Luxury purchase intentions: the role of individualism-collectivism, personal values and value-expressive influence in South Africa

Abstract This study examines the underlying mechanisms for predicting luxury purchase intentions in the context of an emerging market. The authors propose a conceptual model that tests the predictive role of culture and the mediating effect of individual values and value-expressiveness to explain luxury purchase intentions. To determine the antecedents of luxury brand purchase decisions, a survey of South African luxury consumers was carried out. Based on the hierarchy-of-effect model of values, this study supports the notion that higher-order cultural values—individualism and collectivism have predictive influence on individual-level values, self-enhancement, and materialism, which in turn, affect luxury brands’ value-expressiveness to predict luxury purchase intentions. The study results also reveal that luxury brands’ image interacts with value-expressiveness to explain luxury purchase intentions. Also confirmed was the direct effect of individualism and luxury purchases, whereas collectivism was not confirmed.


Subjects: Brand Management; Consumer Behaviour; International Marketing; Marketing Management; Retail Marketing
Divaries Cosmas Jaravaza ABOUT THE AUTHOR Mr Divaries C. Jaravaza is a tenured, full-time lecturer at Bindura University of Science Education (Zimbabwe). His research interests are in transformative consumer research, consumer behaviour and luxury consumption in subsistence marketplaces, entrepreneurship and accounting.

PUBLIC INTEREST STATEMENT
This study was conducted in South Africa with the primary aim of understanding how cultural personal values influence consumers' decision to buy luxury products. The study finds that consumers who endorse individualism (i.e., those who construe themselves to be independent from others) are more likely to pursue self-enhancement values and materialism than those who endorse collectivism (i.e., those who construe themselves to be closely connected with others). Individualist were also more likely to prefer luxury brands than collectivists. Furthermore, the study also revealed that consumers who desire to express their identity do so through luxury brands. However, this effect is more pronounced when the image of the luxury brands is more favourable. The effective management of luxury brands' image is crucial for achieving customer patronage of luxury products. The study concludes that cultural factors and personal values are effective mechanisms for building positive attitudes luxury products and enhancing purchase of luxury brands in emerging markets.

Introduction
The luxury market is growing rapidly due to the heightened demand for premier brands worldwide (Shukla et al., 2016). Emerging markets contribute significantly to this growth (Kaufmann et al., 2016). Although limited spending power is a major hindrance to luxury consumption, more middleincome consumers are developing a unique taste for luxury brands (CPPLuxury, 2019). Also, affluent African populations are growing with an expanding aspirational middle class (Atwal & Bryson, 2014). In 2017, Africa had a US$6.0 billion luxury sector, and this is projected to grow due to increasing urbanization and industrialization, an increasing number of millionaires, and a youthful population (Angus, 2018;Van Wyk & Posel, 2019).
Global wealth estimates indicate that there are about 166,790 high net wealth consumers in Africa with a combined investible wealth of US$1.6 trillion (World Wealth Report, 2019). Furthermore, Bain and Company previously estimated Africa to have 120,000 millionaires (valued in United States Dollars), with South Africa commanding the most considerable portion: about 71,000 millionaires (Crosswaite, 2014). South Africa is regarded as the gateway to Africa's luxury market, with the most prominent luxury brands and the fastest-growing luxury market in Sub-Saharan Africa. This has attracted some of the leading luxury brands, such as Louis Vuitton, Gucci, and Burberry, to maintain a presence in the country's elite retail districts. Despite all this, little is known about the underlying mechanisms that explain consumers' beliefs and attitudes toward luxury consumption in this crucial market. However, to add to the body of knowledge, the present study seeks to investigate these mechanisms.
Social orientation perspectives have dominated luxury consumption research at the expense of personal orientation perspectives (Tsai, 2005). Wiedmann et al. (2009) stressed that social motives alone have not provided a complete picture of luxury consumption. Instead, the focus should also be directed at individual, functional, and financial motives (Wiedmann et al., 2009). Against this background, this study combines relevant theoretical and empirical findings to explain why consumers in emerging African markets buy luxury brands, through the mechanisms of cultural and individual values. It also examined how consumers' value-expressiveness and brand cues influence luxury consumption behavior, a critical issue for luxury researchers and practitioners.
Previous studies find that consumers in Africa predominantly emphasize interdependence and collectivistic cultural beliefs (Gyekye, 1992;Schwartz, 2008). However, as the appetite for luxury brands continues to grow in Africa due to growing affluence, consumer sophistication, urbanization, etc (Basade & Ros, 2005;CPPLuxury, 2019;Signé, 2020), more consumers emphasize individualistic beliefs in collectivist African societies to guide their buying decisions. Studies on collectivism and luxury brands consumption have been done in Eastern markets such as China, Malaysia and Korea (Trompenaars & Hampden-Turner, 2011) and are sparse in Sub-Saharan Africa. In line with Western societies, the homogenization of global consumer culture across distinct cultural contexts promotes individualistic beliefs to influence luxury consumption patterns (D. Dubois & Ordabayeva, 2015;Steenkamp, 2019).
Previous studies have been based on identifying a whole market or country as individualist or collectivist (Aliyev & Wagner, 2018;Hofstede, 1980;). However, more recently, Nabi et al. (2019) noted that some consumers in the same country could have a collectivist orientation whilst others are having individualist orientation. This dual cultural orientation perspective was tested among luxury brands consumers in the emerging market of South Africa.
Individual cultural beliefs and how they impact consumer buying behavior have been explored in various business and consumer behavior research streams (Cukur et al., 2004;Lee & Choi, 2005; X.-P. Chen & Li, 2005). And while significant effort is committed to explaining the psychological, economic, and social factors as antecedents of status-enhancing luxury consumption, limited research effort is devoted to cultural drivers of luxury consumption (B. Dubois & Duquesne, 1993;Nabi et al., 2019). Overall, based on the hierarchy of effects model of values, this study contributes to the role of cultural orientations on luxury consumption through the underlying mechanism of self-enhancement values, materialism, value-expressiveness, and brand image. Three main questions were addressed in this study: (1) Do individualism-collectivism directly influence selfenhancement values and materialism, on the one hand, and luxury purchase intentions, on the other hand? (2) Do self-enhancement values and materialism influence value-expressiveness attitudes? (3) Do value-expressiveness and brand image interact to influence luxury consumption?
The next section discusses the hierarchy-of-effects model of cultural values. The study also provides preliminary results suggesting that both self-enhancement values and valueexpressiveness serially mediate the relationship between cultural individualism and luxury purchase intentions. The same processual influence does not hold for collectivism.

