The role of brand utilities: application to buying intention of fair trade products

Abstract The authors focus on the ethical consumption and propose a model of buying intention of fair trade products, including the utilities of the fair trade generic brand as direct determinants. The authors measure the functional and symbolic utilities provided by this brand, together with the attitude towards the commercialising organisations, consumer concern and perceived knowledge about fair trade issues. The model is tested through a structural equation model on a sample of members (students, lecturers and staff) of a ‘Fair trade University’. The results confirm that perceived functional utility is the most important antecedent of the buying intention, while the symbolic dimension has a significant but weaker explanatory power. Conversely, the consumer attitude towards the organisation has no influence. The authors also highlight the importance of communication and concern to stimulate consumer behaviour.


Introduction
It has long been observed that consumers are increasingly interested in the social and ethical components of the products (Andorfer & Liebe, 2015). In this sense, ethical trade initiatives, such as fair trade, have arisen in the context of economic globalisation, and they have acquired considerable prominence in recent years (Bezençon & Blili, 2010). Fair trade is an alternative market approach that aims to improve the well-being of small producers in developing countries (Randall, 2005). It implies a long-term and direct relationship with the producers, with a price for the products that covers the production costs and a premium that organisations can reinvest either in business or in social and environmental schemes amongst the wider community (Wright & Heaton, 2006). This alternative trade experienced a period of expansion in the early 1990s in Europe, Japan and North America (Wright & Heaton, 2006), and currently the fair trade products sales keep in constant increase (Yamoah, Duffy, Petrovici, & Fearne, 2016). However, its market share remains very small compared to traditional commodity networks, leaving much room for market growth (Schollenberg, 2012). This fact makes specially interesting the study of the ethical purchase. It is necessary a better understanding of consumer buying behaviour, in order to engage the interest of the mass market and attract more and more new consumers (Cailleba & Casteran, 2010).
Academic literature about fair trade is growing. From the quantitative perspective, it is common to find different versions of hierarchical beliefs-attitudes-behaviour models, supported on the Theory of Planned Behaviour (Ajzen, 1985) (Shaw, Shiu, & Clarke, 2000;Shaw & Shiu, 2002Ozcaglar-Toulouse et al., 2006 andYamoah et al., 2016), as well as models inspired in this theory but incorporating new constructs with potential influence on the buying intention (De Pelsmacker & Janssens, 2007;Kim, Lee, & Park, 2010). These authors highlighted the lack of understanding on what determines consumers' responses to fair trade products, and the need of analysing new proposals. To shed further light on this issue, we take as a basis the De Pelsmacker and Janssens (2007) and Kim et al. (2010) works and propose a new model, which includes constructs that have hardly been measured in this field empirically.
In the first place, we consider the consumer-based brand equity. Taking into account the works of Kamakura and Russell (1991), Cobb-Walgren, Ruble, and Donthu (1995) and de Chernatony (1993), Del Río, Vázquez, and Iglesias (2001) and Vázquez, del Río, and Iglesias (2002) define consumer brand equity as the overall utility that the consumer associates to the consumption of the brand, including associations expressing both functional and symbolic utilities. The functional utility is linked to 'doing good' (de Chernatony, 1993) and it meets the practical needs of consumers, whereas the symbolic utility refers to the emotional evaluation of the brand.
The consumer-based brand equity, and specifically its utilities, have not been measured in the fair trade realm. With regard to this, fair trade mark has proved to be very effective in generating overall brand awareness (Nicholls & Lee, 2006). Opposite the little or no coherence across specific brands, fair trade becomes the only consistent feature on products (Nicholls, 2002), and the one that distinguishes them from their competitors (Castaldo, Perrini, Misani, & Tencati, 2009). Consumer perceptions about it may serve as a halo for the attributes of the products, influencing consumers' responses (Kim et al., 2010), so it is advisable to promote it as a brand or meta brand (Bezençon & Blili, 2010;Griffiths, 2012;Wright & Heaton, 2006), becoming a master brand, that is, the umbrella for various products offered (Saunders & Guoqun, 1997).
