Premium private label: how product value, trust and category involvement influence consumers willingness to buy

Premium private labels (PPLs) are applied to products with distinctive features with prices equal, and sometimes even higher, than those of the category leaders. The objective of the retailers is to obtain, in the minds of consumers, the same positioning of national brands. So, PPLs are becoming more and more crucial for competitive advantage and store loyalty. Starting from this evidence, this paper aims to contribute to the advancement of knowledge on the subject, analyzing the impact of consumer trust in retailer, consumer involvement and product value on consumer willingness to buy PPLs. Data for hypotheses testing were collected through a 2 × 2 × 2 experiment between subjects, in which different groups of consumers were randomly exposed to different experimental conditions. Results show that: (a) consumer trust in retailer positively influences consumer’s willingness to buy in case of products with hedonic value; (b) consumer involvement has a significant main effect and there is a significant two-way interactions between trust and involvement; (c) when consumer trust in retailer is low, non-involved consumers show a greater willingness to buy the PPL for products with utilitarian rather than hedonic value.


Introduction
Although the revenue market share of private label in Italy (20.1%) is still lower than in other European countries (on average 26%, but in several countries it exceeds 40%-Nielsen 2018), their strategic role for retail companies is increasingly important (Sansone 2016;Fornari 2018). This is clearly highlighted by the growing depth and breadth of private label portfolios (Kumar and Steenkamp 2007;Geyskens et al. 2010), within which premium private labels (PPLs) are becoming more and more crucial for competitive advantage and store loyalty. According to IRI data (2020), their incidence on total Italian store brands revenues has now exceeded 20%, compared to 8% in 2010.
PPLs are applied to products with distinctive features (in terms of quality, origin, safety, variety, authenticity, sustainability, ethics, etc.) with prices equal, and sometimes even higher, than those of the category leaders 1 (Feetham and Gendall 2013;Cristini et al. 2018;Martinelli 2018).
The objective of retailers is to obtain, in the minds of these consumers, the same positioning of national brands (ter Braak et al. 2014;Schnittka 2015). The PPLs brand names often includes adjectives such as "top", "finest", "premium", "selection" and others with the same meaning. In addition to a careful selection of copackers, shelf positioning, packaging design and of the whole retailing mix are oriented to signal product quality (Feetham and Gendall 2013;Nenycz-Thiel and Romaniuk 2016;Abril and Rodriguez Canovas 2016). This is a significant change, considering that, also in Italy (Fornari 2007;Ceccacci 2013), private labels have traditionally emphasized their price convenience compared to national brands (Richardson et al.1994;Burt and Davis 1999;Sethuraman 2003). Kumar and Steenkamp (2007, p. 41) identified PPLs as "one of the hottest trends in retailing" more than a decade ago, but literature on the subject is still limited and mainly focused on: (a) strategic motivations underlying their introduction (differentiation in order to escape price competition- Kumar and Steenkamp 2007;Huang and Huddleston 2009); (b) PPLs impact on incumbents, represented by both national brands and other private labels (Geyskens et al. 2010;Nenycz-Thiel and Romaniuk 2016).
This paper aims to contribute to the advancement of knowledge on the subject, analyzing the impact of consumer trust in retailers, consumer involvement and product value on consumer willingness to buy, variable that has been overlooked in the literature on PPLs. Sethuraman and Gielens (2014), on the basis of a meta-analysis of the literature, showed that the willingness to buy traditional private labels depends on the differential value perceived by consumers between store and national brands. Several drivers of price utility and nonprice utility explain this differential value. Price utility drivers include the difference in the prices applied (mainly as a result of structural market conditions), price promotions and consumers price sensitivity (in turn, a function of demographic characteristics, purchasing behaviour and perceived risk). Nonprice utility drivers are the perceived quality and image of the retailer's brands. The results of the analysis highlight the relevance of the latter determinants, while the situation is more multifaceted as far as price utility determinants are concerned. First of all, there is a positive effect on the difference in price level, albeit less strong than usually assumed and in any case moderated by various factors. Price sensitivity is relevant, but it is not so much influenced by demographic variables such as income and household size. Consumer's level of education, frequency of purchase and perceived risk have a stronger influence. At the end of their analysis, Sethuraman and Gielens (2014, p. 149) observe that "few studies have estimated how the effects of store brands drivers systematically vary with product characteristics such as functional versus hedonic".
