To meat or not to meat? Comparing empowered meat consumers’ and anti-consumers’ preferences for sustainability labels

Highlights

• Using a choice experiment we evaluate preferences for sustainability labels.

• We compare the preferences of meat eaters, meat reducers and vegetarians.

• Type of meat, price and fat content labels have the largest overall impact on consumer choices.

• Latent class analysis identified different segments within the three consumer groups.

Abstract

An increasing awareness of the impact of high levels of meat consumption on health and environmental sustainability is leading to a growing number of consumers reducing or avoiding meat. To address gaps in the literature, we compare and contrast the importance of the seven sustainability-related labels for three consumer groups (meat eaters, meat reducers and vegetarians) using a choice experiment involving 600 UK respondents (200 meat eaters, 200 meat reducers, 200 vegetarians). Type of meat, price and fat content labels have the largest overall impact on consumer choices. The impact of carbon footprint, method of production, origin and brand labels varies across consumer groups.

We subsequently use latent class analysis to identify heterogeneous intra-group consumer segments, based on their preferences, and highlight the socio-demographic differences between them. For meat eaters, three consumer segments are identified (empowered, traditional and price conscious meat eaters). Meat reducers are divided into health curtailers and sustainable consumers, while only one segment of vegetarians is identified. By drawing on signalling theory and the consumer empowerment and anti-consumption literature, we identify links between sustainable consumption, consumer empowerment and anti-consumption and provide valuable insights for policymakers and practitioners seeking to utilise food labels to encourage more sustainable consumption.

Introduction

Meat is increasingly being criticised as an unsustainable and unhealthy food choice, due to health risks and environmental concerns associated with its high carbon footprint and inefficient use of resources (Cliceri et al., 2018, de Boer and Aiking, 2017, de Boer and Aiking, 2019, Weibel et al., 2018). In 2015, the World Health Organisation called for a reduction in meat consumption, characterising several processed meat products as carcinogenic (Bouvard et al., 2015). Furthermore, a recent study by Oxford University reports that, in most middle- and high-income countries, red meat consumption exceeds recommended levels (Springmann et al., 2018). The same authors argue that a health tax on red and processed meat could prevent more than 220,000 deaths and save over US$40 billion in healthcare costs every year. An alternative, arguably more effective and less disruptive approach, involves encouraging individuals to voluntarily adopt more sustainable meat consumption patterns by targeting specific segments of consumers with interventions that motivate behaviour change (Apostolidis and McLeay, 2016a, Weibel et al., 2018). This includes strategies and policies to encourage meat reduction or the substitution of meat with more sustainable protein products, such as plant-based meat-free alternatives, also known as meat substitutes (de Boer et al., 2014, Verain et al., 2015).

Globally, businesses and policymakers are searching for effective ways to encourage more sustainable food consumption and inform consumers of the social, environmental and economic sustainability-related characteristics of their food (e.g. Grunert et al., 2014, Hawkes et al., 2015). As these characteristics generally involve ‘invisible’, credence attributes (such as method of production, country of origin and animal welfare), it is often difficult for consumers to acquire all the relevant information necessary to inform their choices (Fernqvist and Ekelund, 2014, Grunert et al., 2014). In the case of food products, food labelling schemes have been advocated to communicate these unobservable characteristics and enable better-informed consumer choices (Ardeshiri and Rose, 2018, Van Loo et al., 2014). For example, the UK’s Food Standards Agency (2013) suggests the use of a traffic light labelling system (green, amber, red) to communicate the nutritional content of food products, while the Carbon Trust has developed a carbon footprint label to communicate the impact of production on the environment. However, information overload and cognitive biases may mislead consumers and limit the effectiveness of food labels to provide information on credence attributes (Schuldt, 2013, Leathwood et al., 2007).

Providing consumers with easily accessible information when they are making purchasing decisions is becoming increasingly important, due to the increasing presence of ‘empowered’ consumers in the market. These consumers use their purchases as ‘votes’, to express their values and beliefs and influence businesses to develop ‘better’ products (Shaw et al., 2006, Spaargaren and Oosterveer, 2010). In the context of meat, ‘consumer empowerment’ can refer to ‘meat eaters’, who purposefully purchase more sustainable meat, as well as ‘meat reducers’ who decide to consume ‘less but better’ meat products (Apostolidis and McLeay, 2016b, de Bakker and Dagevos, 2012). In addition to empowered consumers, a growing number of meat ‘anti-consumers’ abstain from consuming meat or particular meat products (Armstrong Soule & Sekhon, 2018). In several countries, including the UK, there is a growing number of consumers reducing meat consumption or adopting vegetarian or vegan diets (Rosenfeld, 2019, Mintel, 2017a).

