects-based foods can be authorized before being placed on the market (Lähteenmäki-Uutela et al. 2018; Lotta 2019).² What seems to be generally accepted, is the recognition that consuming insects does not belong to the traditional Western diet (Dagevos 2020). Besides the disgust factor, consumers are also averted to insects as a food source because of the knowledge of the origin and habits of the insects, as well as of the potential perceived negative effects that consumption might have for their health (Rozin and Fallon 1980). The literature has also pointed to attitudes and perceptions as important factors that have a significant effect in shaping consumers’ acceptance of edible insects (Elorinne et al. 2019; House 2019; Iannuzzi et al. 2019; Jensen and Lieberoth 2019; Mancini et al. 2019).

Despite their low acceptability, foods based on edible insects are considered an excellent alternative source of protein for humans. In fact, due to the continuous increase of the world population and their nutritional needs, research for innovative food products that are rich in nutrients is expected to intensify (Nakagaki and Defoliart 1991). Consumption of foods based on insects can be particularly beneficial because of the high nutritional value that insect-based food has. More specifically, eating foods based on edible insects can constitute important source of vitamins, minerals, and especially proteins, contributing significantly in solving food scarcity problems of developing countries, while also being an important complementary food source in developed countries (Ramos-Elorduy 1997; Caparros Megido et al. 2014).

The environmental impact that mass breeding of insects has in relation to traditional livestock like pork, cows, and chicken (Oonincx et al. 2015), is an additional benefit of food based on insects. Farmed insects require small areas to grow and have low water and feed needs (Rumpold and Schlüter 2013; Muller et al. 2016). Furthermore, an important advantage of insects is that they can also be used as a raw material for animal feed, and especially so for fish, poultry, and swine where so far, the replacement of fish/soy meal components in the diet of some farmed fish and poultry species has not shown any negative effects on the flavor, juiciness, or texture of the final products (Sogari et al. 2019). The use of insects as feed has beneficial properties and can be used as a replacement for conventional animal-derived proteins, especially in aquafeeds (Gasco et al. 2020). For example, Iaconisi et al. (2017) demonstrate that mealworm larvae meal is a promising alternative that can be used as a protein source for the partial replacement of meals in the diet of the blackspot sea bream. Insects can replace a significant percent of the expensive forms of proteins, which are currently used in feed production and on fish farming (Halloran et al. 2014). It is also worth mentioning that according to Piccolo et al. (2017), mealworm larvae meal can replace fish meal up to 25 per cent of inclusion in the diet for gilt-head sea bream without negative effects on weight gain, crude protein, and ether extract digestibility. In Uganda, roughly 5 per cent of the farmers use termites for feeding fish and the quantity of produced termites depends largely on the number and size of termite hills on the farm as well as on the termite species (Van Huis et al. 2013). In this way, significant quantities of raw materials that are used in the production of animal feed can be diverted to human consumption. Using insects as a cheap source of protein can lower production costs of breeders without deteriorating the diet quality of fish, while at the same time an important source of protein is added in the diet of farmed fish (Halloran et al. 2014).

Given the importance of insect-based proteins in foods and feed production, we explore the factors that potentially affect consumer acceptance and willingness to pay (WTP) for farmed insects. Because consumer acceptance for farmed insects likely differs with direct or indirect consumption of insects (direct vs. indirect entomophagy, see La Barbera et al. (2020)), we elicit valuations for different products by randomly assigning respondents to one of two treatments. For roughly half of the respondents, we elicit valuations for insect-based food, that is, food in which insect proteins have been directly integrated in the product (direct entomophagy). For the other half of the respondents, we elicit valuations for farmed fish that have been fed with insect-based feed (indirect entomophagy).

The products we use are an energy bar with insect protein and a biscuit with insect flour (for the direct entomophagy treatment) and a farmed gilt-head (sea) bream that has been fed with insect-based feed (for the indirect entomophagy treatment). In the next section, we review the relevant literature in order to set the context of our study. In Section 3, we present our data collection and value elicitation methods. We then present our results in Section 4 and conclude with a discussion and implications of our findings in the last section.