The hierarchy-of-effects model of values
The conceptual model of the study was based on the hierarchy-of-effects of global values. According to Vinson et al. (1977), global values are centrally held beliefs that underlie behavior and are transcendental of situations. The hierarchy-of-effect model was defined by Homer and Kahle (1988) as a theoretical flow from "abstract values to mid-range attitudes to specific behaviors" (p. 638). Also, Vinson et al. (1977) highlighted that values and attitudes are linked in a hierarchical arrangement. According to Rokeach , individual values are derived from cultural orientations to define, maintain, and regulate social systems. Cultural values constitute global values that are widely held in society and distinguish what individuals consider desirable (Markus & Kitayama, 1991;Schwartz, 2012;Steenkamp et al., 1999). Craig and Douglas (2006) describe culture as a persistent force that underlies all aspects of social behavior and interactions and is experienced through the norms and values that shape society. Culture is contained in the objects people use in their daily lives, and the values and belief systems they endorse.
At the aggregate level are cultural values. Individualism-collectivism (I-C) is the most basic cultural value that distinguishes societies (Cleveland & Bartikowski, 2018). Triandis (1995) define individualism as a social pattern where individuals regard themselves as independent of others and autonomous, motivated to seek their own interests, and prioritize their own goals over collective goals (Triandis, 1995). On the other hand, collectivism was defined as a social pattern where individuals regard themselves as an integral part of the community of family, co-workers or the nation as a whole and are motivated to live by norms and responsibilities imposed by the ingroup and prioritize the in-group goals over personal goals (Triandis, 1995). People can hold individualism and collectivism simultaneously, although the salience of each may vary depending on an individual's beliefs, norms, and cultural experiences (; Nabi et al., 2019). Yet, individualism is more conceptually consistent with self-enhancement and materialistic values than collectivism, since prioritizing these beliefs enables individuals to accentuate their uniqueness and superiority in social networks and thereby influencing the desire for power, achievement, and material possessions.
Personal values are assumed to be influenced by cultural values or global values, and in turn, influence domain-specific attitudes and actions (Steenkamp et al., 1999). Schwartz defined values as "desirable, trans-situational goals, varying in importance, which serves as guiding principles in people's lives" (Schwartz, 1992, p. 122). Schwartz's work on individual values is the most comprehensive and theoretically encompassing framework in the values domain (Schwartz, 1992). A distinguishing feature of the content of Schwartz's value theory is the motivational goals that specific values express and the outcomes they produce (see Schwartz, 1992). The present study adopts the Schwartz value theory to guide the conceptualization of personal values.
In this study, the global values, which are more abstract and centrally held, are individualism and collectivism, followed by less abstract values in the hierarchical ordering, which are selfenhancement values and materialism (Steenkamp et al., 1999;Vinson et al., 1977). Selfenhancement and materialism are regarded as generalized consumption values because they relate to many contexts, including luxury consumption. These less abstract personal values, in turn, impact on domain-specific value-expressive attitudes (Wilcox et al., 2009). Value-expressive attitudes highlight the role that luxury products help to symbolize and express a person's core values and self-identity (Johar & Sirgy, 1991;Wilcox et al., 2009). Moreover, luxury brand image is also a domain-specific cue that is theorized to be predicted by materialistic values and, in turn, influence luxury purchase intentions (Shukla, 2011;Sun et al., 2017).

The direct effect of individualism on self-enhancement values
Cultural individualism and collectivism have various facets; each expressed through unique cultural elements (Singelis et al., 1995). In this study, we focus on the composite construct of individualism, as it appears more closely related to luxury consumption behavior. Individualism is "a cultural pattern in which an autonomous self is postulated. The self is independent and different from the self of others" (Singelis et al., 1995, p. 245). Individualists are driven by the need to be unique, achieved through competition, social advancement, and power (Wong, 1997). They may use luxury consumption to achieve their objectives. Individualism is theorized as a higherorder cultural value from which less abstract personal values derive to influence consumptionspecific values.
Individualism is very much prevalent in almost all economies of the world, although it is particularly widespread in Western society where independence is strongly valued. However, due to the globalization of cultures achieved through media, technology, travel, and education, many consumers in developing nations also adopt individualistic orientations as cultural values (Alden, Steenkamp and Batra, 1999). Singelis et al. (1995) note that the growth of individualistic societies is fostered by modernity, urbanization and industrialization, rapid shifting of culture, changes in political systems Figure 1, and technological advancement.
The characteristics of individualism are identified as market pricing and democracy, low equality, and high freedom (Singelis et al., 1995). The widespread existence of these elements indicates that the cultural pattern is in existence and is growing globally. Prior studies have reported a significant relationship between individualism and cooperative decision making (X.-P. Chen & Li, 2005) and have also reported online consumers' responses to persuasive advertising (Lee & Choi, 2005). We expect that individualists would regard selfenhancement values defined by power, achievement, and hedonism to be an important value orientation. Cukur et al. (2004) noted the link between individualism and the self-enhancement values of hedonism and achievement. It is therefore predicted that: H1a: Individualism is positively related to self-enhancement values. Richins and Dawson (1992) conceptualization of materialism as a consumer value was adopted in this study. Dittmar et al. (2014, p. 880) defined materialism as "individual differences in people's long-term endorsement of values, goals and associated beliefs that center on the importance of acquiring money and possessions that convey status." Materialism is regarded as an individual value (see. Grougiou & Moschis, 2015;Kasser & Ahuvia, 2002). Gurel-Atay et al. (2014) also conceptualized materialism as a value by distinguishing between instrumental materialism and terminal materialism. Following Rokeach's values system approach, Gurel-Atay et al. (2014) categorized Richins and Dawson (1992) three elements of materialism into the instrumental value of "acquisition centrality" and the terminal values of "happiness and success".