The fair trade brand has an identification function; at the same time, it has many possibilities to generate functional and symbolic utilities to consumers, that is important to measure. Functional issues related to fair trade products and its commercialisation have been studied (De Pelsmacker & Janssens, 2007;De Pelsmacker, Janssens, Sterckx, & Mielants, 2006;Kim et al., 2010;Yamoah et al., 2016), but no previous research has measured the functional utility of fair trade brand as a multidimensional concept. With regard to symbolic utilities, related to psychological issues linked to the buying, to the best of our knowledge they have not yet been included in models of buying intention. Against this background, this research aims to provide the two first contributions to the academic literature on this topic: (1) to measure fair trade brand utilities, both functional and symbolic; (2) to test empirically their direct effect and importance on the buying intention.
In the second place, we study the influence of the attitude towards the companies that sell fair trade products. In general terms, past research suggests that corporate evaluation can be separated from product evaluation and that the attitude towards the corporate brand has a direct and positive influence on behavioural intentions (Goldsmith et al., 2000;Lafferty, Goldsmith, & Newell, 2002) and product evaluations (Berens, van Riel, & van Bruggen, 2005;Keller, 1993). In this sense, Kim et al. (2010) analysed a for-profit company (Starbucks) and showed its influence on the buying of fair trade products depending on the country analysed. However, the fair trade movement is mainly characterised by networks of relatively small companies, non-profit organisations (NPO), rather than large, profit-making corporations. Llopis (2007) found that one key factor influencing the consumption of fair trade products is trust in these social entities, but this issue has not been included in intention models of these type of products. Its possible impact on the perception of brand utilities have not considered either. Therefore, the third contribution of the article is to analyse the possible influence of consumer attitude towards the commercialising NPO on the intention model. Finally, the model is completed with the inclusion of consumer concern or general attitudes towards the fair trade issue (Castaldo et al., 2009;De Pelsmacker & Janssens, 2007;Shaw & Shiu, 2002;Shaw et al., 2000), and the perceived knowledge (De Pelsmacker & Janssens, 2007). Knowledge has been identified in qualitative studies as a very important issue to promote consumption and influence on attitudes (Bray, Johns, & Kilburn, 2011;Wright & Heaton, 2006), so it must have a key role in the intention model (De Pelsmacker & Janssens, 2007).
The challenge to expand the consumption of these products to mass market, not only to the niche market of ethical buyers, implies several challenges. Thus, together with the academic contribution, the study of these variables will let us to identify motivators and barriers to the buying, an information that can be useful to fair trade products sellers when planning their marketing strategies.

Buying intention of fair trade products
One research line in fair trade field is the analysis of consumer behaviour. Specifically, it is highlighted the development of models of ethical purchasing behaviour that have incorporated into Ajzen (1985) Theory of Planned Behaviour (TPB) constructs of 'ethical obligation' and 'self-identification with ethical issues' (Shaw et al., 2000;Shaw & Shiu, 2002Ozcaglar-Toulouse et al., 2006) or 'self-direction and universalism values' (Yamoah et al., 2016) in an attempt to better understand the phenomenal growth in the Fair trade products market. The inclusion of these constructs assumes that people who buy these products are ethically minded consumers, that is, they are concerned with ethical issues and feel obligated to buy Fair trade products (Carrington, Neville, & Whitwell, 2010). However, these purchases are not driven by ethical concerns alone (Yamoah et al., 2016) and the inclusion of additional variables is not only empirically pertinent, but also conceptually desirable in this behavioural context (Shaw et al., 2000).
With regard to this, De Pelsmacker and Janssens (2007) and Kim et al. (2010) analyse the antecedents of buying behaviour including new variables. For example, whereas the construct of Perceived Behavioural Control of previous works includes items related to problems which may affect the buying, such as availability, range, location of retail outlets, price or availability of information, De Pelsmacker & Janssens (2007) analyse each issue separately, that is, they study constructs like Quality of Information, Quantity of Information, Convenience, Product interest, Product likeability and Price acceptability. These authors include both positive and negative attitudes towards fair trade (concern and scepticism), whereas Kim et al. (2010) analyse the influence of fair trade corporation evaluation in the buying intention (see Table 1).
We support on De Pelsmacker and Janssens (2007) and Kim et al. (2010) to propose our buying behaviour model, which includes the measurement of the fair trade generic brand equity (brand utilities) and the attitude towards the commercialising entities as main antecedents. The details of the model will be explained in the following sections.