2 Literature review

Despite the growing interest for insect-based foods and the gradual popularity that insect-based products are gaining, there are still important challenges that insect-based products need to overcome in order to gain a wider acceptance in Western countries. For EU countries in particular, after the approval by the European Commission of the new Regulation on Novel Food (EU Regulation No 2015/2283), insect-based products were gradually made available in some countries after January 2018 and in some others even before that date (e.g., in the Netherlands, Belgium, Germany, United Kingdom, Spain, Finland, Denmark, France), where a total of fifty-nine companies are involved in the production, processing or sale of insect-based food (Pippinato et al. 2020). However, in many other countries (e.g., in Greece, Cyprus, Hungary) insect-based products are still not available likely due to limited consumer acceptability (Gere et al. 2017).

Consumer acceptance is a key factor of wider availability of insect-based products in Western markets. Due to the high interest for such products, the literature on consumer acceptance of insect-based products is rapidly expanding and several studies have evaluated consumer acceptance of insect-based products for different parts of the world. Lensvelt and Steenbekkers (2014) report that approximately 35 per cent of Dutch and Australians consumers that took part in their online survey had tasted insect-based products before. Furthermore, in their choice experiment in Australia at the same time period, they report that approximately 34.5 per cent of the participants had tasted insect-based food before. Van Thielen et al. (2018) report a lower rate for Belgian consumers: just 11.2 per cent of Belgian consumers had already tasted foods with processed insects. This is in contrast to Caparros Megido et al. (2016) that report that 33 per cent of subjects from Belgium had tasted insect-based products before.

A few other studies have explored the factors that may determine acceptance for insect-based products. Curiosity has been shown to be one of the most important reasons for consumers to try insect-based foods (Sogari et al. 2017; Van Thielen et al. 2018). However, trying once a product may not be enough to retain a consumer in the long run, since consumers may not be willing to repeatedly consume insect-based food if it is not regarded to be tasty or appropriate for consumption and factors related to the product preparation seem to play a major role in determining acceptability rather than taste familiarity and individual traits (Tan et al. 2016). As Rozin and Fallon (1987) show, disgust plays an important role in food rejection. This result is in line with Sogari et al. (2017), where 25 per cent of their sample in Parma in Italy stated that they would not taste edible insects because they perceive insects as disgusting.

Before we present the set of studies that examine the factors that play a crucial role in the acceptability of insect-based products, it is worth discussing how product form affects willingness to taste insect-based foods. Gmuer et al. (2016) point out in their survey carried out in Switzerland that products that contained the processed and therefore less visible insect ingredients (tortilla chips made of cricket flour and tortilla chips containing deep-fried cricket bits) had lower negative emotional evaluation and higher positive emotional evaluation than the products that contained visible whole cricket bodies (a snack consisting of tortilla chips and deep-fried crickets). Similarly, Orkusz et al. (2020) find that willingness to eat processed insect foods is far higher than for unprocessed whole insects. Moreover, a cultural factor may be at play since Hartmann et al. (2015) found that although willingness to eat was significantly higher for processed food items (i.e., drinks and cookies) than for unprocessed food items (i.e., crickets and silkworms) in Germany, willingness to eat ratings in China did not differ significantly between processed and unprocessed food items, with the exception of silkworm protein. Labeling in relation to flavor expectations might also play a role with respect to the product form. Le Goff and Delarue (2017) assessed consumers’ non-verbal reactions to insect-based products and their flavor expectations. Consumers were presented with flavored potato chips labeled either as ‘protein enriched’ or as ‘insect protein enriched’. They found that congruent flavors (i.e., chicken and barbecue) generated significantly more negative expressions, higher negative affect scores, and lower liking scores than incongruent flavors (i.e., strawberry and blackcurrant) and while consumers rejected the idea of tasting chips, they seemed to accept it after the first bite, indicating that westerners might be willing to take a first step toward insect consumption.

Kornher et al. (2019) suggest that apart from the taste factor, consumers who are concerned about environmental and nutrition issues are more likely to try foods with processed insects. The importance of environmental and nutrition issues in insect-based food consumption is supported by other studies as well (Menozzi et al. 2017; Van Thielen et al. 2018). For example, Palmieri et al. (2019) found that Italian consumers who were careful about food sustainability and collective health, were (on average) about 22 per cent more likely to be willing to consume insects than those who were not. There are, however, many concerns whether nutritional benefits and the eco-friendly footprint of consuming insects as a source of food in comparison to meat products, is likely to make consumers adopt insect-based products in their diet (Laureati et al. 2016; Wilkinson et al. 2018).