The current paper reports the results of the second and final stage of a two-part project, that investigated consumer preferences for the sustainability-related attributes and associated labels in the case of meat/meat-free products. In the first part of this project, we examined the impact of sustainability-related food attributes on the choices of UK consumers, using a representative sample of UK consumers (Apostolidis & McLeay, 2016b), revealing that consumer preferences vary across meat eaters, meat reducers and vegetarians. Other studies also corroborate that meat consumption (and anti-consumption) can be influenced by ethical, health and sustainability concerns (e.g. Hodson and Earle, 2018, Latvala et al., 2012, Peschel et al., 2016, Tosun and Gürce, 2018). Despite the suggested links between sustainability-related food attributes and meat consumption patterns, previous researchers have not compared the preferences of consumers with different meat consumption patterns for the labels used to communicate the sustainability-related characteristics. Therefore, in this paper we report the findings of the second stage of the project in which, by drawing upon signalling theory and the consumer empowerment and anti-consumption literature, we extend our previous choice experiment in order to compare the preferences between equal numbers of UK meat eaters, meat reducers and vegetarians. The specific objectives of this research are to:

• Compare and contrast the importance of sustainability labels for three consumer groups (meat eaters, meat reducers and vegetarians).

• Use latent class analysis to identify heterogeneous intra-group consumer segments, based on their preferences for different sustainability labels.

• Highlight any socio-demographic differences that exist between the segments.

• Draw on signalling theory and the consumer empowerment and anti-consumption literature to explain the results and their implications.

Mince (also known as ground meat) and meat-free mince are used in the experiment, due to their popularity and availability in the market (de Boer et al., 2014, Koistinen et al., 2013). By using both meat and meat-free mince in our experimental design, we are able to examine and compare preferences of people belonging in different consumer groups and segments, thereby contributing to existing knowledge and practice. As vegetarians and meat reducers have an increasing presence and power in the market, our research can assist efforts encouraging more sustainable diets and demarketing (i.e. discouraging the consumption of) unsustainable products (Armstrong Soule & Sekhon, 2018).

The role of empowered consumers, who ‘vote’ through their purchases for the product attributes they like, has been widely recognised in the literature (e.g. Spaargaren and Oosterveer, 2010, Shaw et al., 2006). The active engagement of empowered consumers and their significant power in the marketplace is driving changes in the market, as businesses do not generally invest in products that consumers are unwilling to buy (Papaoikonomou and Alarcón, 2017, Shaw et al., 2006). In contrast to empowered consumers, who use their purchasing power to influence businesses within a market, ‘anti-consumers’ reject the consumption of particular products, or abstain from the market altogether (Lee, Roux, Cherrier, & Cova, 2011).

Vegetarians and vegans are obvious examples of anti-consumers in the meat market. Recent reports suggest that in the UK as many as 9% of the population follow a vegetarian diet. Furthermore, over a quarter of UK meat consumers (28%) have limited the amount of meat they eat, while a further 14% has expressed intentions to reduce their meat consumption in the future (Mintel, 2017a). This highlights another aspect of anti-consumption which refers to practices of avoidance of particular products, instead of completely abstaining from the market (Cherrier et al., 2011, Black and Cherrier, 2010). Meat reducers (also known as flexitarians) can be considered as empowered consumers, if they only eat products that fulfil specific criteria (i.e. ‘less but better’ meat), as well as anti-consumers, if they generally avoid specific meat products (Apostolidis and McLeay, 2016b, de Boer et al., 2014). Chatzidakis and Lee (2013) explain this distinction between empowered consumers and anti-consumers, as the difference between reasons for and reasons against particular products. This means there is a difference between consumers whose choices are based on reasons in favour of particular product attributes (e.g. consumers eating meat substitutes because they like their taste) and those who are opposed to specific product attributes (e.g. consumers that choose meat substitutes because they disagree with animal killing). Accordingly, “an anti-consumer of meat (vegetarian) may avoid meat owing to concerns about animal welfare, but it is unlikely that those who consume meat do so because they want animals to be killed” (Chatzidakis & Lee, 2013, p. 191).

The difference between the reasons for and against consumption is particularly important in the cases of anti-consumers, considering that moving from consumption to anti-consumption is not a one-way process. For example, research suggests that meat anti-consumers may decide to move from following a meatless diet to consuming meat, if they feel that certain meat products fit their lifestyles, ideologies and beliefs (Hodson & Earle, 2018). Therefore, it is not appropriate to consider the behaviour of vegetarians or vegans as directly opposite to that of meat eaters. This emphasises the importance of research which simultaneously compares and contrasts the preferences of meat eaters, vegetarians and meat reducers. In the current study we address this issue, by comparing the impact that food labels have on the choices of these three consumer groups.

Sustainability is a multidimensional concept. According to the World Commission on Environmental Development (WCED, 1987), sustainability has a temporal dimension (related to trade-offs between present and future) and a social dimension (related to trade-offs between consumers and the society in general). In the late 1990’s, Elkington (1998) introduced the Triple Bottom Line approach to explain the different aspects of sustainability; social, environmental and economic sustainability. In this study, we follow the Triple Bottom Line view when exploring the links between sustainability and meat labels.

