Psychological switching mechanism of consumers' luxury and non-luxury brand attitude formation: the effect of store location prestige and self-congruity.

This paper discusses the relationships between store location prestige, location self-congruity, brand luxury, and brand attitude. The result of structural equation modeling revealed that location prestige impacted ideal social location self-congruity positively contributing to consumers' luxury brand attitudes, while it impacted actual location self-congruity negatively contributing to consumers' non-luxury brand attitudes. The findings also implied that consumers switched referential self-concept when they evaluated different brands. This psychological switching mechanism of consumers' brand attitude formation suggests it difficult for a non-luxury company to raise its brand equity by acquiring a prestigious store location, referring to the luxury strategy.

Introduction

The Japanese apparel market is showing a decreasing trend (Japanese Ministry of Internal Affairs and Communications, 2018), in which most apparel companies are benchmarking each other while suffering from tough price competition. Therefore, not only the companies' products but also their marketing strategies are becoming homogenized, leading to commoditization, a serious dynamic in contemporary Japan (Onzo, 2007).

On the other hand, the luxury market is expanding globally. Even in Japan, the luxury market is currently showing signs of growth (D'Arpizio and Levato, 2018). Once consumers have used luxury goods with unique attributes, such as status and exclusivity (e.g., Vigneron and Johnson, 1999, 2004; Dubois et al., 2001; Heine, 2012), it seems difficult for them to lower the product prestige upon their next purchases. From this perspective, previous literature suggests that the distinctiveness of luxury brands, such as the ‘ratchet effect’ influencing consumers to continue luxury consumption even when their income is diminished (Kapferer and Bastien, 2012). Even after the financial crisis in 2008, leading luxury conglomerates, such as Moët Hennessy-Louis Vuitton, continued to expand their businesses while numerous firms in other sectors suffered from a serious period of depression. Consequently, certain luxury brands are assumed to be ‘recession proof’ (Bellaiche et al., 2010; Som and Blanckaert, 2015). Therefore, because luxury brands seem to have acquired sustainable competitive advantages, even non-luxury apparel brand managers deem the luxury strategy (Kapferer and Bastien, 2012) to be an effective anti-commoditization tactic and try to differentiate their brands by conforming to the luxury strategy to a greater or a lesser extent (Kumagai and Nagasawa, 2016).

Today, the structure of the mainstream apparel business is not a classic wholesale model, but embodies vertically integrated operations, including own retail. It includes not only well-known top-class luxury brands such as Hermès and Louis Vuitton, but also affordable apparel labels for the masses, such as H&M and Uniqlo (Millward Brown, 2018), who all distribute their products via their own stores. Therefore, regardless of the extent of brand luxury, store investment is among the most important managerial issues for apparel companies today. Noteworthy here is that an apparel brand managers frequently enforce an investment in acquiring a store space in a prestigious shopping mall or in a prestigious area (hereafter jointly referred to as the store location), even when the rent is extremely high and expected profitability is very low. This decision indicates that managers expect not only direct cash flow generated by a new store, but also a contribution on the form of secondary brand association (Keller, 1993) to raise their brand equity based on location prestige (Kumagai and Nagasawa, 2017).

Store location selection is especially important when focusing on non-luxury apparel companies attempting to raise their brand equities by implementing the luxury strategy, since they are motivated to acquire prestigious store locations to build point-of-parity association with luxury and to acquire prestigious category membership (Kotler and Keller, 2007). However, in practice, the effect of location prestige is commonly estimated only by utilising brand managers' business experiences, since a quantitative evaluation of this effect is difficult. Additionally, previous literature has reported that luxury brands are unique and differentiated from general brands (e.g., Vigneron and Johnson, 1999, 2004; Dubois et al., 2001; Heine, 2012). Based on the differentiated attributes of luxury, previous studies on luxury place a great emphasis on store location prestige (e.g., Nagasawa, 2007; Kapferer and Bastien, 2012; Som and Blanckaert, 2015). On the other hand, general marketing studies have not focused on location prestige but have often discussed accessibility and usability (e.g., Noguchi, 1987; Jones and Simmons, 1990; Pan and Zinkhan, 2006; Chebat et al., 2010; Kumagai and Nagasawa, 2018). Therefore, there is no certainty for non-luxury companies as to whether store location prestige actually contributes to their brand equities, even though the luxury strategy suggests it to be contributable to luxury. If location prestige is not effective for non-luxury brands, costly marketing efforts by non-luxury brand managers to acquire a prestigious location by merely copying the luxury strategy will be in vain.