Willingness to taste and acceptability of insect-based products is also affected by appearance (Caparros Megido et al. 2014; Wilkinson et al. 2018), food neophobia and disgust (Sogari et al. 2019; Barton et al. 2020; Orkusz et al. 2020). Lammers et al. (2019) identified sensation seeking and food disgust as the most important predictors of willingness to consume insect-based foods, in addition to other known influential factors such as food neophobia and food technology neophobia. Similar findings are presented in Woolf et al. (2019), who concluded that disgust-based rejection, food neophobia, and low availability are the main determinants that hinder consumption of insect-based foods.

In contrast to the previous cited studies, Schlup and Brunner (2018) do not find food neophobia to be a key predictor of willingness to consume insects for subjects in Switzerland where one-fifth of the participants had consumed insects at least once in their lifetime. Prior consumption of insect-based food as well as information provision have been shown to affect consumer acceptance of food containing processed insects by reducing the feeling of disgust (Barsics et al. 2017) while taste exposure increases acceptance for these products (Sogari et al. 2018). In another study, acceptability ratings of Italian consumers significantly increased after they had received information about the benefits of consuming insects (Laureati et al. 2016). Moreover, the level of sensory-liking of a mealworm burger among Dutch consumers increased to a similar level as that of the beef burger after tasting it, whereas prior to tasting it, the mealworm burger was rated significantly less positively than the beef burger (Tan et al. 2016). Schouteten et al. (2016) compared three burgers (insect-based, plant-based, and meat-based) and report that overall acceptance and perceived nutritiousness of the insect-based burger were significantly higher than the meat-based burger during an informed condition for a sample of Belgian young adults. These studies indicate the importance of prior consumption, taste exposure, and information provision with respect to consumer acceptance of insect-based foods. However, Alemu and Olsen (2020) find that peer effects (i.e., observation of peers reacting negatively in terms of disliking an insect-based product) may counter any positive effects from tasting insect-based foods.

In a somewhat different approach, Hamerman (2016) investigated the factors that influence willingness to attend an event at which foods that contain insect-based ingredients were served. Undergraduate students completed a 25-item disgust scale (Olatunji et al. 2007), which contained three facets of disgust: core disgust, animal reminder disgust, and contamination-based disgust (Olatunji et al. 2008) and were informed about a local museum (an ‘insectarium’) that runs a program called ‘Bug Appétit’. Hamerman (2016) found that a priming manipulation related to cooking increased willingness to attend the ‘Bug Appétit’ program for a subset of the population, that is, those who were low in sensitivity to animal reminder disgust. Moreover, high levels of sensitivity to core disgust and animal reminder disgust—but not contamination disgust—reduced preference for attending the ‘Bug Appétit’ program.

As far as demographics is concerned, several studies find that males and younger consumers are more likely to adopt insects as a novel and more sustainable protein source than females or older consumers (Verbeke 2015; Tan et al. 2016; Menozzi et al. 2017; Wilkinson et al. 2018; Castro and Chambers 2019; Orsi et al. 2019). In general, Onwezen et al. (2021) show that there are several attitudinal and other factors that have a consistent effect on consumer acceptance for insect-based foods such as healthiness, taste, familiarity, food neophobia, disgust, social norms, etc.

Besides insect-based foods, the use of insects for animal feed is gaining a lot of attention. Spartano and Grasso (2021) investigated willingness to try (WTT) and WTP for eggs produced from insect-fed hens in the United Kingdom. They found out that only 17 per cent of participants were aware of insects as a potential animal feed while most consumers were WTT (72 per cent) and 87 per cent were WTP a 18.3 per cent premium price for eggs from insects-fed hens. Moreover, Bazoche and Poret (2021) show that providing information on the negative effects of overfishing and on insect feed as a viable alternative to fish-meal in aqua-feed, can improve acceptability of insects as fish feed. In addition, males were more likely to eat insect-fed fish and 76 per cent of the informed participants had a positive attitude to taste them. Similarly, Ankamah-Yeboah et al. (2018) find that the majority of consumers are not concerned about the type of feed when buying fish and only a 23 per cent of the people who took part in the survey, exhibited negative attitudes for fish fed with insects rather than standard feed. Ferrer Llagostera et al. (2019) investigated the perceptions and WTP of Spanish consumers for insect meal as a sustainable feeding alternative in aquaculture. Their results showed that Spanish consumers were WTP, a premium for a gilt-head sea bream fed with insect meal compared to fish produced with different feeds (wild feed, fish feed, mix vegi feed) while the gilt-head sea bream fed with insect meal was the most valued with respect to its environmental impact.

