How does animal welfare taste? Combining sensory and choice experiments to evaluate willingness to pay for animal welfare pork

Highlights

• Combining sensory and choice experiments to evaluate the importance of animal welfare.

• Information regarding animal welfare affects hedonic liking and willingness to pay.

• Consumers rated the organic product higher than those with animal welfare labels.

• General, but not specific information on husbandry practices affects preferences.

Abstract

Although consumers show considerable interest in higher animal welfare, the market share for such products remains low. To provide consumers with a choice regarding different levels of animal welfare, animal welfare labels have been developed. Such product options are intended to serve as intermediate products between conventional products and more expensive organic options that are often associated with high animal welfare standards. This study aims to evaluate whether information about animal welfare practices affects consumers’ liking and willingness to pay (WTP) for a pork product. Using a within-subject design with three evaluation rounds (blind, expectation, and full information) we combine hedonic liking where subjects rate four different types of ham with a choice experiment. Hams differ in the animal husbandry conditions during the rearing and fattening process: conventional rearing, a specific treatment of mother sows, a general animal welfare label, and organic production. Results show an effect of information on consumers’ sensory evaluation of the different products, although products were evaluated to be similar in the blind condition. Consumers rated the organic product higher than those with animal welfare labels while the conventional option had the lowest liking scores. Results from the choice experiment concur with the hedonic rating. Estimates indicate that consumers are willing to pay more for ham bearing a general animal welfare or organic label in the expectation and full information condition. Consequently, information regarding animal welfare affects both consumers’ hedonic liking and WTP for ham.

Introduction

Farm animal welfare continues to be a salient issue for European citizens. Ninety-four percent of Europeans indicated protection of farm animals to be important (European Comission, 2016). Although the concern about animal welfare seems to be high, the share of products which declare an improved animal welfare is rather low. For instance, 95% of German respondents perceive the protection of farm animals to be important (European Comission, 2016), but the market share for organic products in Germany is only 5.7% (GfK, 2017) and below 2% for meat and meat products (Bundesanstalt für Landwirtschaft und Ernährung, 2020). Organic livestock production requirements in Germany include, but are not limited to, access to the outdoors, sufficient space, and appropriate housing, which has been shown to impact animal welfare (Federal Ministry of Food and Agriculture, 2020). Previous research has also shown that consumers associate organic with higher animal welfare (Baker et al., 2004, Bundesanstalt für Landwirtschaft und Ernährung, 2020, Von Meyer-Höfer et al., 2015). Furthermore, animal welfare has been cited as a significant reason for consumers to purchase organic animal products (Harper and Makatouni, 2002, Honkanen et al., 2006, Hughner et al., 2007, Padilla Bravo et al., 2013, Zanoli et al., 2013). Nevertheless, the small market share could be caused by the price premium of organic meat products compared to conventional products, which may be too high for consumers (Aschemann-Witzel and Zielke, 2017, Schröck, 2013, Spiller, 2001). Although it has been found that consumers are willing to pay more for organic, price has been cited as one of the deterrents to consumers adopting organic items (Buder et al., 2014, Nasir and Karakaya, 2014, Yiridoe et al., 2005). Therefore, de Jonge, van der Lans, and van Trijp (2015) suggest the generation of compromise goods to help increase the consumption of animal friendly products. According to the authors, compromise goods are products which provide a higher animal welfare standard but are below the organic label requirements. Consumers then have the possibility to change from conventional meat to an animal welfare labeled meat with moderate expenditure changes. Furthermore, their study showed a high-level animal welfare label is able to attract organic and meat replacement consumers. This ability of compromise goods to draw different consumer segments can help increase choice to consumers, improve animal conditions, and raise profits for the producer.

Another reason for the discrepancy between concern and purchase behavior could be that consumers may forego buying animal welfare products because other more important product attributes of animal welfare products are not up to their standards. Consumer quality expectations and acceptance of food and beverages are formed due to a complex combination of cues and factors. Intrinsic sensory characteristics (e.g. taste, texture) of products play an important role in the consumer purchase decision. External cues, such as brand and certification labels, also affect how consumers perceive product quality. Extrinsic cues that are not able to be confirmed after consumption are known as credence attributes (Darby & Karni, 1973). In terms of meat products, information regarding rearing conditions would be considered such a credence attribute. Based on these attributes consumers form expectations that inform their purchasing decision. The purchasing environment, personal characteristics and prior experience with the product are also factors that have been shown to affect expectations (Deliza and Macfie, 1996, Piqueras-Fiszman and Spence, 2015, Steenkamp, 1990). Product liking expectations and the actual experience may or may not be matched. Once a consumer has evaluated the product a confirmation or disconfirmation of expectations occurs. Disconfirmation is measured as the acceptance difference between the expected liking and the blind assessment of the product. When the product is better than expected there is a positive disconfirmation and when it is worse than expected a negative disconfirmation occurs (Cardello and Sawyer, 1992, Deliza and Macfie, 1996). If sensory expectations are not met due to a negative disconfirmation, the consumer may not choose to repurchase the product (Deliza and Macfie, 1996, Lange et al., 1999), although there is limited evidence that animal welfare and nutrition information can improve acceptance scores despite a negative disconfirmation (Napolitano, Caporale et al., 2007). Therefore, it is important for producers to be aware of how extrinsic cues can affect consumer expectations in order to ensure their product is well-received and viable.