Even from an academic perspective, the location prestige effect on luxury versus non-luxury brands remains a contended issue. Discussions concerning store location in general marketing studies are mostly based on the viewpoint of shopping cost in relation to accessibility (Pan and Zinkhan, 2006; Kumagai and Nagasawa, 2018), while the effect of secondary brand association based on location is often addressed in luxury researches (Som and Blanckaert, 2015). Based on these previous works, it seems that studies on luxury and general marketing studies have taken different approaches to the store location effect. Therefore, a comparative analysis of the location prestige effect between luxury and non-luxury brands based on this particular research focus is meaningful for both managerial and academic purposes.

From this perspective, this study conducts a consumer survey and then analyses the effect of location prestige on general consumers' attitudes toward apparel brands by adopting a psychological approach based on ‘self-congruity theory’ (Sirgy, 1982). We focus on the variations in location prestige effect based on the extent of a brand's luxury. In addition, the location prestige effect is discussed with respect to an unknown brand. This provides additional implications for cases where a company launches its brand business in a new market in which no one knows it. The findings of this study imply that consumers switch their referential self-concept according to the brand luxury levels, proving that the location prestige effect varies between luxury and non-luxury brands. This result provides managerial implications that will be helpful in developing efficient store location strategies. It also provides an academic insight in both luxury researches and general marketing studies.

Theory

Effects of store location prestige on consumers' brand attitudes

Tversky and Kahneman (1981) demonstrated the variations in consumers' decisions across different psychological dimensions and conceptualized ‘the framing of decisions’. This concept implies that an individual's decision may change according to his/her subjective conception of the acts, outcomes, and contingencies associated with a particular choice, even when the decision problem remains objectively stable (Tversky and Kahneman, 1981; Moriguchi and Takemura, 2012). Therefore, we can assume that a consumer will change his/her brand attitude based on a psychological cue which frames his/her subjective conception of a brand evaluation problem.

Kojima (1986, 1994) discussed the framing effect based on the image of a shopping location via several experiments, reporting that the subjects' perceptions of product value were high in the setting of a prestigious department store, while their perceptions were low in the setting of a supermarket with the image of a bargain sale. Moreover, previous studies have demonstrated that a prestigious store image raises brand and product images (e.g., Collins-Dodd and Lindley, 2003; Vahie and Paswan, 2006; Diallo et al., 2013).

Meanwhile, Kumagai and Nagasawa (2017) presented their work on the impact of store location status—which differs from the store's own image—on luxury brand attractiveness. Som and Blanckaert (2015) used case studies to suggest that the overall perception of a store location (hereinafter referred to as store location image) was an influencing factor on apparel brand equities. Even concerning non-luxury, Kinoshita (2004) suggested that store location image influenced consumers' perceptions of apparel brands operating direct retail. Consequently, not only store image, but also store location image including location prestige, can be proposed as a psychological frame influencing consumers' brand attitudes.

It is well-known that a consumer perceives social risk, such as anxiety about others' assessment of himself/herself, when he/she purchases apparel (Koyama, 1997). Based on this perceived risk, a consumer activates an information search for brand evaluation (Blackwell et al., 2001), referring to various aspects of brand information such as store image, in order to reduce this risk (Diallo, 2012; Roselius, 1971; Semeijn et al., 2004). In addition, existing literature reports that store image reduces the social psychological risk associated with a store brand, such as the anxiety of making negative impressions by purchasing a certain product (Liljander et al., 2009; Delgado-Ballester et al., 2014).

Assuming that store location can be a psychological frame for consumers, we consider that consumers refer to not only the store image, but also its location image when they evaluate apparel brands under social psychological risk, such as anxiety concerning others' assessment.

