Willingness to Pay for Optimal ‘Anjou’ Pear Quality

In light of increasing consumer demand for optimal fresh fruit quality, experience attributes are crucial to ensure repeated purchases and price premiums. Challenges in offering consistent quality throughout the marketing year make ‘Anjou’ pears an interesting case in which to analyze the effects of experience quality attributes on willingness to pay. We analyzed data from choice experiments conducted along with sensory tests at two different times during the marketing year. Results indicate that individuals are willing to discount between 15.43 cents/kg to 37.48 cents/kg for a one-unit increase in pear firmness and pay between 19.84 cents/kg to 24.25 cents/kg for a one-unit increase in soluble solid concentration. This information would help the industry assign priority to the factors likely enhancing the characteristics leading to repeated sales at price premiums. Also, this information is valuable to related sciences because it provides market information for focusing research portfolios on quality characteristics likely to increase pears’ commercial viability. Marketing of fresh fruits is evolving as consumers are becoming increasingly demanding of quality. Today’s breeders, postharvest scientists, and the industry in general have to pay closer attention to consumers’ quality demands to guarantee commercial viability. Nelson (1970) objectively defined quality by describing it in three dimensions: search (or appearance), credence, and experience. In the case of fresh fruits, quality of appearance is important because, in most cases, consumers rely on it to decide whether to buy. Moreover, standards and grades for fresh fruits are typically based on appearance characteristics such as size, color, shape, and external damage. Another quality dimension includes credence attributes, which require a third party verification because it is not possible to confirm their presence by simple inspection or after tasting the product. Examples of credence attributes include organic, sustainable, functional, locally grown, and so on. The third quality dimension is experience, which relies on individuals’ overall sensory perceptions, because attributes are difficult to observe by visual inspection alone. It is believed that experience attributes are determinant in assuring repeated food sales, because it is the quality of similar products bought in the past (that is, one’s previous experiences) that serve as an indicator of present and future quality (Shapiro, 1983). Also, several studies have demonstrated that optimal quality characteristics, as defined by consumers, are associated with willingness to pay premiums for fruit (Carew, 2000; McCluskey et al., 2007; Quagrainie et al., 2003). This article investigates the effects of eating quality characteristics on the value consumers place on fresh ‘Anjou’ pear quality. This is an important research question for the pear industry because the potential of increasing commercial viability is dependent on the industry’s capability of supplying consistent and optimal quality. Hence, information on consumers’ willingness to pay for quality attributes would help assign priority to the factors likely enhancing the characteristics leading to increased commercial viability. This information is also valuable for scientists such as fruit breeders and postharvest researchers to help them focus research efforts on quality attributes likely to be responsible for making a product successful in the marketplace. To accomplish the study’s objectives, we used data from two sensory tests in which pears under different post-harvest treatments were analyzed. Treatments consisted of applying varying amounts of ethylene, a naturally occurring hormone, to pears at two different times during the marketing season (December and March). Why ‘Anjou’ pears? Besides being one of the most popular pear varieties grown in the Pacific Northwest with an average production of 2.2 million metric tons per year (Washington Growers Clearing House, 2009), ‘Anjou’ pears present an interesting case in which to analyze the interactions between experience quality attributes and willingness to pay. First, the pear industry faces no small challenges to obtain and deliver consistent quality to consumers. Such challenges start at the orchard level and remain throughout the marketing chain (Kupferman et al., 2009). At the orchard, fruit firmness varies by location within the tree, harvest date, and orchard elevation. Challenges from the retail and distribution network are presented by retailers who prefer hard pears because there is less loss to marking or decay; however, this comes at the expense of having pears less appealing to consumers who want ready-toeat fruit. Second, there is the challenge of consistently supplying pears throughout the entire marketing year. Winter pears need to be put in cold storage (–1 C) for 2 to 8 weeks after harvest to induce ripening and prevent putting hard fruit in the market (Chen et al., 1996; Kappel et al., 1995; Kupferman, 1994; Shang Ma et al., 2000). This chilling period delays the entry into the market and likely deters sales (Schotzko et al., 2001; Shang Ma et al., 2000). A long-studied alternative is to induce and control ripening by using ethylene gas (Chen et al., 1996; Shang Ma et al., 2000). Despite the strong link between marketing decisions and final product quality of ‘Anjou’ pears, little empirical work on consumers’ values for pear eating attributes has been conducted. To our knowledge, only the studies conducted by Combris et al. (2007) and Zhang et al. (2010) elicited consumers’ economic values for pear eating quality. Combris et al. (2007) conducted experimental auctions along with sensory tests and found that consumers were willing to pay premiums ranging from 8 cents/lb for a oneunit increase in soluble solids concentration (SSC). Zhang et al. (2010) used results from a sensory taste with a contingent valuation questionnaire to measure consumers’ willingness to pay for pear eating quality. They found that on average, consumers were willing to pay 5.7 cents, 3.7 cents, and 8.5 cents more for a one-unit increase in the liking rate for pears sweetness, juiciness, and firmness, respectively. (The liking rate was measured Received for publication 15 Oct. 2010. Accepted for publication 14 Dec. 2010. We gratefully acknowledge Hayley Chouinard and Jill McCluskey from the School of Economic Sciences at Washington State University for their helpful comments that improved the manuscript. We also thank Chris Sater, former Associate in Research at the Tree Fruit Research and Extension Center at Washington State University, for her valuable help toward the completion of the manuscript. Finally, we thank the Pear Bureau Northwest for funding this research. Assistant Scientist and Extension Specialist. Postharvest Specialist. Sensory Program Manager. To whom reprint requests should be addressed; e-mail [email protected]. 