There has been extensive research investigating the effect of external information on consumer liking and expectations for a variety of foods and beverages (Asioli et al., 2017, Fernqvist and Ekelund, 2014, Hersleth et al., 2011, Piqueras-Fiszman and Spence, 2015). A common methodology to study the effect of information is to compare evaluations of products before and after information is given to consumers. Some studies also incorporate an expected liking round before the information round. Several studies have evaluated the interaction of consumer liking and organic claims. Many findings show organic labels to have a positive impact on consumer acceptance for products including bread (Kihlberg, Johansson, Langsrud, & Risvik, 2005), beef (Napolitano, Braghieri et al., 2010), and wine (Apaolaza, Hartmann, Echebarria, & Barrutia, 2017). However, little work has explored the impact of animal welfare information on consumer liking scores of animal products. Carlucci, Monteleone, Braghieri, and Napolitano (2009) found consumers to prefer yogurt that was made with milk from cows with more freedom of movement and cleanliness. Another study by Musto, Cardinale, Lucia, and Faraone (2015) showed increased hedonic expectations for milk from goats in semi-extensive conditions, which are associated with higher animal welfare, compared to goats in intensive conditions. Specific rearing conditions for meat have been assessed in only a few studies. Research by Napolitano et al., 2007, Napolitano et al., 2007 reported that information about animal welfare increased expected and actual liking scores for beef and lamb. Dransfield et al. (2005) found that consumers preferred pork labeled as originating from their own country and raised outdoors.

In addition to examining consumer acceptance, it is important to understand whether and how much consumers are willing to pay for animal products that follow higher animal welfare standards because providing better animal welfare can be costly for producers. Some previous studies have found willingness to pay (WTP) estimates for animal welfare (Gerini et al., 2016, Lagerkvist and Hess, 2011). Norwood and Lusk (2011) showed that more space and outdoor access significantly increase the WTP for pork chops. Interestingly in the case of pig gestation crates, a label indicating non-usage has a higher WTP than a full ban of gestation crates (Tonsor, Olynk, & Wolf, 2009). Furthermore, a study by Gracia, Loureiro, and Nayga (2011) showed consumers are willing to pay more for a comprehensive animal welfare label for ham rather than for labels referring to particular attributes such as housing conditions. Studies have shown that the value of animal welfare varies across products and also differs in the potential to deliver an added value compared to other product attributes. For instance, Dahlhausen, Rungie, and Roosen (2018) found that certified antibiotic free production and local production for all examined products (eggs, pasta made with eggs, and pork minute steaks) were more valuable than an animal welfare label. However, animal welfare was valued more by the consumers than organic in the case of pasta and pork minute steaks.

There is also a sub-set of literature that has combined sensory evaluations with an economic experiment to assess how information affects liking and consumer WTP. Many of these studies use auctions (Cardoso et al., 2013, Costanigro et al., 2014, Gallardo et al., 2018, Teuber, Dolgopolova, & Nordström, 2016), conjoint analysis (García-Torres, López-Gajardo, & Mesías, 2016), or hypothetical discrete choice experiments (Bi et al., 2015, Enneking et al., 2007). Overall, these studies show that information can impact consumer sensory ratings and WTP. There are some examples of non-hypothetical binding choice experiments where participants receive one of the chosen products to reduce hypothetical bias as participants will be more motivated to accurately reveal their preferences (Aoki et al., 2010, Chen et al., 2013, Gracia, 2014, Loomis, 2014, Lusk and Schroeder, 2004, Moser and Raffaelli, 2012). However, only a few studies have also included hedonic ratings in addition to the non-hypothetical choice experiment. For instance, Roosen, Marette, Blanchemanche, and Verger (2007) combined preference ratings with an economic choice procedure using endowment constraints and participants received the products from one randomly selected choice occasion. Lange, Issanchou, and Combris (2000) also gave participants the opportunity to take their choice in one randomly selected choice set home as compensation for their participation after they had evaluated appearance and taste of different orange juices. More recently, Kallas et al. (2019) examined products coming from indigenous pig breeds and made with additional health related ingredients using a non-hypothetical choice experiment and hedonic evaluations for expected and actual liking.

While previous research has examined the effect of animal welfare on consumer acceptance and WTP for a variety of dairy and meat products, to our knowledge no previous work has compared the consumer liking and WTP of conventional, specific rearing conditions, overall animal welfare, and organic production methods. Furthermore, using a binding discrete choice experiment combined with sensory evaluations for three rounds (blind, expectation, and full information) has not been done previously. The objective of our study is twofold. First, we want to evaluate if there are subjective differences of sensory characteristics between conventional, organic, and animal welfare meat products. Secondly, we aim to determine whether information about animal welfare practices affects consumers’ sensory liking and WTP for pork using ham as a representative product. For this study, consumer panels with sensory evaluations of four types of cooked ham varying in animal welfare standards were conducted. Two levels of animal welfare were examined: non-fixation of the sow and overall animal welfare, in addition to an organic product and a conventional (without label) product as the reference product. Sow fixation was chosen to be of interest in regard to animal welfare because some consumers perceive sow fixation to have a negative impact on the sow’s well-being and according to Grunert, Sonntag, Glanz-Chanos, and Forum (2018), 41% of German respondents stated non-fixation of the sow to be an important attribute when buying pork. A within-subject design was implemented with participants evaluating the ham in three stages: blind assessment, expected liking after receiving information on sow husbandry practices, and re-evaluation with the information given in the second round. A non-hypothetical discrete choice experiment was used to estimate the WTP for ham samples after each stage. Our study helps to gain further insights into how consumers perceive varying degrees of animal welfare in relation to conventional and organic production using a newer approach to sensory and consumer science research.