Self-congruity theory

Previous literature defines self-concept as the perception of oneself (Kinch, 1963; Grubb and Hupp, 1968). The literature also reports that self-concept does not consist of only one component but is comprised of multiple dimensions (Belch and Landon, 1977; Belch, 1978). Reviewing these studies, Sirgy (1982, 2015) suggested four types of self-concepts: (1) Actual self-concept: The truthful image that an individual has of himself/herself; (2) Ideal self-concept: The ideal image that an individual would like to be; (3) Actual social self-concept: The image that an individual holds of himself/herself, based on his/her beliefs about others' assessment; (4) Ideal social self-concept: The ideal image that an individual would like others to have of himself/herself.

Self-congruity is defined as the psychological closeness between these four types of self-concepts and brand user imagery (or brand image). When the distance between these elements is small, the level of self-congruity is deemed to be high (Sirgy, 1982; Sirgy et al., 1997). Sirgy (1982, 1985) conceptualized the relationships between self-concept, brand user imagery (or brand image), and consumer behaviour as the self-congruity theory, suggesting that an individual evaluates a brand favourably when his/her self-congruity based on this brand is high. Generally, in the field of self-congruity research, two methods of self-congruity measurement are widely used (Sirgy et al., 1997):

Absolute Difference Score: Σ|Bi-Pi| or its derivations, where Bi is item i of the personality scale measuring brand user imagery (e.g., Aaker, 1997) and Pi is item i of the personality scale measuring self-concept;

Direct questions: Questions simply asking the psychological closeness between brand user imagery and self-concept.

Numerous previous studies have confirmed the positive influence of actual and ideal self-congruities on consumers' brand attitudes (e.g., Sirgy, 1985; Assarut, 2008; Liu et al., 2012). The influence of self-congruity concerning brand information outside of a company's direct communications, such as sponsorship events, has also been confirmed in previous studies (Sirgy et al., 2008). Concerning store location image, Chebat et al. (2006) suggested that the upscale image of shopping malls positively influenced perceived store image via self-congruity with user imagery of shopping malls. Meanwhile, Donvito et al. (2013) suggested a positive effect of self-congruity with store image on consumers' luxury brand evaluations. While these studies do not discuss the direct relationship between consumers' brand evaluations and location image, in this study, we consider that consumers refer to self-congruity based on store location, even when they form brand attitudes, since this congruity affects store evaluations and self-congruity with store image affects brand evaluations (Chebat et al., 2006; Donvito et al., 2013).

In addition, the actual and ideal self-congruities and their influences on consumers' attitudes were confirmed to vary across brand attributes, social psychologies, behavioural contexts, and consumer characteristics (e.g., Aaker, 1999; Graeff, 1996, 1997). Moreover, previous literature reported the framing effect of store location on consumers' brand perceptions (Kumagai and Nagasawa, 2017). Consequently, we expect that a psychological frame, such as store location image, influences self-congruity, and this situational self-congruity psychologically affects consumers' brand attitude formation.

Therefore, this study employs self-congruity theory to analyse the psychological mechanism of consumers' brand attitude formation due to store location prestige as a psychological frame. To assess the effect of location prestige, this study focuses on store location self-congruity (hereafter referred to as LSC) by assessing the psychological distance between self-concept and consumer imagery of certain shopping locations without including brand images by direct questions, based on the work of Sirgy et al. (1997). Among the four types of self-concepts (Sirgy, 1982), the social psychological influence, such as others' assessment, is relevant to ideal social self-concept, since it pertains to a consumer's desire for positive social assessment. For the purposes of this study, the ideal social LSC is analysed to test this influence; it is discussed comparatively with actual LSC. The two types of LSCs discussed here are defined as follows:

Actual LSC: The closeness of psychological distance between the actual self-concept and shopper imagery in a particular location.

Ideal social LSC: The closeness of psychological distance between the ideal social self-concept and shopper imagery in a particular location.

To test the effect of store location prestige on consumers' brand attitudes and its psychological mechanism under social influence, this study conducts a pass analysis employing the above two LSCs, as illustrated in Fig. 1.

Fig. 1

Analytic model: Relationships between location prestige, LSC, and brand attitude.