Taking this suggestion as our starting point, in this paper we analyze the impact of these two characteristics on PPLs consumers' willingness to buy. Moreover, as branding extension literature asserts that private labels could be considered an extension to the brand name of the store (Dacin and Smith 1994;Collins-Dodd and Lindley 2003;Aliwadi and Keller 2004;Burt and Davies 2010), we intend to verify the role played by consumer trust in retailers on consumer willingness to buy PPLs. Finally, as these brands present distinctive features that have to be perceived and appreciated in the choice process, it is also important to understand the influence of consumer involvement, which directly influences the motivation to process information (Hawkins and Hoch 1992;Petty et al. 1983).
The paper is structured as follows: the next section illustrates theoretical background and research hypotheses; Sect. 3 presents the experimental design, measures, stimuli and sample; Sect. 4 summarizes the main results obtained. In the final section we discuss limitations, theoretical and managerial implications of the study.

Theoretical background and research hypotheses
To our knowledge, in the literature on PPLs there are no studies focused on the impact of consumer trust in retailer, consumer involvement and product value; very few contributions have investigated consumer willingness to buy. Beneke et al. (2012) analyzed the impact of perceived risks that consumers associate to PPLs, to understand which of these risks significantly affect their purchase intention. This study revealed that functional and time risk both have a significant negative influence on consumers' purchase intention of PPL, while financial, physical, psychological and social risks do not significantly influence their purchase intention. More recently, Martinelli and De Canio (2019), based on the literature of traditional store brands, have examined the drivers of PPLs consumer buying intention. Results of their research show that perceived product quality, label consciousness and PPL familiarity exert a positive impact toward PPL products. Conversely, consumers do not choose a PPL product to conform to others. Moreover, despite the growing extension of the PPL assortment with geographical indications, no significant effect was found between the PPLs product branded with these indications and attitude 1 3 toward PPLs. Last but not least, findings show that the higher the level of consumer familiarity to the PPL, the higher the intention to buy PPL products.
The topic of willingness to buy PPLs, therefore, remains worthy of further research. As Walsh and Mitchell (2010, p. 4) stated: "now that the market for PLB (private label brand) is maturing and any retailers are offering premium PLB that compete directly with national brands on the basis of quality and branding to attract a different segment of the market, research needs to look at other explanatory variables that might explain consumers' intention to buy private label brands".
Among these explanatory variables, as mentioned before, in this paper we consider product value, consumer trust in retailers and consumer involvement.
About product value, while standard private labels have focused on traditional categories and price convenience over national brands, PPLs are applied to higher quality products and to new product categories. The influence of product categories on private label willingness to buy was mainly investigated by Batra and Sinha (2000). The Authors first noted that this willingness is greater for product categories for which consumers "perceive lower consequences of making a mistake in their brand selection". Secondly, they pointed out that the perception of the probability of making such a mistake is increased when there is a high heterogeneity in quality among the various brands within the selected product category. Thirdly, and this is the most relevant aspect for our research, the perceived risk is linked to the "search" or "experience" nature of the attributes that characterize product category. While search attributes can be verified before purchase by direct inspection or through easily accessible information sources by the consumer, experience attributes can only be verified through the use of the product (Nelson 1974). As a result, the level of perceived risk by the consumer is higher for experience attributes and this leads to giving greater importance to the brand, which through its trustworthiness and symbolic meanings can reduce the risk. In this perspective, for products with a high number of experience attributes, consumers tend to prefer the leading national brands (Hoch and Ha 1986;Erdem and Swait 1998;Richardson et al. 1996).
The same argument applies if, instead of the "search" or "experience" nature of the attributes, we consider the functional or experiential value of the product. In particular, in the literature a well-known distinction is between "utilitarian" and "hedonic" value (Holbrook and Hirschmann 1982;Chadhuri and Holbrook 2001;Dhar and Wertenbroch 2000;Okada 2005; Alba and Williams 2013). The first derives from tangible attributes mainly related to functional benefits, while the second comes from symbolic attributes that determine sensory stimuli and benefits related to emotions deriving from consumption experience. Since it is very difficult to verify the hedonic value before the experience of consumption, it is reasonable to assume that in this case the perceived risk is greater and therefore the role played by brand trust is crucial.