Brand utilities for consumers
Brand equity is a core concept of marketing. An extensive research has been conducted on the topic over the last decades, resulting in highly diverse definitions and in a great variety of methods and proposal to measure it (Buil, de Chernatony, & Martínez, 2013;Vázquez et al., 2002). Against the financial dimension, the research has tended to focus on the consumer perspective. Consumer brand equity denotes the added value endowed by the brand to the product (Keller, 1993). This definition is related to brand name utilities, which are associations added to the product thanks to its brand name (Del Río et al., 2001). The theoretical and empirical literature on this issue suggests classifying the utilities according to two basics dimensions: the functional and the symbolic one (De Chernatony & McDonald, 2003;Vázquez et al., 2002).
On the one side, functional utility is associated with 'doing good' (de Chernatony, 1993), it meets the practical needs of consumers and has a guarantee function. It is linked to brands with the ability to offer products that meet the market needs (Keller & Aaker, 1992), good quality, convenience (Ambler, 1997) and value for money (Aaker, 1996).  Shaw et al. (2000), Shaw & Shiu (2002 ethical obligation, self-identification with ethical issues, Subjective Subscribers to the uk ethical consumer magazine norm, attitude, Perceived Behavioural control, Purchase intention On the other side, symbolic utility refers to the emotional evaluation of the brand, linked to intangible issues, such as personal and social identification (Vázquez et al., 2002). Product ownership and use help consumers define and live out their identity (Mittal, 2006). The act of buying can become an act of expression and projection of oneself, necessary for consumers' personal definition (Belk, Bahn, & Mayer, 1982). It lets consumers communicate their link to certain social groups, values and personal features (Keller, 1993). Brands act as communication instruments, fulfilling a social identification function by allowing consumers to manifest the desire to integrate themselves with or dissociate themselves from the groups that make up their closest social environment. Consumers will positively value those brands with a good reputation among the groups to which they belong or aspire to belong (Long & Schiffman, 2000). These needs are linked to the maintenance of self-esteem, understood as the motive to seek experiences that enhance or protect the self-concept and the acceptance by significant others, with approach and avoidance behaviours (Banister & Hogg, 2004), for example rejecting products or brands with negative imagery (Sirgy, 1982).
The special characteristics of fair trade products make interesting the study of brand utilities. Their purchase is related to the ethical buying, in the sense that fair trade generic brand provides people with a guarantee that producers have not been exploited (Alexander & Nicholls, 2006), among other social issues. Given its strong ethical component, it is expected that emotional or psychological issues have a relevant role in the model, that is, the personal identification with the values of the cause, social identification and self-esteem associated with the purchase of products with fair trade brand will have a certain degree of influence on the buying. However, the lack of empirical works measuring these issues lead us to consider an unanswered question the role of symbolic utility (versus the functional utility) in the intention model. On the side of functional associations, the ethical issue explains the premium pricing of many fair trade products and the need for greater reasoning to explain it (Wright & Heaton, 2006). Furthermore, consumers found availability of fair trade products to be limited and insufficient (Shaw & Clarke, 1999), which may be a barrier to purchase. Finally, its quality has been traditionally called into question (Obermiller et al., 2009) so quality of a product is needed to come before any efforts to communicate the fair trade message (Wright & Heaton, 2006). Quality, convenience and value for money are three key issues of functional utility, particularly important in this type of products, that should be globally taken into account to explain the buying intention.