The influence of brand luxury and LSC on consumers' brand attitudes

Because shopping is a socially visible behaviour, the prestige of store location can be an influencing factor on the social evaluation of shoppers. In addition, conspicuous consumption that signals social status is a general consumer behaviour that is not confined to the wealthy class (Eastman et al., 1999; Mason, 1998). Therefore, shoppers in a prestigious location are highly evaluated by others, and their imagery is close to general consumers' ideal social self-concept, reflecting consumers' desires for positive assessment by others.

On the other hand, consumers often prefer stores with the image of low price (Noguchi, 1987; Blackwell et al., 2001; Pan and Zinkan, 2006). Therefore, general consumers would ordinarily shop in a non-prestigious location where shopping costs are expected to be low; the imagery of shoppers in a non-prestigious location is therefore close to general consumers' actual self-concepts. Meanwhile, prestigious location, such as shopping malls where luxury stores are located and where shopping costs are expected to be high, is considered far from general consumers' actual lives. A store location where there are many shoppers and brand users whose imagery is close to the actual self-concepts of general consumers is therefore not assumed to be prestigious but rather general. This means that store location prestige is expected to influence actual LSC negatively, while its influence is positive on ideal social LSC.

Previous literature suggested that the role of luxury brands, such as consumers' status display (Vigneron and Johnson, 1999; Han et al., 2010) and social statements (Kapferer and Bastien, 2012), meaning that a luxury brand is an object of conspicuous and high-status consumption by which consumers display their status socially (Kumagai and Nagasawa, 2017). Other studies have suggested that the main value of luxury brands was situational utility rather than functional benefit (Nueno and Quelch, 1998; Heine and Phan, 2011). Consumers are therefore concerned about how they are assessed by others socially referring to their ideal social self-concept, especially when they purchase luxury products.

‘Construal level theory’ (Trope and Liberman, 2003; Trope et al., 2007) proposed the interplay of psychological distance and consumers' responses to others' recommendations. It suggests that an individual's construal level becomes high based on a socially and temporarily distal target. In this case, an individual tends to evaluate the target in terms of both its abstract and central features (Trope et al., 2007; Togawa et al., 2016). Additionally, Zhao & Xie (2011) reported that an individual with high construal level was more likely to be influenced by socially distal persons than by close peers. When we focus on luxury brands, we consider them to be socially and psychologically distal to a general consumer, since luxury is inaccessible and exclusive; it is a general consumer's dream (Dubois and Paternault, 1995; Katahira, 1999). Moreover, shoppers close to ideal social self-concept is assumed to be socially distal to a general consumer because this concept is not actual but his/her ideal. Consequently, even from the viewpoint of construal level theory, when consumers evaluate luxury brands, they are likely to focus on unique luxury value such as prestige and exclusivity, rather than concrete functionality in the product's attributes, under influence of shoppers close to their ideal social self-concept.

Conversely, the main value of non-luxury brands is assumed to be functional (Heine and Phan, 2011). In addition, they are affordable, compared to luxury brands. Therefore, consumers expect this functional benefit and mainly refer to their actual lifestyle when they evaluate non-luxury brands. Moreover, since non-luxury brands are more accessible, they are psychologically closer to a general consumer. Based on construal level theory, an individual evaluates the target in terms of concrete and feasible features when it is psychologically close to him/her, thereby lowering the construal level (Trope and Liberman, 2003; Trope et al., 2007). Furthermore, with a low construal level, an individual tends to be influenced by socially close people (Trope et al., 2007; Zhao and Xie, 2011). Therefore, when consumers evaluate non-luxury brands, they are likely to focus on concrete product value, such as functionality, under the influence of shoppers close to their actual self-concepts.

Considering the above, we can state that ideal social LSC affect consumers' luxury brand attitudes more positively and strongly than in the case with non-luxury brands, while actual LSC is considered to affect consumers' brand attitudes toward non-luxury more than toward luxury. Based on these assumptions, we propose the following three hypotheses:

The influence of store location prestige on actual LSC is negative, while it is positive on ideal social LSC.

Actual LSC influences consumers' attitudes toward non-luxury brands more positively and strongly than toward luxury brands.