452 HORTSCIENCE VOL. 46(3) MARCH 2011 on a 9-point hedonic scale with 1 being ‘‘extremely dislike’’ and 9 ‘‘extremely like.’’) Because liking rates are mostly subjective and highly variable across individuals, in this study, to measure the effects of quality characteristics on willingness to pay, we use physiochemical measures for firmness and SSC. Materials and Methods Sensory tests. Sensory tests were conducted in Dec. 2008 and Mar. 2009 at the Food Innovation Center, Oregon State University in Portland, OR. Participants for each test were recruited using an online screening questionnaire sent to 5000 consumers in the Portland metropolitan area. A between-subject design was used, because different consumers were recruited and randomly assigned to trials in December and March. In each, a planned sample size of 120 consumers [The standard sample size for a central location test of over 100 consumers was used and 20 extra individuals were recruited for each test to cover for last minute cancellations by participants (Meilgaard et al., 1999).] was selected to participate in the sensory test. Because our objective was to assess how pear eaters would define optimal quality, we based our sampling on the sociodemographic criteria given by the Pear Bureau Northwest for pear consumers [75% to 80% female, white, annual household income above $25,000/ year, 35 to 65 years of age, and post-secondary education (Moffitt, 2002)]. Compared with the city of Portland population (U.S. Census Bureau, 2000), our two samples of panelists over-represented the groups 45 to 54 and 55 to 64 years old, female, white, advanced college degree, and income range $75,000 to $99,000 (Table 1). Before being served, individuals were given a brief set of instructions on how to complete the test. Once in the sensory testing booths, panelists were presented with pear samples that were labeled with random threedigit codes representing different postharvest treatments. Just before being served, each pear was tested for firmness using a Fruit Texture Analyzer penetrometer (Güss Manufacturing, Stand, South Africa). One half of the fruit used for the firmness testing was juiced to determine SSC on each pear; the other half was served to panelists. Postharvest treatments. Several postharvest treatments were studied to measure their effects on pears’ final quality. The treatments used were selected based on the maturity of the fruit at the time of the experiment. In general, the longer pears are held in storage, the less time they need in a conditioning environment to ripen properly. Pears used in December were kept in controlled atmosphere (CA) storage for 70 d. Twenty-four h before conditioning, fruit was placed in a warm room (23 C). Conditioning was done at 23 C in shroud-covered box pallets using ethylene release canisters (ERCs) (Balchem Corporation, New Hampton, NY) and at least 50 ppm ethylene after 6 h. One of four treatments was applied to each test group: 1, 2, or 4 d with ethylene or no ethylene exposure. After treatment, all fruit was returned to cold storage (–1 C) to simulate transit to retail market, and 2 d before consumer evaluation, all fruit was removed from cold storage and held at 21 C until testing. Pears used in March were kept in CA for 169 d. Twenty-four h before being treated, fruit was placed in a warm room at 22 C. Like with the trial in December, treatment was done in shroud-covered box pallets at 22 C using ERCs and reached at least 50 ppm ethylene after 6 h. Five treatments were applied: 2 d in ethylene, 1 d in ethylene, 1 d in ethylene followed by 1 d in warm air, 2 d in warm air, and no treatment. After being treated, fruit was returned to cold storage (–1 C) for 7 or 8 d to simulate transit to a retail market. Three d before consumer evaluation, all fruit was removed from cold storage and held at 22 C until testing. Questionnaire. After tasting each sample, individuals were asked to rate overall liking, pear flavor, sweetness, juiciness, firmness, and texture liking using a 9-point hedonic scale with 1 at ‘‘dislike extremely’’ and 9 at ‘‘like extremely.’’ Also, individuals were asked about their purchasing habits such as price usually paid for pears, frequency of consumption, planned grocery shopping, and attitudes toward locally grown and organic fresh produce as well as demographic questions. The final step was a set of choice experiment questions to elicit the willingness to pay. Choice experiment. Participants were asked to answer hypothetical purchasing questions. In each question, individuals were asked to choose the option describing the pear sample they preferred out of five (December) or six (March) options. A ‘‘none’’ option was also included. See Figure 1 for an example of the choice question. Each option was randomly assigned prices obtained from grocery stores in Portland 1 week before each experiment. In the December trial, prices ranged from $1.49/lb ($3.28/kg) to $1.99/lb ($4.39/kg) in $0.10 or $0.20 increments. A fractional factorial design was used to randomize price and alternative matching. The design yielded 32 questions that were divided into four groups; thus, each respondent answered eight questions. For the trial conducted in March, prices ranged from $1.39/lb ($3.06/kg) to $2.19/lb ($4.83/kg) in $0.20 increments. The fractional factorial design yielded 25 questions that were divided randomly into two groups of 12 and 13 questions each. Model specification. We assume a random utility model in which the i individual’s utility from choosing treatment j is given by: Uij = V ij + eij 1⁄21 where Vij is the deterministic portion of the utility and eij is the identically independent distributed stochastic error component. In this study, Vij is represented by: V ij =l1ij + l2Firmness ij + l3SSC ij + b Pr ice ij 1⁄22 where l1ij is the alternative specific constant denoting the utility derived for choosing pears under treatment j relative to not choosing any Table 1. Comparison between sensory taste participants and Portland demographics. Demographics Dec. 2008 (N = 111; %) Mar. 2009 (N = 120; %) US Census Portland Population = 529,121 (%)