Brand utilities and attitude towards organisations
Functional utility is based on objective characteristics, thus it is expected that this dimension influences the buying intention directly and positively. Several studies analyse specific variables of this type of utility. With regard to the price, there are a lot of empirical research whose goal has been identify how much more individuals are willing to pay for a fair trade product (Salvador, Merchant, & Alexander, 2014), highlighting the power of price to attract new consumers (Hainmueller, Hiscox, & Sequeira, 2015). In this sense, researchers and practitioners should not forget that although consumers state that they would pay a premium for socially responsible products, they will only purchase the products if they perceive them to be of high quality (McCluskey & Loureiro, 2003). Thus, product likeability is determinant of the buying behaviour (De Pelsmacker & Janssens, 2007). Alexander and Nicholls (2006) establish that to grow their market share beyond the population that constitutes their natural consumers, it is necessary for companies to support the fair trade positioning on quality and differentiation, focusing on new product development and increasing the retail availability of products. Therefore, all the aspects included in the functional utility of the fair trade generic brand are key factors to encourage consumers to buy. That is: H1: The functional utility of the fair trade generic brand influences buying intentions positively On the other hand, symbolic utility refers to satisfying the needs of the psychological and social environment, for example communicating desirable impressions about consumers to others and helping consumers to live out their self-concept (Vázquez et al., 2002). Specifically, the greater the consistency between the brand image and the consumer's self-image, the greater the consumer's intention to buy the product (Hogg, Cox, & Keeling, 2000). In the same way, self-esteem linked to the purchase is an important motivational driver for consumption, involving both the acceptance and the rejection/avoidance of products and brands (Banister & Hogg, 2004).
While some people question the quality or functional attributes associated with the fair trade products (Bray et al., 2011;De Pelsmacker & Janssens, 2007;Wright & Heaton, 2006), the social benefit of products is beyond doubt (Nicholls & Lee, 2006;Randall, 2005). Thus, the buying of these products can have a strong symbolic component. An individual's values are likely to play an influential role in shaping aspects of her or his decisions to purchase or support fair trade products (Salvador et al., 2014). According to Varul (2009), buying these products is not only a means of 'doing good' but also a way of expressing consumer identity as a moral person ('being good'). With their purchase, consumers express their concerns about mass consumption and distinguish themselves from conventional shoppers. Consumers can then enrich their self-image and transmit information to others through the images of the brands that they buy, reinforcing their self-esteem. Based on these ideas, it is proposed that: Fair trade products are commercialised by importing organisations; it is then interesting to analyse the role of the attitude towards these organisations in buying intention. Specifically, attitude is related to the corporate image; it is an overall evaluation of the company, a subjective and intangible judgement that includes feelings and associations (Barich & Kotler, 1991). Corporate associations might influence product imagery, in the sense that a good view of an organisation would lead to an overall positive evaluation of its products and would favour the buying intention (Brown & Dacin, 1997;Saunders & Guoqun, 1997;Brown, 1998;Berens et al., 2005). With regard to this last issue, Llopis (2007) establishes that one factor influencing the consumption of fair trade products is consumer trust in NPOs. This relationship has been analysed in the fieldwork of Kim et al. (2010), although they considered the case of a for-profit company and obtained different results depending on the country analysed. Thus, we propose that: H3: The attitude toward the organization that commercializes fair trade products influences positively on (a) functional utilities; (b) symbolic utilities; (c) buying intentions

Consumer concern and perceived knowledge of fair trade issue
With regard to ethical buying intentions, it is believed that general attitudes towards fair trade will lead to more specific consumption-related attitudes (De Pelsmacker & Janssens, 2007). According to these authors, one dimension of consumer attitude is concern, which measures respondents' support for the fair trade issue. This variable influences the cognitive and behavioural processes of the consumer. In fact, inherent in the purchase of a fair trade product is the consumer's concern that producers in developing countries receive their fair compensation for what they produce (Doran, 2010). Concern leads to a greater perception of the attributes of a product and greater interest in its characteristics (Bloch & Richins, 1983). In this sense, most of the participants in the study of De Pelsmacker and Janssens (2007) stress that their concern about fair trade issue would enhance their interest in these products and would lead them to pay less attention to inconvenience and high prices. Furthermore, those people concerned about ethical issues may be guided by a sense of ethical obligation to others and self-identification with these issues (Shaw et al., 2000). Therefore, consumer concern would lead to better perceptions of the functional and symbolic utility of the fair trade generic brand. Moreover, it is expected that highly concerned consumers have more elaborated evaluations of the NPOs that commercialise fair trade products. Consequently, they will evaluate them more favourably. That is: H4: Consumer concern influences positively on (a) the functional utility of the fair trade generic brand; (b) the symbolic utility of the fair trade generic brand; (c) the attitude toward the fair trade organization.