Ideal social LSC influences consumers' attitudes toward luxury brands more positively and strongly than toward non-luxury brands.

Method

Method

In order to test these three hypotheses, this study analyses the relationships between location prestige, LSC, and consumers' attitudes toward three different brand types: a well-known luxury brand, a well-known non-luxury brand, and an unknown brand (hereafter jointly referred to as the researched brands) via structural equation modelling (SEM) based on the analytic model presented in Fig. 1.

The researched brands were Louis Vuitton as a well-known luxury brand, Uniqlo as a well-known non-luxury brand, and Toast as an unknown apparel brand. Louis Vuitton is a world-famous and prestigious apparel and leather goods brand founded in France in 1854 (Moët Hennessy - Louis Vuitton, 2017). Because of its attributes, numerous previous studies have treated it to exemplify a luxury brand (e.g., Nagasawa, 2007; Kim and Ko, 2010; Kapferer and Bastien, 2012). Millward Brown (2018), an influential consulting firm, classified Louis Vuitton as one of the top 10 valuable brands in the luxury category, together with Hermès and Prada, among others. Among these listed luxury brands, it was the best-selling brand in Japan (Yano Research Institution, 2017). Uniqlo is the best-selling Japanese apparel brand, of which the first store was launched in 1984 (First Retailing, 2018). It is well known for its low prices and high quality. Because of its chain operations for the mass market, previous literature has regarded it as comparable with other global mass apparel brands, such as H&M and Zara (Saito, 2014; Lee et al., 2014). Along with H&M and Zara, it was categorized as one of the top 10 valuable apparel brands—a different category from luxury—by Millward Brown (2018). Toast is an apparel brand established in Wales in 1997. It operates 12 stores only in the United Kingdom (Toast, 2017), and has not yet launched its business in Japan.

Concerning store location frames for brand evaluation, six locations were chosen for this study. These include ‘Isetan Shinjuku Department Store’, ‘Marunouchi Building’, ‘Ginza area’, ‘Aeon Lake Town’, ‘Lalaport Tokyo Bay’, and ‘Ikebukuro area’. Isetan Shinjuku is located in Shinjuku, Tokyo, and is the best-selling department store in Japan (Nikkei Shinbun, 2018a). There are numerous luxury boutiques as well as upper-middle class stores located here. Marunouchi building is a symbolic building accommodating both shopping floors and offices and is located in the centre of Tokyo, just in front of Tokyo station. Numerous boutiques selling high-end fashion products are located there. Ginza is not a mall, but a major shopping area in central Tokyo of which the property price is the highest in Japan (Nikkei Shinbun, 2018b). Numerous flagship stores of luxury and non-luxury apparel brands are located in and around this area. Aeon Lake Town is a representative shopping mall located in Koshigaya, Saitama, a residential city within the research area for this study. In this mall, there is a supermarket selling groceries, as well as numerous casual boutiques. Lalaport Tokyo Bay is a well-known shopping mall located in Funabashi, Chiba, which is also a residential city in the study area. It holds numerous casual boutiques. Ikebukuro is a major shopping area in central Tokyo. There are numerous streets with casual apparel stores in this area, and a few department stores located next to the Ikebukuro station. These six store locations represent major and influential commercial centres for apparel retail in and around Tokyo. In addition, the characteristics of these locations are different from each other. Some of them are shopping locations where numerous luxury and premium boutiques are located, while others are casual locations where various mass brands operate their stores. In line with our research purposes, these are appropriate locations to use for the comparative analysis and to discuss whether a prestigious location image is effective even for non-luxury brand value. These well-known locations are also easy for respondents to recognize when completing the survey.

In the survey, respondents were asked to answer according to a 7-point Likert-type scale, where 1 = absolutely disagree and 7 = absolutely agree. The questionnaire items are as follows:

Store location prestige: The location prestige of stores including shopping malls and shopping areas was measured by four items suggested by Donvito et al. (2013): ‘upper class’, ‘prestigious’, ‘high priced’, and ‘upscale’.