The direct effect of individualism on materialism
Recent market changes, including shifts of political systems to capitalist structures, have accelerated the growth of materialism in emerging markets (Ger & Belk, 1996), as witnessed in several countries as well as South Africa. An improvement in technological advancement, especially in communication, has accelerated the emergence of the globalized consumers and the convergence of materialistic values (Van Wyk & Posel, 2019). Materialism is closely related to individualism (Wong, 1997). Materialistic individuals are basically self-centered (Richins, 2004;Richins & Dawson, 1992). Consumers who endorse individualism achieve social status by engaging in conspicuous consumption of global brands and prefer brands that enable them to compare better with others (Cleveland & Bartikowski, 2018).
Individualists in emerging markets have been described as cosmopolitan elites who are concerned about status and achievement (Batra et al., 2000). Such individuals tend to be highly interested in possessions, which helps them to be highly competitive and express their achievement and success. We expect individualism to be directly related to materialism because possessions and acquisitions are instrumental in accentuating inequalities among individuals.
H1b: Individualism is positively related to materialism.

The direct effect of collectivism on self-enhancement values
Following the definition by Triandis (1995), collectivism relates to individuals construing themselves as an integral part of a group and living according to the expectations of the collective. Collectivists depend on their group affiliation for personal and social meaning. Group goals, norms, and values trump individual goals, norms, and values. Maintaining cordial relationships is a prime aim of collectivist consumers. In this study, we also focused on collectivism. Collectivism "is a cultural pattern in which the individual sees the self as an aspect of an in-group" (Singelis et al., 1995, p. 244). Collectivists emphasize their group membership as an important aspect of their identity; where there is a conflict between group goals and personal interests, group interests take precedence and they eclipse personal goals.
Since collectivism rests on interdependence (Markus & Kitayama, 1991), but may encourage and accept hierarchy in social relations, we expect collectivism to be positively related to selfenhancement values. In line with Triandis theorizing (Triandis, 1995); in horizontal collectivism individuals are interdependent to the in-group there is equality. However, vertical collectivists consider themselves to be part of the in-group but members in the in-group embrace inequality. This inequality acceptance by collectivists bring materialistic tendencies to collectivist members. It was noted that collectivists aim for self-enhancement through the approval by others (Aliyev & Wagner, 2018). Thus: H2a: Collectivism is positively related to self-enhancement values.

The direct effect of collectivism on materialism
We also expect collectivism to influence materialism. Although horizontal collectivists are motivated to surrender individual needs to propel group needs, vertical collectivists emphasize on the achievement of superiority among members of the collective and they will be more willing to pursue material possessions (Triandis, 1995(Triandis, , 2001. Such vertical collectivists will use the consumption of objects to facilitate their competitive self in their social networks. Further, collectivists would use material possessions to derive social meaning since they are instrumental in sustaining their daily lives and function as symbols of status and self-expression (Dittmar et al., 2014;Karabati & Cemalcilar, 2010).
In examining why consumers desire status-imbued products, Nabi et al. (2019) reported that collectivism has a strong impact on status consumption because these consumers want to compare favorably among their in-groups. Collectivists are so much attached to group sentiments such that they may want to own material brands for the sake of their public image and group approval, compliments and praise (Aliyev & Wagner, 2018). The self-expansion theory note that materialism is driven by the need to have people and objects to propel an individual's selfrealization (Aron et al., 2013). Materialist consumers may also be intrinsically motivated to augment their self-esteem, status, and individual achievement (Siahtiri & Lee, 2019). These underlying intrinsic drivers of materialistic behavior are consistent with a vertical collectivist orientation to have a higher status within the in-group.
H2b: Collectivism is positively related to materialism.

The direct effect of materialism on value-expressiveness and brand image
The growth of capitalism has promoted the acceptance of materialism as a goal in life. Materialism is related to conspicuous consumption (Veblen, 1965). Evolutionary theorists noted the importance of materialism to signal individual wealth and self-concept (Smith & Bird, 2005). The costly signaling theory also affirms the need for individuals to signal their self-image through the possession of luxury brands (Burroughs & Rindfleisch, 2002). Self-expression is a fundamental element of the value-expressive influence of attitudes (Wilcox et al., 2009). Possession-defined success denotes the value of material things owned, and this gives the possessor the "desired selfimage" (Richins & Dawson, 1992, p. 304). The ownership and use of a luxury brand give the consumer uniqueness and self-respect required for purposes of value expression (Gurel-Atay et al., 2014;Hudders & Pandelaere, 2012).
Materialism also allows us to consider brand image as a costly signal (Smith & Bird, 2005). Veblen's theory of the leisure class (Veblen, 1965) alludes to luxury brand image as part of conspicuous consumption, which is considered in contemporary markets, whether individualist or collectivist, as essential. A profoundly entrenched brand image thrives in the "attraction economy" (Kevin, 2004), such that luxury brand consumers can reach a stage of "loyalty without reasoning" (Kevin, 2004, p. 2), where their choices are driven purely by their materialistic values and love of the brand image. This is because materialistic consumers rely on luxury possessions to communicate their status, premier brands with associations of craftsmanship, timelessness, exclusivity, uniqueness, and favorable reputation are deemed very appealing (Zhan & He, 2012). It is therefore posited that: H3: Materialism is positively related to the value-expressive influence of luxury brands.
H4: Materialism is positively related to luxury brand image.