Finally, in models of buying intentions, knowledge has an impact on attitudes, which in turn have an impact on behaviour (McEachern & Warnaby, 2008). In the case of fair trade products, it is assumed that better knowledge leads to more positive attitudes towards the issue and product-specific characteristics (De Pelsmacker & Janssens, 2007). The challenge is to gain consumer understanding of the issues associated with fair trade because the level of knowledge heavily influences the process of the formation of buying intentions (Kim et al., 2010). According to De Pelsmacker, Janssens, Sterckx, & Mielants (2005), almost half of the respondents felt they did not have enough information to be convinced, so they did not buy fair trade products. In the same line, one of the reasons offered by most people for their inaction is ignorance of how the system works (Castaldo et al., 2009). The research of Bray et al. (2011) leads to the same conclusion, that is, the without prominent communication of these issues, lack of knowledge would continue to limit ethical consumption. Among the reasons that explain this relationship is that a consumer who is becoming more familiar with one particular issue will have a different frame of reference for evaluations than a consumer exhibiting a low level of familiarity (Söderlund, 2002). Roughly speaking, the high-familiarity consumers have encountered the fair trade issue more often, and therefore they have a larger pool of evaluations stored in their memory. This fact will influence positively their concern, the utilities of the fair trade generic brand and the attitudes towards the NPOs involved in this trading. With regard to this last issue, Llopis (2007) shows that people with a higher level of knowledge trust NPOs to a greater extent. Therefore, it is proposed that:

Methodology
The research focuses on the Spanish market. This market still has great potential for growth in this country, representing an annual per capita consumption of €0,71, far below the European average of €11,43 (National Fair Trade Coordination Committee). Despite this difference, the sales have not stopped growing since the year 2000, the use of the Fairtrade mark certification being an important stimulus in this development. Taking into account the fact that the fair trade system is still emerging in Spain, we decided to carry out the study in an environment in which there are frequent awareness-raising campaigns and points of sale of fairly traded products throughout the year. Specifically, the NPO IDEAS, with the support of the European Commission and the Spanish Agency for International Development Cooperation, coordinates the 'Fair Trade University' initiative. A university can achieve this status if: (a) it approves an institutional statement favourable to fair trade; (b) it uses fairly traded products on an institutional level; (c) it offers fairly traded products on the university campus; (d) it promotes and raises awareness of fair trade issues in the university; and (e) it creates a work group about fair trade. In Spain, there are eight 'Fair Trade Universities' , and this empirical study was carried out in one of them.
A web survey was published and the link was emailed to the university community in March 2014. The size of the final sample was 292 valid cases, once we had removed all the incomplete surveys (65.4% students, 24.3% lectures and 10.3% staff and 52,3% women and 47,7% men). The most commonly consumed category of fair trade products was coffee, with a total of 54.5% of respondents reporting its eventual consumption in the past.
The hypotheses were contrasted with a causal analysis studying the relationships among the variables through a structural equation model (SEM). All the variables were measured with 10-point Likert-type scales (see Appendix 1 with items and source of the scales). Specifically, we propose that functional utility included issues related to the product, perceived value and convenience. With regard to symbolic utility, we include items related to consumer personal identification, social identification and self-esteem (scale adapted from Ellemers, Kortekaas, & Ouwerkerk, 1999). Regardless of these constructs, and to avoid consumer confusion, an explanation about the meaning of the Fairtrade mark was included in the questionnaire, indicating that it is a 'brand' that identifies products that have been fairly produced and traded.
With regard to the attitude towards the organisation, we asked the respondents to select a entity that commercialised fair trade products in the region. A total of 71.8% respondents selected Oxfam Intermon, 11.8% selected Espanica and the rest of the respondents selected other minority institutions.

Perceived functional and symbolic utilities
To fulfil the first objective of the research, we carried out two confirmatory factor analyses with maximum likelihood estimation (Table 2). In the first place, we evaluated the global model applying absolute, incremental and parsimonious measures of fit. We also examined the measurement model by individual factors, confirming the statistical significance of each loading obtained between the indicator and the construct. With these measures, they analysed the convergent and discriminant validity of the proposed model (Hair, Black, Babib, & Anderson, 2010).