3 Data collection methods and experimental design

To evaluate consumers’ attitudes and behavior toward insect-based foods, we designed and distributed a web questionnaire in April 2020. The choice of a web questionnaire vis-à-vis face-to-face interviews was dictated by the coronavirus pandemic.

The questionnaire was first pilot tested in a small convenience sample and adjustments for understanding and better flow were made based on the feedback received. The link to the questionnaire was distributed in social media as well as directly emailed to a list of subjects from the general population that had previously participated in laboratory experiments at the Laboratory of Behavioral and Experimental Economics Science (LaBEES-Athens) of the Agricultural University of Athens.

The final sample consisted of 451 consumers (52.11 per cent female) and the only criterion that participants had to meet was being over 18 years old. Subjects were randomly allocated to one of the two versions of the questionnaire. The only difference between the two versions was about the valuation questions. In one version subjects were asked about their WTP for products that were based on insects, while in the other version we elicited WTP for a farmed gilt-head (sea) bream that had been fed with insect-based feed. Elicitation of valuations for these different products allows us to explore differences in WTP for direct and indirect entomophagy.

Moreover, before eliciting WTP, subjects’ were provided with information about edible insects as well as information about nutritional and environmental benefits that may result from insect-based food consumption. Additional questions tried to assess various factors such as whether subjects are knowledgeable about the existence of insect-based products, the importance consumers attach to the nutritional value and the environment footprint of their food choices, their desire to try innovative food products, the importance of safety certifications and confidence in innovative products, etc. Standard demographic characteristics were also elicited.

3.1 Value elicitation

To elicit valuations, we employed the contingent valuation method (CVM). The CVM is the main workhorse when it comes to measuring WTP values for public and private goods, services, or amenities. Most CVM studies are conducted in hypothetical contexts, particularly in environmental valuation studies where a real market with salient payments is difficult to establish (Carson and Hanemann 2005; Carson 2012; Kling et al. 2012; Haab et al. 2013).

The favored elicitation format in the CVM literature has been the dichotomous choice (DC) format because of its well-known property of incentive compatibility.³ Although alternative elicitation methods are hypothesized to give rise to strategic and untruthful responses, researchers often use alternative elicitation formats driven by efficiency gains (e.g., lower sample size requirement), reduced complications associated with experimental design (e.g., bid design) and the possibility to increase the power of the experimental design (e.g., by asking about multiple goods in the same survey and/or by eliciting more precise information on preferences) (Johnston et al. 2017; Vossler and Holladay 2018). Vossler and Holladay (2018) identify assumptions under which open-ended and payment card formats are incentive compatible and find that incentive compatibility may not be an elusive goal when considering alternative elicitation formats or when studies include two or more value elicitation questions.

As mentioned before, subjects were randomly allocated to one of two treatments. In one of the treatments subjects valued food products that directly integrated insect proteins while in the other treatment, subjects valued a farmed fish that had been fed with insect-based feed. The insect-based products were a 60 g energy bar with chocolate that contained insect protein and a 60 g biscuit with chocolate that contained flour made from insects. Subjects were asked to indicate the premium (if any) they would be willing to pay over the price of a conventional product priced at € 2. In the second treatment, subjects were asked to value a gilt-head sea bream that had been fed with insect-based feed. Subjects were asked to indicate the premium (if any) they would be willing to pay over the price of a 500 g conventional farmed gilt-head bream priced at € 4 that had not been fed with insect-based feed.

Because the CVM involves creating a hypothetical valuation scenario in which consumers are asked to state their WTP for the product in question, we tried to mitigate hypothetical bias by preceding the valuation questions with a cheap talk script and a budget constraint reminder.⁴ The cheap talk script was compiled from several sources as well as our own previous work (e.g., Lusk 2003; Bulte et al. 2005) and reads as follows:⁵

• You will be presented with two hypothetical scenarios about whether you are willing to pay a certain amount of money for two products that contain edible insects in different forms. These two products will be a high insect protein energy bar with chocolate and a cookie with chocolate which is based on insect flour.

• It is important to remember that the questions are hypothetical and you will not be asked to pay anything. But we need from you to answer as if you had to pay the corresponding amount of money that you will state, since what happens often in hypothetical questions is that consumers state they are willing to pay a larger amount of money than what they are actually willing to pay. Your honesty is of great importance for us in order to be able to draw reliable conclusions.