The introduction of a PPL is indeed a high-end brand extension and implies the achievement of a high degree of trust in the retailer (Dacin and Smith 1994;Richardson et al. 1994;Collins-Dodd and Lindley 2003;Aliwadi and Keller 2004;Burt and Davies 2010). Trust literature offers a huge variety of definitions of the construct (Castaldo 2008). According to Rotter (1967), trust is the expectation that a party will adopt behavior that is consistent with what she has promised. Expectation is sometimes associated with the antecedents of trust, such as the level of competencies, honesty and goodwill of the trustee (Blomqvist 1997), and non opportunistic motivations to act, too. Trust is the product of past behaviors (the history of the relationship) and of the beliefs that each subject has formed about the honesty of behaviors and reliability of the other parties. It has been referred to the trustee's willingness to keep promises and to fulfill obligations (Dwyer et al. 1987).
In consumer marketing research, studies on trust have mainly focused upon the analysis of consumer-specific concepts, such as customer satisfaction, loyalty and brand. Some works (e.g. Chadhuri and Holbrook 2001) underlined the centricity of trust as a main cognitive prerequisite of customer's loyalty by suggesting, on this basis, specific loyalty and trust management approaches. These studies prove that trust takes a specific relevance when consumers have little chances, before making his/her purchase, to have access to reliable information on the quality of goods or services under evaluation.
With reference to the relationship retailer-consumer, trust increases the certainty of performance expectations (Sethuraman and Gielens 2014), as well as beliefs on the retailer's ability and lack of opportunism. Trust, therefore, reduces the perceived risk, in particular for the intangible attributes linked to the psycho-social and experiential benefits underlying the hedonic value of the product. Consumers who have a high level of trust in the retailer should therefore show a greater willingness to buy hedonic products identified by PPL. An opposite result is expected if trust is low. Accordingly: H 1 When consumer's level of trust in retailers is high (low), PPL willingness to buy is greater in case of hedonic (utilitarian) value.
Regardless of the type of value perceived in the product, consumer choices depend on multiple factor (Baltas and Doyle 1998), which directly influence the complexity of the purchase and consumption processes as well as the behavioral orientations of individuals. Among these factors, the level of consumers' involvement is particularly important (Mitchell 1979;Traylor 1981;Zaichkowsky 1985;Mittal 1989).
Consumers elaborate information in different ways according to their level of involvement. In particular, low involvement learning occurs when consumers deal with to marketing communications without any explicit intention to process and evaluate the message (Hawkins and Hoch 1992). Consumers who are low involved in the decision task generally use simple schemas or cognitive heuristics. As conceptualized by Petty and Cacioppo (1981) in their Elaboration Likelihood Model, there are two basic routes for attitude change. In the so called "central route", consumers carefully consider information that she/he feels is central to finalize the purchase process, while in the "peripheral route" they rely on a set of simple cues in the decisional context. In this stream of research, Petty et al. (1983) demonstrated that different features of an advertisement may be more or less effective, depending upon consumer involvement: under conditions of low involvement, peripheral cues are more important than issue-relevant argumentation, but under high involvement, the opposite is true. So, consumers experiencing high levels of involvement are more inclined to accurately process informative messages, thus avoiding the use of heuristics. They allocate greater cognitive resources and invest more energy in their comprehension processes than customers who experience less involvement.
To sum up in the context of retailer brand choice, as consumer involvement increases, the deepening of the evaluation processes and therefore the ability to appreciate the distinctive features of PPL grow as well. Therefore, we expect that the willingness to buy such a brand increases for highly involved consumers. All this, of course: a) if trust in the retailer offering the PPL is high, since the perceived ability and fairness of retailer are crucial drivers of consumers choice; b) to a greater extent in presence of hedonic value, which depends on attributes that are more difficult to assess. Thus: H 2 Regardless of the type of value, if consumer's level of trust in retailers is high (low), the PPL willingness to buy is greater when consumer involvement is high (low).
H 3 When consumer involvement and consumer's level of trust in retailers are low (high), the PPL willingness to buy is higher in case of utilitarian (hedonic) value.