The direct effect of self-enhancement values on value-expressive influence
This study focused on self-enhancement values because its motivational goals are largely associated with luxury brand consumption intentions. According to Sedikides and Gregg (2008), one can self-enhance either by self-advancing or self-protecting, augmenting positivity or diminishing negativity. This study focuses on the self-advancing dimension.
Schwartz (2012, p. 8) defined self-enhancement values as "values that emphasize the pursuit of one's own interests." The motivational goals of self-enhancement values are achievement, power, and hedonism (Schwartz, 1992). Achievement is a value associated with personal success. Moreover, success is expected to come from outstanding performance and competence (Schwartz, 1992). The desire for power may influence consumers to purchase luxury brands to express their status-oriented and prestigious self. Hedonism relates to valuing pleasure, optimizing excitement, and general self-indulgence (Schwartz, 1992(Schwartz, , 2012. Consumers purchase luxury brands to express their achievement goals, communicate their power and use possessions to signal enjoyment in life. Successful people tend to purchase luxury products for value-expressiveness purposes; to show prestige and acquire things they can proudly display (Wiedmann et al., 2009). Stathopoulou and Balabanis (2019) find that self-enhancement values are related to the uniqueness, usability, and symbolic value of luxury brands because the exclusivity of luxury products helps consumers to infuse their self-identities with the symbolic meanings of luxury brands (Zhan & He, 2012). Previous findings support the notion that specific aspects of consumers' self-identity (e.g., success, power, modernity, and sophistication) are best communicated to others through the use of luxury brands (Schade et al., 2016). We expect consumers who emphasize achievement, power, and hedonism to use luxury brands to support their self-identity in South Africa. Hence, it is posited: H5: Self-enhancement values are positively related to the value-expressiveness of luxury brands.

The direct effect of value-expressiveness on luxury purchase intentions
Value-expressive influences highlight consumers' use of luxury brands to communicate their "central beliefs, attitudes, and values to others" (Wilcox et al., 2009, p. 248). Value-expressiveness is essential for consumers who desire to achieve favorable self-image through the consumption of luxury objects (Schade et al., 2016;Wilcox et al., 2009). Luxury brands possess qualities that allow consumers to construct their ideal self-image (Johar & Sirgy, 1991). The costly signaling theory noted that some people are willing to undertake very costly strategies of communicating information about the self, such as buying luxury brands (Smith & Bird, 2005). The theory of the snob value of luxury brands is value-expressive and provides an important signal. In a cross-national study, Shukla and Purani (2012) found that other value perceptions significantly influenced consumers' overall luxury value perceptions among UK and Indian consumers. Schade et al. (2016) found that young adults purchased luxury brands due to their value-expressive influence. Thus, since emerging market consumers need to identify with global consumer culture, achieve self-expression, and strengthen their image through conspicuous consumption (Shukla & Purani, 2012), we expect value-expressiveness influence to predict luxury brands' purchase intentions. We posit tha : t H6: Value-expressive influence is positively related to luxury brand purchase intentions. Keller (1993) defined brand image as a psychological association held by a consumer concerning a particular brand. However, brand image has not attracted significant interest in luxury research (Shukla, 2011). Brand image was developed using marketing communications, consumption experience, and social influence (Riezebos & Riezebos, 2003). Being a successful luxury brand implies delivering products that are associated with sophistication, timelessness, pleasure, innovation, premium quality, authenticity, and excellence (Heine & Phan, 2011). These qualities give a strong image to luxury brands, delivering a competitive advantage over non-luxury brands (Aaker, 2009). Luxury products are guaranteed premium pricing due to a superior brand image (Yeoman & McMahon-Beattie, 2006).

The direct and moderating effect of brand image
Extant literature confirms that luxury brands rely on a strong brand image to attract customers (Chevalier & Mazzalovo, 2015;Di Mattia et al., 2018). Therefore, brand image becomes a pertinent asset to a luxury brand designer (Kapferer & Bastien, 2009). Specifically, luxury brands are symbols of prestige and serve as visible markers of status and, hence, are consumed for their conspicuous benefits in society (Wiedmann et al., 2009;Wilcox et al., 2009).
The distinctive brand image becomes a salient sales asset and therefore, is directly related to purchasing intentions (Shukla, 2011;Wilcox et al., 2009). Product extrinsic cues have been noted to be critical in influencing purchase decisions (Di Mattia et al., 2018). These extrinsic cues are even more pertinent to luxury brands. Among luxury brand cues, brand image is a pertinent cue (Chevalier & Mazzalovo, 2015). Purchase intentions decisions are taken after considering brand image (Shukla, 2011). Consequently, we anticipate that the favorability of luxury brands' image would positively influence purchase intentions for luxury brands.
Moreover, brand image is expected to interact with the value-expressive influence of luxury brands to predict luxury purchase intentions. Due to luxury brands' image as unique, strong, and favorable, they can facilitate consumers' maintenance of their self-identity. In a related study, Shukla (2011) found that brand image moderated the relationship between normative interpersonal influence and luxury purchase intentions in the United Kingdom and India. Consequently, we predict that consumers' luxury image perceptions will interact with value-expressive to impact luxury goods purchase intentions. Thus, it is hypothesized: H7a: Brand image is positively related to luxury brand purchase intentions.
H7b: Brand image moderates the relationship between value-expressive influence and luxury purchase intentions.