Communicating the benefits of sustainable food is often challenging, as many ethical, environmental and health related food characteristics are credence attributes that cannot be easily evaluated by consumers (Grunert, 2005). This can create information asymmetries between producers and consumers (for example sharing information on the method or environmental impact of production), which can make product choices more difficult (Grunert et al., 2014). Consequently, when consumers are uncertain about the quality of a product, they may opt for cheaper alternatives to minimise their risk, driving more expensive (but higher quality) products out of the market (Moussa & Touzani, 2008). For instance, although people express an interest in sustainability-related characteristics of meat, the demand for more sustainable meat and meat-free products is still low, as they tend to be more expensive than conventionally produced options (Mintel, 2017b, Apostolidis and McLeay, 2016b).

Product labelling can enable the effective communication of information on the credence attributes that individuals are interested in (Micheletti & Stolle, 2015). In his seminal work, Spence (1973) defines this communication of information about the characteristics of a product as ‘signalling’. Signals are “a marketer-controlled, easy-to-acquire informational cue, extrinsic to the product itself that consumers use to form inferences about the quality or the value of the product” (Bloom & Reve, 1990, p. 59). These signals are particularly important in situations where consumers need to form judgements of product quality under uncertainty such as when products are characterised mostly by credence attributes (Grunert, 2005). In the food marketing literature, food labels have been advocated as reliable and useful signals, which allow consumers to make inferences regarding the quality of the products (Apostolidis and McLeay, 2016a, Koistinen et al., 2013, Grunert, 2005). As a signal, a food label is designed to assist consumer choices by transforming credence features into attributes that consumers can search for prior to purchasing, hence reducing potential information asymmetry (Karstens and Belz, 2006, Jahn et al., 2005). For example, food labels may enable consumers to identify characteristics related to social, environmental and economic sustainability, such as nutritional values (Koistinen et al., 2013), origin (Kuchler et al., 2010, Pouta et al., 2010), carbon footprint (Apostolidis & McLeay, 2016b) and production method (Van Loo et al., 2014).

Despite consumers’ interest in sustainability food labels (e.g. Koistinen et al., 2013, Pouta et al., 2010), this interest is not always translated in higher sales, due to low levels of motivation or a poor understanding of these labels (e.g. Grunert et al., 2014, Larceneux et al., 2012, Tobler et al., 2011). For instance, the results of stage one of this project suggest that although labels relating to type of meat, fat content, origin and price are major factors that influence choice, other labels such as carbon footprint and production method labels play a secondary role in determining consumers’ choices of meat/meat substitutes. These findings are also supported by other authors who found that consumer preferences, willingness to pay and adoption of some of the sustainability-related labels (including carbon footprint, organic and animal welfare labels) is still relatively low (Peschel et al., 2016, Van Loo et al., 2014). Nevertheless, the impact that sustainability labels have on consumer choices and willingness to pay is largely dependent on consumer segments. Several studies have identified small segments of consumers (based on their knowledge, preferences, pro-environmental attitudes, psychometrics and demographics) for whom sustainability-related labels had a significant impact on their choices (Peschel et al., 2016, Koistinen et al., 2013). In the first stage of the project, we identified that meat consumption patterns can also influence the impact of these labels (Apostolidis & McLeay, 2016b). Our initial results supported the findings of Latvala et al. (2012) who argue that several reasons may lead to changes in meat consumption patterns, including healthiness, animal welfare and environmental reasons. The above discussion highlights the importance of investigating the differences in consumer preferences for sustainability food labels between groups of consumers with different meat consumption patterns.

From a signalling perspective, labels might complement but also compete with one another or with other signals, such as brands and prices, in their influence on consumer choices (Larceneux et al., 2012). This leads to the contradiction that, although consumers may have positive attitudes towards sustainability, they may not always choose the most sustainable products, as they focus on the labels representing the attributes they consider to be most important (Van Loo et al., 2014). In addition, choices are susceptible to various signalling biases associated with the quality inferences consumers make. Schuldt (2013) argues that food labels may create a ‘halo’ effect when positive information on a particular attribute results in positive inferences about other product attributes as well. For example, a product with a ‘locally produced’ label may be in very high demand, because consumers perceive it as a better-quality alternative than similar products without this label (Lee & Yun, 2015). Similarly, Leathwood et al. (2007) describe a ‘magic bullet’ effect, which occurs when consumers overgeneralise information on one product attribute, assuming benefits in other product attributes as well. For example, a consumer might inaccurately infer from a low-fat label that the product will also be low in calories or sugar. Furthermore, Balcombe, Fraser, and Di Falco (2010) suggest that consumers are more likely to use labels to avoid food products high in harmful ingredients (e.g. high fat content) than use labels to consciously identify healthier products (e.g. low fat content), which emphasises the importance of food labels for demarketing unsustainable and unhealthy products. However, the use of labels to demarket unsustainable products may cause a ‘boomerang effect’, as consumers may attach positive attributes to these ‘forbidden’ product labels (Van Kleef and Dagevos, 2015, Yakobovitch and Grinstein, 2016). For instance, higher fat content in meat may be associated with better quality and taste (Grunert, 2005, Scozzafava et al., 2016).

Given the inconsistencies in the results of previous studies that have explored food labels and consumer product choices, further research focusing on the effect of a range of signals is needed. In this paper, we contribute to the ongoing debate and calls for further research by comparing the preferences of meat eaters, as well as people avoiding or reducing meat, for sustainability labels.