Methods
Data for hypotheses testing were collected through an experiment between subjects, in which different groups of consumers were randomly exposed to different experimental conditions. The experiment is based on a 2 × 2 × 2 factorial design, i.e. three independent variables, each articulated on two levels, and a dependent variable. More precisely, the independent variables are the value of the product (utilitarian vs. Low hedonic), consumer trust in retailer (high vs. low) and consumer involvement in the product category (high vs. low). The dependent variable is the consumer's willingness to buy. Table 1 summarizes the experimental design. As can be seen, it is composed of eight possible conditions, corresponding to the intersection of the three independent variables and their respective levels. The first independent variable (product value, i.e. value perceived in the product) was proposed to the sample involved in the experiment in the form of stimuli; the other two variables (consumer trust in retailer and consumer involvement in the product category) were measured using scales tested in literature and then divided ex post into two levels (high vs. low) based on the median value.
The stimuli, represented by products with utilitarian and hedonic value, were proposed to two different groups of consumers. These groups were then subdivided into two groups characterized, respectively, by low and high retailer trust. Each of these four groups was further distinguished into two subgroups according to consumer involvement in the product category (high vs. low), thus obtaining eight different samples, one for each of the experimental design conditions.

Measures and stimuli
The variables used in the experimental design were measured using scales already tested in the literature. Specifically, the variables consumer trust in retailer, perceived value in the product and consumer involvement in the product category were measured on seven points Likert scales. Table 2 shows, for each variable, the number of items used, the reference sources, original scale reliability coefficients and, in the last column, those found in this study.
The retailer selected for our experimental study is Esselunga, an italian retail company which has always been at the forefront of retail marketing activities and has a well-structured portfolio of its own brands in various product categories. The retailer portfolio includes the following private labels: Esselunga Top, Esselunga Equilibrio, Esselunga Bio, Esselunga CheJoy, Esselunga Naturama and Esselunga (the standard version of the private label, which now has over 1600 references distributed in many product categories). Since 2013, the retailer also has production facilities where bakery products, pastries, fresh pasta and stuffed pasta are made. The research was focused on the Esselunga Top brand, which covers about 60 products (jams, cakes, biscuits, pasta, olive oil, ready-made sauces, and so on) designed to "bring to the table" the excellence of raw materials at various times of the day. It is a brand that is well differentiated and recognizable as premium in terms of price, quality and appeal, as well as being widely present in the retailer points of sale. Within the above mentioned product categories, through some preliminary investigations two products have been identified: one with a utilitarian value and the other with hedonic value, which were used as stimuli to be submitted to the sample of consumers investigated.

Preliminary analysis
In order to verify the correct perception of the proposed stimuli and to better develop the questionnaire to be submitted to the consumer involved in the experimentation, some pre-tests were carried out.
First of all, a focus group was conducted with ten subjects, equally divided between males and females, aged between 18 and 25, all resident in the city of Milan and responsible for the buying process of grocery products. The focus group has made it possible to identify at a general level the habits linked to the shopping behaviour of consumers, their means-end chains and the associations developed towards the various retailers (beyond Esselunga) present in the area. After having clarified the conceptual distinction between utilitarian and hedonic value, the "Esselunga Top" product range was presented by using digital devices and surfing the company website, asking participants to select the three products with the most functional values and the three with the most hedonic connotations. Among the first, the most suitable ones were: butter, extra virgin olive oil and ready-made sauces. Products of the second type were found in: fresh pasta, extra-fine dark chocolate and chocolate biscuits.
In the light of these findings, a second pre-test was then carried out to: verifying the understanding of the measurement scales identified; to detect the value (utilitarian or hedonic) associated with Esselunga Top products selected through the focus group; identifying the products to be used as stimuli for the consumer sample analyzed. To this end, fifteen in-depth interviews were conducted. Participants were recruited through a non probabilistic extraction and interviewed in a large point of sale of Esselunga located in Milan. The selected group was balanced in term of gender, age, education and jobs. The interview covered the following aspects: shopping habits and behaviors; store loyalty; trust toward Esselunga; attitude toward Esselunga PPL (Esselunga Top). On the basis of the surveys carried out, the utilitarian product was identified in butter, while the hedonic one in extra-fine dark chocolate.