The direct effect of individualism on luxury purchase intentions
Individualism is theorized to relate positively to luxury purchase intentions in South Africa. It has been noted in extant literature that consumers buy luxury brands because they promote inequality and reinforce hierarchical arrangements in society (D. Dubois & Ordabayeva, 2015;Tsai, 2005). Cultural orientation is recognized to influence luxury consumption (D. Dubois & Ordabayeva, 2015;Nabi et al., 2019). Individualist consumers are self-centered and invest resources and effort to uphold inequalities through brand preferences (Vigneron & Johnson, 2004;Wong, 1997). If a consumer can buy Louis Vuitton accessories, it suggests luxury class membership and a sense that one is better than others. Prior research has provided empirical evidence on the positive relationship between individualism value orientation and luxury purchase intentions (Ricca & Robins, 2012). Thus, we hypothesize tha : t H8: Individualism is positively related to luxury purchase intentions.

The direct effect of collectivism on luxury purchase intentions
The in-group focus of a collectivist implies maintaining harmonious relations as well as accepting and promoting hierarchical arrangements in society (Cleveland & Bartikowski, 2018). Even collectivists tend to buy luxury goods to meet their need for status, superiority, and achievement whereas individualists would have self-expression as the main motive to buy luxury goods (Zhan & He, 2012). Some of the purposes of purchasing luxury brand include an exhibition of social status through conspicuous consumption and prestige (Steenkamp et al., 2003;Zhan & He, 2012).
Collectivists will extensively desire luxury brands because they have a symbolic value that signals consumers' success, wealth, power, and social achievement to others (Hudders & Pandelaere, 2012;Richins & Dawson, 1992). Although collectivists' consumers have a sense of duty and obligation towards the group, their orientation predisposes them toward luxury products as a vehicle to separate higher status individuals from lower-status individuals and reinforce their competitiveness (Basade & Ros, 2005). We expect that collectivists will desire luxury products to facilitate their intentions to achieve high status and superiority in their groups. Thus: H9: Collectivism is positively related to luxury purchase intentions.

Data collection and sample characteristics
The conceptual model was tested using data from an online survey. The survey was done through Qualtrics, from a sponsored link targeting social media users in South Africa. The questionnaire contained a cover letter that informed respondents about the rationale for the study. They were assured that the data was collected for research purposes only, and that their anonymity was guaranteed as no personal data that is traceable to any particular respondent was collected.
Respondents were asked to indicate their informed consent to participate in the study by answering to a yes or no question. Survey distribution criteria were determined by potential respondents' income, location, and age. Online surveys have become popular among researchers and deemed suitable for this study because they are cost-effective, flexible in design, and they provide an opportunity to achieve a large sample of diverse respondents across various locations in a short period (De Gregorio & Sung, 2010). South Africa was purposively selected for this research because it represents one of the leading markets in Africa, where several luxury brands are sold (CPPLuxury, 2019).
To promote the internal and external validity of the results, a screening question addressed whether respondents had recently bought a luxury clothing/accessories product brand or had plans to do so in the near future (Kastanakis & Balabanis, 2012). This question was designed to ensure that only consumers with positive luxury brand attitudes were included in the survey; hence, only those who answered yes could proceed to complete it.
The data was collected between 29 March and 26 April 2020. To reach a larger sample size, the survey was promoted through Facebook and Instagram ad campaigns. After one week of publishing the survey, a sample of 404 participants began the online survey. A total of 135 respondents who answered no had their survey terminated. Of those who completed the survey, 16 responses included excessive missingness, resulting in a total of 253 responses for data analysis, yielding a 63% response rate. Such response rate reflects good sampling techniques, ensuring that a sufficient number of respondents with knowledge and experience in luxury products in an emerging market context participated in the study.
The sample comprised mainly of females (86.2%), with few males (12.6%). The majority of the participants were aged between 25 and 34 years (41.5%) and 35-44 years (31.2%). Approximately, 75% were of African descent, and as many as 36% had postgraduate degrees while 33.2% had bachelor's degrees. Also, the majority of participants (72.7%) resided in Gauteng province (the largest and most commercially vibrant province of South Africa). Lastly, the highest reported monthly income was R19,999 ($1,348) or less (35.2%), followed by those who reported a monthly income of between R 20,000 ($1,348)-R 39,999 ($2,696) (29.6%). About 19% of the respondents' earned R 60,000 ($4,045) or more. The differences in the demographic profiles represent missing responses. Table 1 provides full overview of respondents' profile.

Measures
We employed established multi-item scales to measure the focal constructs (see Table 1). Sevenpoint Likert scales (1 = strongly disagree to 7 = strongly agree). Individualism and collectivism were measured with three items for each construct from Cleveland and Bartikowski (2018). Materialism was measured with four items from Richins (2004). Three items derived from Schade et al. (2016) were used to measure value-expressiveness. Also, brand image was measured with two items from C.-F. Chen and Tseng (2010). These two items are highly correlated and produce excellent factor loadings, hence they were retained to measure the construct. Unlike these scales, selfenhancement values were measured with a portrait scaling approach where portraits of individuals were presented to the respondents, who were asked to indicate how much they resemble the portraits being described. Five items derived from Burgess and Steenkamp (1999) were used to measure self-enhancement values on a six-point scale, from (1 = not at all like me, to 6 = very much like me). Lastly, three items from Shukla and Purani (2012) were used to measure luxury purchase intentions. The scale items were presented randomly to reduce sequencing effects.