Brand attitude in the hypothetical settings of store location frame: Brand attitude was measured according to three items, following the work of Adaval and Monroe (2002): ‘attractive’, ‘desirable’, and ‘like’. Respondents were presented with the following statement referenced from Graeff (1997): ‘Please imagine a situation where you go shopping at [the name of shopping mall (area)]. If you find [the name of researched brand] in this mall (area), please share your impression about [the name of researched brand]’.

LSC: Both actual LSC and ideal social LSC concerning each researched store location were measured by the responses to three statements, referenced from Sirgy et al. (1997) and Assarut (2008). Actual LSC: ‘Shoppers in this mall (area) are consistent with my actual personal image’, ‘Shoppers in this mall (area) reflect my actual personal image’, and ‘Shoppers in this mall (area) are similar to my actual personal image’. Ideal social LSC: ‘Shoppers in this mall (area) are consistent with the personal image that I would like others to hold of me’, ‘Shoppers in this mall (area) reflect the personal image that I would like others to hold of me’, and ‘Shoppers in this mall (area) are similar to the personal image that I would like others to hold of me’.

Samples

For data collection, we conducted an online survey in collaboration with Macromill, the leading online research firm in Japan. It regularly conducts governmental and business surveys as well as academic researches. The number of its registered monitors is more than two million, and their demographics are approximate to the data of the Japanese governmental census (Macromill, 2013). For this survey, random sampling was conducted from the pool of its monitors living in the Tokyo area (Tokyo, Kanagawa, Saitama, and Chiba prefectures). Data were collected in compliance with the ethical code of Japan Marketing Research Association, as Macromill was an official member of this institution. The survey was announced online and participants responded on their personal computers. The age composition of the respondents was based on the population aged from their 20s to 60s in the Tokyo area, according to the Japanese governmental census (Japanese Ministry of Internal Affairs and Communications, 2010). This range is the general target population of the apparel business in Japan. The data were collected online until each age composition was completed. The screening process confirmed that all respondents knew Louis Vuitton and Uniqlo but did not know Toast; it also confirmed that they knew each store location discussed in this study. Socioeconomic status was not factored because the purpose of this research was to analyse general consumers' attitudes toward a luxury brand and a non-luxury one. The survey was conducted on 1–2 July, 2016. Adopting a list-wise deletion method (Miura and Kano, 2003), 312 valid data samples were collected. Table 1 presents the demographic characteristics of the respondents.

Table 1

Demographic variables of respondents.

Variable		n	%
Gender	Male	159	51.0
	Female	153	49.0
Age	20–24	16	5.1
	25–29	38	12.2
	30–34	32	10.3
	35–39	42	13.5
	40–44	35	11.2
	45–49	30	9.6
	50–54	35	11.2
	55–59	20	6.4
	60–64	36	11.5
	65–69	28	9.0
Household Income	<2.0	11	3.5
(Million Yen)	2.0–3.9	53	17.0
	4.0–5.9	67	21.5
	6.0–7.9	47	15.1
	8.0–9.9	33	10.6
	10.0–11.9	16	5.1
	12.0–14.9	8	2.6
	15.0–19.9	3	1.0
	20.0 +	4	1.3
	Missing	70	22.4

Results

Using a confirmatory factor analysis (CFA) regarding store location prestige, actual LSC, ideal social LSC, and consumers' brand attitudes, it was confirmed that construct reliability (CR) of each construct was above .700, each average variance extracted (AVE) was above .500, and each factor loading surpassed .700, satisfying the required levels (Fornell and Larker, 1981; Hair et al., 2014). Internal consistency reliability, convergent validity, and indicator reliability were all assumed to be sufficient. In this test, each AVE was confirmed to be higher than the squared correlation between each of the constructs, showing sufficient discriminant validity (Hair et al., 2014). Each construct's Cronbach's alpha was confirmed as sufficiently high and consistency reliability was also confirmed (Peterson, 1994) (Tables 2 and 3).

Table 2

Result of CFA: store location prestige, LSC, and consumers' brand attitudes.