On this basis, a questionnaire for each product has been structured. After a brief introduction of the research goals, the questionnaire investigated: • buying habits and store loyalty: we ask informations about the frequency of visits to chain stores; how often consumer buys private labels; the incidence of Esselunga Top brand; • retailer trust: this section verifies the behavioral, cognitive and affective dimensions of trust toward Esselunga; • consumption habits: in the third section we investigates the consumption habits of the selected product category (excluding from the survey those who do not consume it); • consumers involvement in the product category, mainly assessed considering its relevance and psychological appeal; • the willingness to buy the Esselunga Top product, after showing its graphic display to the interviewees; • socio-demographic profile of the respondents: gender, family dimension, responsibility of the buying process, job, age, store of Esselunga usually frequented.
Finally, the questionnaire was pre tested by a group of 20 people heterogeneous in terms of age, education and employment in order to assess the clarity of the questions proposed, the full understanding of the measurement scales and the fill in time.

The sample
The questionnaire was submitted through the Qualtrics platform on the web, then on social networks such as Linkedin and Facebook, to residents in Northern Italy and visitors of Esselunga. Data were collected in January and February 2019.
A total of 380 replies were received, from which, however, it was necessary to exclude those who did not correspond to the research target, i.e. not frequenting Esselunga's point of sales and/or non consumers of the product category considered. The final sample is therefore composed of 320 consumers, equally divided between the two product categories. 51% of respondents are male, 84% are from Milan and its province (areas where the Esselunga chain is widely present). Also due to the way the questionnaire was administered, the average age is low: 35% are between 18 and 24 years old, 51% between 25 and 35, 11% between 36 and 50 and 3% over 50. Most respondents (66.1%) have a job (as employees, freelancers, managers), 32.3% are students and recent graduates, while 1.6 are housewives.
In relation to shopping habits, 53% of respondents shop only for themselves or at most for another person, while the remaining 47% provide food shopping for their family, which in most cases consists of two children. Those contacted are familiar with the retailer and its brands: those who shop in Esselunga's stores always or frequently account for 60% of the sample, while those who are used to buying the retailer's brand at least frequently account for 85%.

3 4 Results
The research hypotheses were tested by means of a multi-factor variance analysis, aimed at assessing the main effect and interactions of each independent variable (consumer trust in retailer, product value and consumer involvement) on the willingness to buy the product of premium private label.
Our first hypothesis states that, when the consumer trust in retailer is high (low), the willingness to buy the premium brand is higher for products with hedonic (utilitarian) value. As can be seen from the results presented in Table 3, the values assumed by the variables in the interaction consumer trust in retailerx product value (F = 23.039, p value = 0) confirm the existence of a significant effect on the dependent variable, represented by the willingness to buy. Figure 1 shows the effects of the two independent variables on the dependent variable. It should be noted that, when consumer trust in retailer is high, the willingness to buy the products considered does not show a statistically significant difference (5.33 vs. 5.19). On the contrary, when consumer trust in retailer is low, the willingness to buy diverges significantly (5.11 vs. 3.55). Consequently, the H1 hypothesis is only partially confirmed. More precisely, when consumer trust in retailer is low, it is confirmed that the willingness to buy for the product with utilitarian value (in this case, butter) is higher. On the other hand, when consumer trust in retailer is high, the willingness to buy for the product with hedonic value (in our case, chocolate) is not significantly higher than that characterized by utilitarian value. The second hypothesis introduces the level of consumer involvement. The output of ANOVA, represented in Table 4, summarizes the effects of the independent variables (consumer involvement and consumer trust retailer) as well as the interaction between the two on the dependent variable. As can be seen, there is a main effect of the consumer involvement (F = 41.705, p value = 0.000). The effect of interaction  between the two variables is also significant (F = 8.899, p value = 0.003) and is represented in Fig. 2.
Regardless of the type of product value, when consumer involvement is high the willingness to buy is higher, whatever the degree of consumer trust in retailer. In case of high trust, the willingness to buy assumes higher values when consumer involvement is high (6.08 vs. 4.44). This confirms that the effect of the interaction is not only statistically significant, but also very intense. However, contrary to our assumption, when consumer trust in retailer and consumer involvement are low, the willingness to buy is lower (3.97 vs. 4.58). It follows that also the H2 hypothesis is only partially confirmed.