Measurement model assessment
Structural equation modeling, using MPlus 8.3, was employed for data analysis (Muthén & Muthén, 1998. Before estimating the structural models, confirmatory factor analysis (CFA) was employed to validate the fit of the measurement model for its dimensionality and convergent and discriminant validity. A series of modification indices were performed to improve the overall quality of the model. The final measurement model achieved a satisfactory fit of the data (X 2 (208) = 283.97; X 2 /df = 1.37, root-mean-square error of approximation [RMSEA] = .04, confirmatory fit index [CFI] = .95, Tucker-Lewis index [TLI] = .94, standardized root-mean-square residual [SRMR] = .05). All factor loadings were statistically significant (p < .01), with all items loading above .59 on their respective target constructs. Also, all composite reliability scores were above .70, thus indicating support for convergent validity (Fornell & Larcker, 1981). The average variance extracted (AVEs) was achieved as all items exceeded the minimum threshold of .50. However, selfenhancement items had AVE of .49, which was slightly below the minimum threshold (see Table 2).
Next, discriminant validity was checked by comparing the AVEs for the variables with the shared variance between all variable pairs (Fornell & Larcker, 1981). To conclude that a scale has discriminant validity, the square root of the average variance extracted by the primary latent variable should be greater than the correlation of paired latent variables. All the variables met this criterion. The mean, standard deviations, correlations, and squared AVEs are found in Table 3.
Common method variance (CMV) poses serious methodological problems when self-reported surveys used to obtain responses on independent and dependent variables are derived from one source (Podsakoff et al., 2012). Harman's single factor test was used to assess the likelihood of CMV. Within this approach, it is recommended that the variance explained for a single factor in the factor analytic approach should not be greater than 50%. A variance of 39.8% was achieved in the principal component analysis. Moreover, the common latent marker analysis was carried out to complement Harman's approach, both suggesting that CMV poses no major concerns for the study.  Winning is everything to me. .85 Competition is the law of nature. .66 Collectivism (
Hypotheses H3 and H4 predicted that individual values of self-enhancement and materialism are positively related to the value-expressiveness of luxury brands. The results are supported in the anticipated direction because self-enhancement is significantly and positively related to valueexpressiveness (β = .27; t-value = 3.25; p < .01). Materialism also significantly and positively  Note. N = 253; SD = standard deviation; square root of average variance extracted (AVE) on the diagonal in bold predicted value-expressiveness (β = .62; t-value = 8.30; p < .01). This provides support for hypotheses H3 and H4. In this instance, materialism shows a stronger influence on the valueexpressiveness of luxury brands than self-enhancement. Moreover, materialism was positively and significantly related to luxury brand image (β = .17; t-value = 2.03; p < .05), thus, supporting H5. Consistent with our predictions, the value-expressiveness of luxury brands had a positive and significant influence on luxury purchase intentions (β = .22; t-value = 3.05; p < .01), confirming H6. Also, brand image was positively related to luxury purchase intentions (β = .36; t-value = 4.29; p < .01).

Latent moderation analysis
Several studies test interaction effects using regression analysis to assess hypothesized interaction paths through multiplicative product terms. However, by utilizing the strengths of SEM in moderation testing, this study used the latent moderated structural model (LMS) approach with the XWITH command in MPlus (Maslowsky et al., 2015). SEM effectively accounts for measurement errors;

Figure 3. The moderation of values-expressiveness influence and brand image to predict luxury purchase intentions.
hence, it is assumed to produce robust results also for the interaction effects. In LMS, it is not required that variable means are centered. The MLR function was adopted as a suitable estimator in the LMS estimation (Klein & Moosbrugger, 2000).
Conventional fit estimates are currently not available for LMS (Maslowsky et al., 2015). However, following the successful model fitting for the structural paths determined against established criteria, the interaction terms are subsequently included. The log-likelihood ratio and Akaike Information Criterion (AIC) scores were used to determine the optimal LMS results (Sardeshmukh & Vandenberg, 2017). The LMS result is optimal when the log-likelihood and AIC values are lower than the baseline SEM results.  Δlog-likelihood ratio of −3.33 and ΔAIC of 4.65 reflects a significant loss in fit between the two models.
Therefore, we concluded that the latent interaction of value-expressiveness and brand image in predicting luxury purchase intentions fits the model well. The LMS model predicts the hypothesized interaction between value-expressiveness and brand image to influence luxury purchase intentions (β = .19; t-value = 2.57; p < .05), thus, supporting H7b. The simple slope approach was used to depict the interaction effects (see Figure 3) (Dawson, 2014).

Mediation analysis
A mediation analysis was conducted to complement the current findings. The study adopted the bootstrapping method recommended by Preacher and Hayes (2008) and expounded by Zhao et al. (2010) to assess the serial mediating model investigated in this study. We used the 5000 bootstrapped resamples to test the bias-corrected 95% confidence interval (CI) for the mediation effects. Table 5 shows the mediation results. The results reveal a total effect of individualism on luxury purchase intentions (i.e., the sum of the direct effect and the indirect effects) was significant (β = .25, p < .01, CI [.06, .52]). Moreover, the total indirect effect (i.e., serial mediated effects) of individualism on luxury purchase intentions (the sum of all indirect effects) was significant (β = .05, p < .05, CI [.01, .13]). Specifically, both self-enhancement and value-expressiveness mediated the relationship between individualism and luxury purchase intentions in-sequence (β = .02, p < .10, CI [.00, .06]). Highly individualistic consumers are more likely to desire the autonomy, power and achievement that luxury brands contribute and also enabling self-identity construction (Nabi et al., 2019;Schwartz, 2008;Tsai, 2005;Wong, 1997). However, the specific indirect effect of materialism and valueexpressiveness on the one hand, and materialism and brand image, on the other hand, were not significant. Similarly, the total effect and total indirect effect, as well as the specific indirect effects of collectivism on luxury brand purchase intentions, were not significant. This suggests that the personal values and attitudinal beliefs employed in this study do not independently mediate the relationship between cultural collectivism and luxury purchase intentions.