Construct	Items	Factor loading	CR	AVE	Cronbach's alpha
Location prestige	Upper-class	.912	.953	.836	.953
	Prestigious	.896
	High priced	.928
	Upscale	.922
Actual LSC	Consistent	.947	.971	.917	.971
	Reflect	.964
	Similar	.962
Ideal social LSC	Consistent	.953	.971	.918	.971
	Reflect	.950
	Similar	.971
Brand attitude	Attractive	.958	.960	.888	.959
	Desireble	.952
	Like	.917

Table 3

Assessment of discriminant validity.

Construct	Location prestige	Actual LSC	Ideal social LSC	Brand attitude
Location prestige	.836
Actual LSC	.019	.917
Ideal social LSC	.027	.379	.918
Brand attitude	.050	.033	.041	.888

Note: The diagonal elements represent AVE by the constructs and other elements represent the squared correlation coefficient of each construct.

The relationships between store location prestige, actual and ideal social LSC, and consumers' brand attitudes were analysed via SEM using the analytic model (Fig. 1). The result showed a sufficient level of goodness-of-fit regarding this model (Chi-square = 2166.873, df = 180, p = .000; AGFI = .915; NFI = .977; CFI = .979; RMSEA = .044) (Hair et al., 2014; Miura and Kano, 2007). Path coefficients are illustrated in Table 4 and Fig. 2.

Table 4

Relationships between location prestige, LSC, consumers' brand attitudes, and brand luxury.

Brand	Loius Vuiton (Luxury)	Uniqlo (Non-luxury)	Toast (Unknown)
Location prestige --- > Actual LSC	-.139***	-.139***	-.138***
Location prestige --- > Ideal social LSC	.167***	.163***	.164***
Actual LSC --- > Brand attitude	-.055n.s.	.227***	.112***
Ideal social LSC --- > Brand attitude	.287***	.005n.s.	.161***

Note: ***p < .001; n.s.: non-significant; two-tailed test of significance.

Fig. 2

The relationships between location prestige, LSC, consumers' brand attitudes, and brand luxury.

Note: ***p < .001, n.s.: non-significant, two-tailed test of significance.

The results show negative influence of location prestige on actual LSC, while its impact on ideal social LSC is positive, which supports H1. The results also illustrate that actual LSC positively affects consumers' attitudes toward non-luxury brand, while it does not have a significant impact on luxury brand attitudes, which supports H2. Conversely, the results show that ideal social LSC positively affects consumers' attitudes toward luxury brand, while its impact is not significant toward non-luxury brand, supporting H3. In addition, the results show that consumers' unknown brand attitudes are influenced by both actual and ideal social LSCs.

Discussion and conclusion

Discussion

The data of this study imply that consumers feel the imagery of shoppers and users of an apparel brand which operates its store in a prestigious location is close to their self-concept they hope others to hold socially. However, they also think that the shopper and brand user imagery in a non-prestigious location is close to their self-concept in their actual lives. These results imply that store location prestige may encourage consumers' social desires, which may be converse to their actual lifestyles.

The results also demonstrate that consumers evaluate a luxury brand positively when they find the luxury brand store in a location where other consumers are close to their ideal social self-concept. These results imply that consumers focus on whether the consumer imagery in a shopping location is close to their socially desirable self-concept, especially when they purchase a luxury product. On the other hand, the results also imply that consumers evaluate a non-luxury brand positively when they find the non-luxury brand store in a location where other consumers are close to their actual self-concept. When consumers evaluate a non-luxury brand, they may focus on the functional benefit, such as usability in their actual life, that is irrelevant to status signalling.

Ideal social self-concept is distal to a consumer's actual image, therefore shopper imagery close to this self-concept is psychologically distal. Additionally, luxury brands are also relatively distal to a general consumer, since they are inaccessible and exclusive (e.g., Vigneron and Johnson, 1999, 2004; Dubois et al., 2001). Therefore, these results reinforce construal level theory (Trope and Liberman, 2003; Trope et al., 2007) suggesting that an individual tends to refer to a socially distal persons' recommendations more than people close to him/her when he/she evaluates a psychologically distal target (Zhao and Xie, 2011). The results show that consumers' attitudes toward Toast, an unknown brand, are influenced by both actual and ideal social LSC. This result points to consumers' indecisiveness in assessing an unknown brand according to only the cue of its store location.