In the case of functional utility, the model presented high goodness-of-fit measures and convergent validity of the dimensions, with all the standardised lambdas above 0.5 (Steenkamp & van Trijp, 1991). In the case of symbolic utility, the goodness-of-fit measures were right and all the items were significant at the confidence level of 95%. However, the standardised lambda coefficient for the item 'fair trade products (with the fair trade generic brand) are linked to a certain type of people; but they are not for me (r)' was below 0.5 (0.26). Thus, this item was removed. Once the model was reformulated, it was possible to confirm its convergent validity and to obtain an adequate specification of the proposed factor structure. To confirm the discriminant validity, we followed the procedure described by Anderson and Gerbing (1988), in which the confidence intervals for the correlation of the constructs were estimated and compared with 1. In none of the cases did the intervals contain the value 1, so the proposed measurement model was right. Finally, the reliability of the measurement scales was evaluated using the Cronbach's alpha. All the factors exceeded the minimum recommended value of 0.7, confirming the internal reliability of the proposed constructs (Hair et al., 2010).
These analyses led us to undertake a second-order factor analysis. The second-order analysis is a statistical method employed to confirm that the theorised construct in a study loads into certain number of underlying sub-constructs or components. It allowed us to confirm the multidimensionality of functional and symbolic brand utilities and to determine the weight of each factor in the global constructs (Figure 1). The factor loadings between the second-order factor and the proposed dimensions were statistically significant in both cases, and the goodness-of-fit measures also indicated that the model was correct. Therefore, it is possible to conclude that, from the consumers' perspective, functional and symbolic utility of Fairtrade brand exist and may influence consumer behaviour. As far as the research hypotheses are concerned, and following Anderson and Gerbing (1988), we first confirmed the reliability and validity of all the constructs of the causal model by means of a confirmatory factor analysis (Table 3).
All the results led us to accept its adequacy. As it can be observed, the most explanatory variable of functional utility is perceived value. In the case of symbolic utility, the personal identification and self-esteem linked to the purchasing have the greatest weight, the social identification/reputation of the fair trade brand less important.
Subsequently, the structural model was estimated. The indices of goodness of fit of the model to the data were correct, as the statistics exceeded or came close to the optimal values ( Figure 2). It was also observed that all the hypotheses were accepted, except H3b, H3c and H5d. That is, buying intentions are explained mainly by the fair trade brand's functional utility (H1 accepted). The symbolic utility provided by the brand followed (H2 accepted). The attitude towards the NPO has not a direct effect on consumer buying intentions (H3c rejected). These findings show the importance of the associations of the fair trade generic brand, above all the functional issues, to explain the buying behaviour. The attitude towards the social entity has not influence on the symbolic associations either (H3b rejected) but it does enhance functional utility (H3a accepted). Thus, the global evaluation of the NPOs does not affect directly on the buying, but indirectly. When the social entities that market the fair products are well valued, the functional associations of products improve, which makes buying easier.
It is also relevant that consumer' knowledge is a key variable in improving their perceptions about the fair trade generic brand, in the sense that greater knowledge about fair trade leads consumers to feel greater concern about its cause and it improves their perceptions   Bi-concern 0.52 fu-concern 0.50 Su-Pk 0.83 of the brand (H5a, b and c accepted). However, the relationship between knowledge and attitude towards the NPO is not significant (H5d rejected). This relationship is indirect, through the higher concern that provokes the knowledge. Finally, the fair trade concern influences the perceived functional and symbolic utilities directly and positively, and also determines the attitude towards the NPO (H4a, b and c accepted).

Conclusions
Fair trade has great growth potential and faces the challenge of reaching not only 'ethical consumers' but also the mass market. For this purpose, marketing strategies are necessary because ethical consumers buy intangibility, justice and perhaps conscience (Bezençon & Blili, 2010), but the mass market needs more than this. According to Nicholls and Lee (2006), ethical products such as fair trade are competing with well-established brands whose owners are able to focus significant effort on brand-building activities. Therefore, the intention to purchase this type of product in preference to a usual branded option is complex. This work tries to provide insights into consumer behaviour in the fair trade field, something important to identify ways to boost its growth, engage the interest of the mass market and attract more and more new consumers (Cailleba & Casteran, 2010). Specifically, a quantitative study was carried out to test a causal relationship sequence (knowledge -associations/attitudes -buying intention).