In the third hypothesis all the independent variables are considered, assessing their joint effect on the PPLs' willingness to buy. To discuss the first part of the H3 hypothesis, only the responses of consumers (N = 165) with low involvement were filtered from the original database. Table 5 presents the results of ANOVA.  Considering these results, the first important evidence concerns the nonsignificant effect of retailer trust on the willingness to buy (p value = 0.107). The effects of product value (F = 20.134, p value = 0) and the interaction between product value and retailer (F = 16.797, p value = 0) are significant (Fig. 3).
As can be seen, when consumer trust in retailer is low, the willingness to buy assumes the maximum value for the utilitarian category (5.12) and the minimum value for the hedonic category (2.90). The difference between these values is statistically significant and, therefore, the first part of H3 is confirmed: in situations of low involvement and low consumer trust in retailer the willingness to buy for products with utilitarian value is significantly higher.
In order to discuss the second part of the hypothesis, only the responses of consumers (155) with a high level of involvement were filtered out of the database. Table 6 shows the results of ANOVA. As can be seen, the main effect of  4 Main effect and interaction of consumer trust in retailer and product value on the willingness to buy for consumers with a high level of involvement consumer trust in retailer (F = 58.166, p value = 0) and the interaction between this variable and product value (F = 9.883, p value = 0.002) are significant (Fig. 4).
The main effect of consumer trust in retailer is clearly visible, as the willingness to buy increases as trust grows. Both straight lines therefore have a positive inclination, but only in the case of the one corresponding to the hedonic value (which has, in fact, a much more pronounced inclination) is the variation in the willingness to buy statistically significant (from 4.19 to 6.23). The difference between the willingness to buy for the product with utilitarian value is not significant (5.10 vs. 5.96).
Equally not significant are the differences between the willingness to buy of products with utilitarian and hedonic value, both in the case of modest trust (4.19 vs. 5.1) and high trust (5.96 vs. 6.23). It follows that the second variant of the H3 hypothesis is not confirmed.

Discussion, implications, limitations and future research
The retailer's premium brand aims to develop the relationship with the consumer and strengthen consumer loyalty (Corstjens and Lal 2000;Ailawadi et al. 2008;Koscate-Fisher 2014;Martinelli and Vergura 2014). As stated: "the relationship between the retailer and his brand is a two-way one: the retailer guarantees the consumer the quality of the private label product, but when the consumer is satisfied with the choice he will remain loyal to the retailer" (Pellegrini 2002, p. 539). The contribution to retail differentiation and customer loyalty by PPL may be higher than that offered by national brands and also by "traditional" (i.e. "standard" and "economy") private labels. The former (national brands) are normally present in almost all retail companies at comparable prices and the latter (traditional private label) are offered by almost all retailers and must compete with those offered in discount stores at often lower prices.
In fact, PPLs are highly interesting and attractive for retailers, given the high growth rates they present, the good margins they assure (Ter Braak et al. 2013) and the resulting impact from the competitive point of view. Due to these positive results, and to the growing importance gained in retailers' strategies, PPLs were defined as the "Holy Grail" of retailers (Pauwels and Srinivasan 2004, p. 279).
On this basis, the objective of this research is to provide a contribution to the literature on store brands by investigating the factors that influence the willingness to buy of PPL, a topic that is still under investigation to date. The research hypotheses were verified with an experimental 2 × 2 × 2 design: the dependent variable is represented by the willingness to buy, while as far as the independent variables are concerned, some factors that influence the consumer's decisionmaking process and increase the level of complexity were taken into consideration: the type of value attributed to the product (utilitarian vs. hedonic), the level of consumer involvement towards the product category (low vs. high) and the degree of trust towards the retailer (low vs. high).