Discussion and conclusion
The consumption of luxury brands is evolving in emerging African markets. Scholars encourage more empirical research on luxury consumption, especially in emerging markets (e.g., Shukla & Purani, 2012). Drawing on the hierarchy-of-effects model, this study offers insights into the influence of individualism-collectivism on self-enhancement values and materialism, which in turn predict the value-expressiveness of luxury brands. We also highlight the interaction between valueexpressiveness and brand image to influence luxury purchase intentions. Finally, we examine the serial mediating mechanisms inherent in the conceptual model in this study. We find that both selfenhancement values and value expressive attitudes serve as an essential underlying mechanism through which individualism orientations move in-sequence to predict purchase likelihood for luxury brands among South African consumers (Nabi et al., 2019). Ultimately, levels of individualismcollectivism relate differentially to luxury consumption in South Africa. This study responds to calls recommending more emphasis on personal orientation since social orientation perspectives dominate luxury brand literature (Tsai, 2005). Several theoretical contributions and practical implications are derived from this study.

Theoretical implications
Theoretically, this study contributes to calls to examine the role of cultural orientations in accounting for status-enhancing luxury consumption (D. Dubois & Ordabayeva, 2015). Our study establishes the validity of the hierarchy of effect model in accounting for luxury consumption behavior among South African consumers. Specifically, the results show that individualism, as cultural orientation, differentially contributes to more emphasis on self-enhancement values and materialism among luxury consumers (Dittmar et al., 2014;D. Dubois & Ordabayeva, 2015). While individualism is typically linked to high-income societies (Triandis, 1995), we found that individualism is also highly valued in developing countries (Basade & Ros, 2005). As income levels and urbanization trends grow in Africa (Signé, 2020), more consumers tend to gravitate towards individualism (Basade & Ros, 2005). The desire to exercise power, own expensive things, achieve more and enjoy life is appealing to individualistic consumers. This result is consistent with prior findings that individuals who emphasize selfenhancement values are more likely to be concerned with social hierarchies (Cleveland & Bartikowski, 2018;D. Dubois & Ordabayeva, 2015), and the need to display progress through power, achievement, and hedonism (Shukla, 2011), than those who do not.
Further, the positive relationship between individualism and collectivism contributes to the literature, suggesting that both orientations may reside in the same person in response to contextual cues (De Mooij, 2014). A South African consumer desiring independence, autonomy, and need for achievement may prioritize individualistic tendencies. On the contrary, the need to promote the interest of the in-group and meet social obligations will cause the same person to prioritize collectivistic tendencies (Basade & Ros, 2005;Dittmar et al., 2014).
Nevertheless, our findings do not support the expectation that collectivism positively predisposes consumers towards self-enhancement. Collectivists appear not to prioritize values of power, achievement, and hedonism in their consumption decisions. Although collectivists prefer hierarchy and demonstrate their superiority within social groups (Singelis et al., 1995), the desire for harmony may influence them to pursue equality instead of competition and superiority in their social groups. This may be explained by the non-significant relationship between collectivism and self-enhancement and materialism values (Dittmar et al., 2014).
This study also corroborates previous findings that establish positive links between individualism and materialism (Karabati & Cemalcilar, 2010b;Wong, 1997), furthering the understanding that cultural orientation is an essential predictor of materialism. South Africa's materialistic beliefs are anchored in people's value systems, whereby importance is ascribed to end goals of achievement, and the knowledge that material possessions are instrumental in achieving these goals (Hunt et al., 1996). However, the present study reports that collectivism does not influence materialism. Although collectivists tend to prefer products that enable them to achieve status and signal social progress to others (B. Dubois & Duquesne, 1993;Nabi et al., 2019), their desire to maintain harmony and in-group integrity may cause collectivists not to pursue materialism. Furthermore, we found that self-enhancement values and materialism directly impact the valueexpressiveness, which in turn influences luxury purchase intentions. The results demonstrate that consumers emphasize power and achievement and the centrality of material possessions to communicate their self-identity and self-image to others. South African materialists desire to have a lot of possessions, good self-image, be attractive, and popular. These study findings are consistent with the value-expressive benefits achieved through luxury brands (Kasser & Ahuvia, 2002). The finding that value-expressiveness directly influences luxury consumption behavior supports prior studies where young adult consumers demonstrate these patterns (Schade et al., 2016).
Also, we found that materialistic consumers think favorably about the symbolic benefits of luxury brand image. This brand image is diagnostic for materialists, such that when a brand has a strong, unique, and favorable image, it greatly improves the value of the brand. This finding ties in with theoretical contributions that consumers with strong impression management motives tend to be more concerned about the symbolic brand benefits than those with low impression management motives (Helgeson & Supphellen, 2004). We support the assertion that luxury products are distinguished by their symbolic meaning (Heine & Phan, 2011). Favorable brand image motivates consumers to rate a product positively (Lin et al., 2015), which contributes to premium pricing (Shukla, 2011).
Another vital contribution of this study relates to the moderating influence of brand image on luxury consumption. The findings reveal that brand image significantly moderates the relationship between value-expressive influences and luxury purchase intentions. This revelation improves our understanding of the role of brand image by providing consistent empirical support for the significant interaction effect of brand image on the relationship between normative and interpersonal influence and luxury purchase intentions (Shukla, 2011).
Consumers who desire to demonstrate their central values and self-concept will prefer luxury brands with a strong brand image to those with a less favorable image. We show that valueexpressive influence relates positively to luxury consumption among South Africans. The unique characteristics of luxury brands enable them to achieve vital value-expressive needs (e.g., making patrons look attractive, modern, classy, and in vogue). A strong brand image improves the luxury brand's ability to deliver value-expressive benefits.
Importantly, our findings provide strong support of the partial mediation effect of personal values and value-expressive attitudes in mediating the relationship between cultural individualistic value orientation and luxury consumption. This finding is revealing since it provides initial evidence of the underlying mechanisms through which cultural individualists show favourable intentions towards luxury purchase intentions in an important emerging African context. Specifically, we show that selfenhancement values and value-expressive attitudes serve as appropriate pathways through which individualism may exert predictive influence on luxury purchase intentions. Unlike materialism and brand image, the route from self-enhancement and value-expressiveness proved to be more steady in predicting luxury purchase intentions. Both materialism and value expressiveness, on the one hand, and materials and brand image, on the other hand, did not intervene in the relationship between individualism and luxury products purchase intentions. The finding suggests that consumers with strong individualistic orientation may not prioritize materialistic and value expressive tendencies when buying luxury products (Nabi et al., 2019). Both materialism and value expressiveness do not exert a favourable influence on the individualism-luxury product purchase link. Also, materialism and luxury products brand image do not cumulatively intervene to influence individualists consumers in their consideration of luxury products. This finding further supports the strength of individualism orientation in driving luxury products purchase intentions. Furthermore, our findings confirm the expectation that highly collectivist consumer may not prioritize self-enhancement and value expressive attitude as intervening variables to influence luxury consumption. Both the indirect and direct effects from collectivism to luxury products purchase were not significant, demonstrating that collectivistic tendencies may not be a suitable route for influencing the purchase of luxury products.
Ultimately, our study contributes to the ongoing discourse about the role of cultural orientation on luxury where the relationship between individualism on luxury purchase intentions supported, but not between collectivism and luxury purchase intentions (B. Dubois & Duquesne, 1993;Nabi et al., 2019). The changing emphasis on cultural beliefs in favor of individualism in predominantly collectivist societies such as South Africa can be attributed to the growing acceptance of global consumer culture, urbanization, and lifestyle orientation (Signé, 2020). Several emerging market consumers use global luxury brands as a passport to enact behaviors similar to those in Western societies. However, the strong emphasis on collectivism does not support luxury consumption. As African countries deepen global integration with growing wealth, many people will obtain the means to pursue personal choices consistent with their interests and aspirations (Schwartz, 2014). Individualism will increase emphasis on self-enhancement, materialism, and luxury consumption practices.