The payment cards for the insect-based products were similar and were constructed with the following amounts: {0, 0.01–0.10, 0.11–0.20, 0.21–0.30, 0.31–0.40, >0.40}. Subjects had to indicate one of these as their preferred option. In the other treatment, where subjects had to indicate their premium for the gilt-head (sea) bream, the list of possible values for the payment card were as follows: {0, 0.01–0.20, 0.21–0.40, 0.41–0.60, 0.61–0.80, 0.81–1.00, 1.01–1.20, >1.20 }. Since we anticipated negative attitudes toward insect-based products, we also asked subjects that indicated a zero WTP, to indicate whether they would be willing to purchase such products with a discount. Subjects that stated they were not WTP a premium, were subsequently asked if they would be willing to purchase the product for a 20 per cent discount with a Yes/No possible answer. If they further indicated they were not willing to purchase the product for a 20 per cent discount, their purchase intention was further solicited for a 40 per cent discount.⁶

4 Data analysis and results

Before analyzing our data, it is useful to check whether there are any significant differences between the two treatments along demographic characteristics and attitudinal variables. Table 1 shows descriptive statistics of several variables and their normalized differences (Imbens and Wooldridge 2009; Imbens and Rubin 2016) in means and in dispersion. Normalized differences in means are given by |$|\bar{x_1}-\bar{x_2}|/{\sqrt{(s_1^2+s_2^2)/2}}$| where |$\bar{x_j}$| and |$s_j^2$| (j = 1, 2) are the group means and variances, respectively. Normalized differences in dispersion are given by ln(s₁/s₂) (Imbens and Rubin 2016). Cochran and Rubin's (1973) rule of thumb is that the normalized difference in location should be less than 0.25. The dispersion difference measure indicates smaller differences in dispersion when its value is closer to zero.

As shown in Table 1, only a couple of variables exceed the rule of thumb of Cochran and Rubin (1973), with a normalized difference in means greater than 0.25. However, these variables exhibit differences in dispersion that are close to zero, indicating a good balance between treatments in this set of characteristics and attitudinal variables.

We can gain some first insights by comparing stated responses regarding WTP. Figure 1 compares WTP responses for the three products.⁷ As evident, the distribution of responses is roughly similar between the cookie and the energy bar. A Pearson’s χ² test confirms this is the case (χ² = 5.992, P-value = 0.541). It is self-evident that the distribution of responses is shifted more to the right for the gilt-head bream compared to payment card responses for the cookie and the energy bar. A Pearson’s χ² test confirms that responses are significantly different between the products (χ² = 181.695, P-value <0.001).

4.1 Econometric analysis

In this section, we explore whether insights gained from the unconditional analysis of the previous section hold under conditional analysis. Given the nature of the dependent variable, we estimated interval regression models with clustered standards errors at the individual level where appropriate.⁸

Table 2 shows results using only the product dummies and dummies for the uncertainty scale. Model (1) shows results using the sample for which valuations were elicited for insect-based products (energy bar and cookie), model (2) is for the sample of subjects that valuations were elicited for the gilt-head bream, and model (3) is a pooled model. Considering both statistical and economic significance, two results come out of Table 2. First, whether the insect-based product comes in cookie form or energy bar form, does not have a significant effect on valuations. However, model (3) indicates that consumers are willing to pay a higher premium for a gilt-head bream that is fed with insect-based feed rather than products that contain insects for immediate consumption. Second, the negative constant term of model (1) indicates that, on average, subjects require a discount for the cookie/energy bar while the coefficient for the constant from model (2) indicates that no significant discount is required for subjects to purchase a gilt-head bream fed with insect-based feed (since we cannot reject the null that the constant is zero). The uncertainty dummies fail to reject the null of no effect at the 5 per cent level, indicating no significant effect of stated uncertainty over WTP.

Table 3 adds a set of demographic and attitudinal characteristics to the set of variables already used in Table 2. As evident, the effects of the product dummies are robust to the inclusion of the set of demographic and attitudinal characteristics. WTP for cookies and the energy bar do not differ significantly but WTP a premium for the gilt-head bream is significantly higher than the other two products. Moreover, effects from the uncertainty dummies are not significantly different from zero.

Table 3 may also help to characterize the profile of potential consumers of insect-based products. First, with respect to demographic characteristics, males and females do not differ in their stated WTP while age has a significant effect. In specific, older individuals are less willing to pay a premium across all products. Education and occupation do not produce systematic effects on WTP but income has a positive effect. Among factors other than demographics, those that seem to affect WTP a premium are the stated willingness to taste an insect-based product and agreeing with the statement that looking for new sources of food is good. Moreover, evaluating certification and trust as important for innovative products is associated with a higher WTP a premium.