Given that, on a general level, consumer trust in retailer positively influences consumer's willingness to buy of PPL, it is important to note that the impact of this variable is only relevant if considered in relation to the choice of products with hedonic value. As we have seen, in fact, the willingness to buy increases significantly only for these products. When retailer trust is high the willingness to buy of hedonic product increases by 50% (from 3.55 to 5.33). This is consistent with the main objective underlying the introduction of PPL by retailer chains: leveraging retailer trust to strengthen customer loyalty through the proposal of high quality products also in categories characterized by symbolic and experiential values. However, our analysis did not fully confirm the first hypothesis. In fact, when consumer trust in retailer is high, the consumer willingness to buy does not significantly differ (5.33 vs. 5.19) across product value conditions (utilitarian vs. hedonic value). Instead, when consumer trust in retailer is low, the willingness to buy is significantly higher for products with utilitarian value (5.11 vs. 3.55). It is therefore possible to state that, in presence of a low level of consumer trust in retailer, the intention to buy products with utilitarian rather than hedonic value prevails, while if consumer trust in retailer is high there is no significant difference.
Secondly, our analysis showed a significant main effect of consumer involvement and a significant two-way interactions between trust and involvement. When consumer trust in retailer is high, consumer's willingness to buy increases by 36.9% (from 4.44 to 6.08). This seems consistent with the role attributed to trust in the buying process (Chadhuri and Holbrook 2001), which can be summarized in the reduction of perceived risk and, therefore, in the simplification of the evaluation process. The influence of consumer trust in retailer is much more intense for consumers who are psychological involved than for those who are not, but also for these the willingness to buy increases as the trust in retailer grows. For that reason, the H2 hypothesis was partially confirmed.
Accordingly to the third research hypothesis, however, it can be stated that when consumer trust in retailer is low, non-involved consumers show a greater willingness to buy the PPL for products with utilitarian rather than hedonic value (5.12 vs. 2.90). Instead, when both consumers involvement in the product category and trust in retailer are high, there is no significant difference between the willingness to buy of products with hedonic or utilitarian value (6.23 vs. 5.96).
On the basis of these results, it is possible to outline the main managerial implications of the study. First, if consumer trust in the retailer has not reached a high level, for example due to the relatively young age of the retailer, and the retailer wants to launch its own premium brand, it is advisable to choose product categories with utilitarian value. When, on the other hand, the consumer trust in retailer is high, the launch can take place in both categories.
Another implication concerns the consumers to target in order to develop sales of the PPL. It seems appropriate to direct efforts primarily towards those highly involved in the product category, but only when there is a high trust in retailer. Of course, it is up to the retailer to stimulate product testing and to communicate its novelty to consumers (especially if they are references proposed in hedonic categories), PPL could obtain not only interesting results in terms of market share and revenues, but also in terms of customer loyalty.
The results of the study also confirm the essential importance of trust in the retailer and therefore the strategies to develop it, starting from the consonance 1 3 between the values of the retailer company and those reflected in the product portfolio and in all the activities carried out at the store level, which represents the physical platform where the relationship with the customer is generated and nurtured (e.g. Baker et al. 2002;Bao et al. 2011;Diallo 2012;Mason et al. 2017;Dalli et al. 2018). From this point of view, the PPL can play an important role, supporting the competitive advantage over discount stores, which mainly rely on price convenience, and leading national brands, which apply increasingly aggressive promotional initiatives.
The study presented here, of course, is not without limits. First, the external validity of the results is influenced by the method used. Secondly, data were collected through an experiment and not in a real purchasing situation. Thirdly, respondents are Italian consumers and findings may therefore not be generalizable; comparing different national context would lead to a better understanding of PPLs consumer behaviour. Fourth, the attention was focused on a specific private label, "Esselunga Top", which is distributed by a chain that can boast a very long history, a marketing policy characterized by strong distinctive features, an innovative vision and an excellent relationship with its customers (also evidenced by the very high redemption of its "Fidaty card"). In-depth study of the subject therefore requires that the study be replicated by selecting brands from other retailers, so taking into account different corporate cultures, marketing policies and service levels. Similarly, different stimuli should be introduced from those considered here for each of the two types of value (hedonic and utilitarian). In addition, the research considered consumer behavior with respect to PPL willingness to buy, but it would also be interesting to consider other dependent variables, such as willingness to pay, the purchase intention of other types of private labels and leading industrial brands in the supply chain assortment and, consequently, the substitution effects between them. Last but certainly not least, an issue of great importance concerns the growing propensity of consumers to make their purchases online. In this regard, it would be interesting to investigate the impact of the growing use of the digital channel on the willingness to buy PPL (Arce-Urriza and Cebollada 2018).