Managerial implications
Luxury brand managers may rely on practical insights from this study to develop effective marketing strategies suitable for South African and other emerging markets. To influence attitudes towards luxury consumption, these managers should emphasize individualism over collectivism as they strive for success, power, and social dominance. The findings support brand communication strategies that present individuals as striving to do better than others, prioritizing power and dominance, and having an enjoyable experience by consuming luxury brands.
Furthermore, positioning strategies may also use user imagery, which highlights the actual and ideal self-image of typical consumers who patronize luxury brands (Johar & Sirgy, 1991). Luxury brand managers must use advertising messages that emphasize value-expressive appeals to target consumers' self-concept. Prior studies note that advertising appeals consistent with people's self-concept are more effective than inconsistent appeals (Hosany & Martin, 2012). Besides, individualism, self-enhancement values, and materialism may serve as a suitable segmentation basis when targeting South African luxury consumers. Managers need to understand the important role that individualism plays in luxury purchase intentions. People who express strong beliefs in competition and being better than others, as well as the desire for power, achievement, and exciting life, should form a unique target audience to receive special attention.
The finding that value-expressiveness was strengthened in the face of the symbolic benefits of luxury brands to achieve favorably luxury purchase intentions should motivate brand managers to continuously invest in maintaining an appropriate level of brand image symbolism. Creating strong and unique brand associations that enable consumers to define their self-concept and express their personality will significantly influence their behavioral intentions.
Similarly, luxury brand managers who desire to target individualist consumers with the luxury brand propositions should evoke underlying factors of self-enhancement and value expressive attitudes to achieve favourable purchase outcomes. Luxury brand communications linking values of power, achievement and dominance in self-concept definitions may be an effective strategy to stimulate purchase intentions among consumer who prioritize individualism orientation in South Africa, and potentially in other emerging African markets. In sum, this study shows that cultural individualism and personal values orientations are vital constructs for understanding luxury consumption in a predominantly interdependent cultural milieu of Africa.

Limitations and future research
These findings improve our understanding of luxury brand consumption in South Africa's emerging markets through the lens of cultural and individual beliefs, attitudinal dispositions, and marketplace cues. However, the present study has some limitations. First, previous studies have established that cultural beliefs can be construed to include the vertical and horizontal dimensions of individualism-collectivism (Cleveland & Bartikowski, 2018), of which the current study was based on the two composite constructs. Future research should incorporate the vertical and horizontal strands of individualism-collectivism simultaneously to appreciate the full spectrum of cultural value influence. Similarly, Schwartz's value theory, which highlights four higher-order value domains of openness to change versus conservation and self-enhancement versus selftranscendence, may be employed to derive a full account of how individual values influence luxury consumption behavior (Schwartz, 1992;Steenkamp et al., 1999).
Value-expressiveness is conceptualized to have both personal and social domains, where individuals express their identity to themselves and others, respectively (Thorbjørnsen et al., 2007). In future research, both domains could be used to ascertain the route that produces the greatest impact on luxury consumption behavior. Also, the findings are constrained to luxury products generally. Future research may use specific brands across a wide range of product categories to understand the differential influence on luxury brand attitudes as the category changes. Crossnational research comparing luxury consumption behaviors across African countries and comparing developing African countries to developed countries (e.g., Shukla & Purani, 2012) should be explored to advance our knowledge of luxury consumption in Africa.
Moreover, screening out nonusers of luxury products and to focus the research on only consumers with positive attitudes or experience with luxury products may lead to sample selection bias. In future research, both users and nonusers of luxury products should be sampled in luxury consumption research to gain better understanding of how different groups of consumers perceive and experience luxury products. Lastly, the study used large number of females compared to males. This can potentially bias the results and limit generalizability of the study to the population. Future studies should consider using about equal number of females and males to provide a more balanced views and insights. The role of gender as moderating variable could be explored to understand how they impact luxury consumption while investigating personal values.