The above implies the latent psychological mechanism of consumers' brand attitude formation. Consumers refer mainly to the ideal social self-concept to form brand attitudes toward luxury brands, while they refer to the actual self-concept to form attitudes toward non-luxury brands. Both actual and ideal social self-concepts are referred to when they form brand attitudes toward unknown brands. This psychological switching mechanism is a major finding of this study. This mechanism of brand attitude formation suggests that consumers may form different or even opposite brand attitudes based on the level of brand luxury and their brand knowledge, even when referring to an identical store location.

Conclusion

Based on the results of this study, it seems difficult for a non-luxury company to raise its brand equity only by acquiring a prestigious store location, following the luxury strategy. In addition, the data suggest it difficult for a non-luxury brand to pretend to be prestigious merely by opening its store in a prestigious location, even when it is first launched in a new market. The managerial implication of this study is that brand managers should investigate their brands' luxury levels, consumers' LSCs, and the extent of location prestige when they construct their store development strategies.

Previous studies on luxury have discussed marketing strategies, including the store location tactics of luxury brands through case studies of luxury businesses to explore luxury brands' key success factors (e.g., Kapferer and Bastien, 2012; Som and Blanckaert, 2015). These studies provided effective implications for luxury managers, such as the contribution of secondary brand association based on location prestige. If a branded store is opened in a prestigious location, consumers deem its status to be high. Therefore, this location image is assumed to be effective when consumers focus on the social value of luxury, such as status signal (Han et al., 2010).

On the other hand, general studies of store development regarding non-luxury products have mainly discussed accessibility and usability, which affects shopping cost, as factors influencing consumers' shopping behaviours (e.g., Noguchi, 1987; Jones and Simmons, 1990; Pan and Zinkhan, 2006; Chebat et al., 2010; Kumagai and Nagasawa, 2018). These studies report that the main value of general products is its functionality (Heine and Phan, 2011). Therefore, consumers are assumed to focus on functionality when they evaluate non-luxury products. However, it is not easy to differentiate a general product based on its functional attributes in a competitive market, and therefore, consumers shift their concern to the cost of acquiring the product. When the shopping cost is lowered, the expected functional benefit per acquiring cost increases (Kumagai and Nagasawa, 2018). It is therefore natural that general marketing studies have not focused on location prestige, but widely have discussed accessibility and usability.

However, since existing literature has discussed store location's effect on luxury and general brands from different points of view separately, the variation of location image effect between luxury and non-luxury brands is still an academic issue. This gap in the literature is also an issue even for the practices of store development, because numerous non-luxury brand managers attempt to raise their brand equities by leveraging the luxury approach.

Donvito et al. (2013) examined the influence of congruity between self-concept, store image, and brand image on consumers' brand evaluation. It discusses the influence of store image by adopting self-congruity theory meaningfully. However, it focuses on store attributes, which is different from location image. In addition, it also discusses luxury brands separately from non-luxury ones.

This study, however, comparatively discusses the effect of location prestige on both luxury and non-luxury brands with an equal approach and examines the possibility of non-luxury brands adopting the luxury strategy. This is an original practical and academic contribution to the literature.

Limitations

It can be expected that consumers' demographics, such as age, gender, and socioeconomic status may influence consumers' brand attitude formation. Although this paper focuses on the brand attitude formation of general consumers, affluent consumers may evaluate luxury brands in different manner than the general class, since their psychological distance with luxury is much smaller. It is also expected that numerous location factors other than prestige influence brand attitudes, while this study focuses on the location prestige effect. There may also be psychologies that latently influence consumers' attitudes more than LSC does, such as assimilation and contrast effects (Hovland et al., 1957; Anderson, 1973). Considering these effects, prestige disparity between the evaluated brand and store location may vary the location image effect on consumer behaviour. Moreover, in this study, the number of researched brands is limited; extending these elements would be appropriate for further generalization of this study. Furthermore, the discussion in this study is based on marketing research conducted in Japan only, although global marketing strategies are required even for apparel brand businesses today. These are the main limitations of this study, which could be remedied through future research.

Declarations

Author contribution statement

Ken Kumagai: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Shin'ya Nagasawa: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.