An additional exercise we can do with our data is to use the predicted values from model (3) of Table 3 to graph the downward slopping curves shown in Figure 2. Each point on the curves indicates the percentage of respondents that would be willing to buy one unit of the energy bar, the biscuit and the gilt-head bream farm products, respectively, at the premium projected on the Y-axis. As evident, predicted WTP premiums span both the positive and negative axis. This is because a significant number of respondents stated they would require a discount to purchase the products. The crossover point of the downward slopping curves with the horizontal axis at zero indicates the percentage of respondents that would be willing to pay a premium and the percentage of respondents that would require a discount. For the energy bar and the cookie, only 24.3 and 28.3 per cent of subjects, respectively, are willing to pay a premium (conversely, 75.7 and 71.7 per cent of respondents would require a discount). The curve for the gilt-head bream is significantly sifted to the right indicating a higher rate of acceptance. About 44.45 per cent of subjects would require a discount to purchase the gilt-head bream fed with insect-based feed, while 55.55 per cent would be willing to pay a premium.

5 Discussion and conclusions

Given the growing interest for insect-based products as an alternative source of protein, acceptability and knowledge of consumer preferences for insect-based food can be a critical factor for the agribusinesses sector. Knowledge of consumer preferences would allow to make informed decisions about producing insect-based foods, especially for Western countries where entomophagy is mostly nonexistent. It would also allow effective targeting of segments of the population that are more likely to accept and purchase insect-based products.

To achieve this goal, we used a survey-based experiment with Greek consumers where we elicited their preferences using the CVM. In order to explore differences between direct and indirect entomophagy, we selected two kinds of products: (a) an energy bar and a cookie, where insect proteins can be directly integrated in the product and (b) a gilt-head bream farmed fish where insects are only part of the diet of the fish and are not directly consumed by people. Our results confirm the low acceptance of insect-based foods, since most subjects require a discount to purchase such products (we find that 75.7 and 71.7 per cent of respondents would require a discount to consume the energy bar and cookie with insect protein, respectively), while we find a much larger acceptance for a gilt-head bream fish that is fed with insect-based feed (55.55 per cent of subjects would be willing to pay a premium to purchase the gilt-head bream fed with insect-based feed). Naranjo-Guevara et al. (2021) find similar results using Dutch and German students. They observed that respondents were more willing to accept insects as animal feed than direct consumption of insects in their own diets, where 84 per cent of the respondents were willing to accept insects as animal feed and less than half of them (48 per cent) stated to be willing to eat products containing insects.

Our study can also be used by food companies that are looking to build the profile of consumers that are more likely to buy insect-based products: younger consumers, regardless of their gender, occupation, or education but of slightly higher income, that already trust innovation in food production, that find food certification important, and are actively looking for new sources of food. Consistent with other surveys by Lombardi et al. (2019) and Spartano and Grasso (2021), gender did not affect WTP while the significant effect of age confirms what is regarded as common knowledge that younger people are exhibiting a higher WTT new products (Schsler et al. 2012; Vanhonacker et al. 2013; Hartmann et al. 2015; Lombardi et al. 2019). Our findings are also interesting for fish farm business that consider partly replacing farmed fish diets with sustainable alternatives (Henry et al. 2015). Our results suggest that fish produced with feeds containing insects may be a larger than a niche market and fish farm businesses are likely to be able to market their products for a premium.

This study is not without caveats. First, due to unavailability of insect-based products in Greece, the participants had to deal with hypothetical products when eliciting their valuations. Although we took all the necessary steps described in the literature to mitigate or account for hypothetical bias, it is still possible that our WTP estimates suffer from it. As long as hypothetical bias uniformly affects the different products, it may be less of a problem in the experimental comparison of the treatments. Thus, our main result that consumers are less averse to indirect entomophagy rather than direct entomophagy, likely holds even in the presence of hypothetical bias.

In addition, we refrain from generalizing our results to the entire Greek population given limitations in data collection, although our findings can be considered as a first perspective about entomophagy in Greece that may help foster future studies in this field. Future research that combines consumer experiments with sensory evaluations is undoubtedly the way forward.

Acknowledgment

We would like to thank the Editor and two anonymous reviewers for helpful comments and suggestions.

Conflict of interest

The authors declare no conflict of interest. Data and codes to replicate the analysis have been deposited at the Open Science Framework: https://osf.io/zd6et/?view_only=468779d181934bbeab5837b57681ab38.

Footnotes

References