In this model, an important concept was brand utility. There are many specific brands of fair trade products, many of which are unknown to the consumer, but they may have something in common: the fair trade generic brand. It is advisable to develop this mark as a brand (Wright & Heaton, 2006), which serves as a guarantee and communicates that products have been fairly produced and traded. To encourage the purchasing of these products, it is necessary to act in relation to the two main utilities that a brand provides: the functional and the symbolic one. First, we confirmed that functional utility includes three dimensions, related to product characteristics, perceived value and convenience. The most important variable is the perceived value, which reinforce the importance of the price for consumers (De Pelsmacker et al. 2005). The higher price of fair trade products should not be a barrier to attracting consumers who are concerned with social responsibility, but it may be an obstacle to attract more market (Obermiller et al., 2009). Taking into account the strong explanatory power of the functional utility in the buying intention, we can say that it is not enough to highlight the ethical value of products to stimulate the buying. Quality and convenience are key variables and fair trade organisations have to act on them. On the one hand, it is necessary to change any negative perception about the products (Wright & Heaton, 2006) and for this, the communication message should focus on the quality of the product, its origin, materials, production process or taste, among other issues. The buying of a fair trade product should not be related to charity, but with good products and ethical. With regard to the convenience, one way to make more accessible the products is to develop the e-commerce, a channel that, besides, allows reaching young people.
It is highlighted that these aspects of functional utility are even more important than symbolic associations, closer to the ethical and solidarity attributes of these products. Symbolic utility includes a social function, a personal identification and a self-esteem dimension linked to the purchase. The last two dimensions, more linked to personal issues, are more important for consumers than the social function, which measures the prestige and recognition of the brand in the social environment. It seems that people do not clearly identify the reputation of the fair trade generic brand, so it may be advisable to act on this. The social function is linked to people as members of social groups. Thus, consumers positively value those brands with a good reputation among the groups to which they belong or aspire to belong. To expand the market, it is important for the Fairtrade brand to improve its prestige and recognition in the market, no longer to be considered an issue for 'a few' ethical consumers. Again, the advice is to raise the communication because, as it has been defined in previous works (Bray et al., 2011;Castaldo et al., 2009;De Pelsmacker et al., 2005), the lack of knowledge will limit the growth of ethical consumption. A higher recognition will strengthen the symbolic associations as well, facilitating the buying.
Communication is important, and it is necessary to increase fair trade awareness among consumers. Taking into account companies' possible lack of resources, encouraging consumer word of mouth can be key, or support on social media and communication on the Internet. Knowledge leads to a higher level of consumer concern, it improves brand utilities and, therefore, influences buying intentions. In turn, interest or concern about fair trade issues favours the buying, because people will be more involved and will improve the brand associations and the attitude towards NPOs that sell the products.
With regard to this variable, it is important to take into account that the attitude towards the NPO does not influence on the buying. People buy a product of this type mainly because it is of fair trade, with a priori independence of the seller. However, the attitude towards the ONG does have a role in the model, in the sense that its image can influence on the functional utility. Perceptions about product, price and convenience can be affected according to the NPO that market the product, influencing the buying. On the contrary, the social function, personal identification and self-esteem is not affected by the trading company. These symbolic issues only come from the Fairtrade brand. This result leads us to support the argument that it so it is advisable to promote the Fairtrade mark as a brand or meta brand (Bezençon & Blili, 2010;Griffiths, 2012;Wright & Heaton, 2006), and give it the greatest importance. This paper is not exempt from limitations. The field study was carried out in a university and we used a convenience sample. This fact can limit the generalisation of the results. Furthermore, we did not include large commercialising companies in the study, even though they are increasingly including fair trade products in their portfolios. It would be interesting to carry out new studies including these companies and make comparisons between them and non-profit organisations. We also studied the functional and symbolic utilities in an aggregated form, so further research could analyse the independent effect of each construct on buying intentions. Finally, we propose as future research to include new explanatory variables in the model, such as consumer values, and it would be interesting to explore how different messages in communication campaigns influence brand utilities and consumer behaviour.

Disclosure statement
No potential conflict of interest was reported by the authors.