Journal of Business Research

value including physical value, economic value, expressive/social value, emotional value, and service value. Perceived quality value remains the main type of brand value perceived by consumers. Perceived quality refers to a consumer's subjective judgment about a brand's overall excellence (Zeithaml, 1988). Uniqueness refers to the degree to which customers believe that a brand is different from competing brands (Agarwal & Rao, 1996; Netemeyer et al., 2004; Wiedman et al., 2007). As an individuallevel perception of luxury brands, conspicuous value involves the brand's perceived use in conveying the consumer's social status. A well-known brand name carries a certain symbolic identity within a given society (Keller, 1993). Consumers use luxury brands to classify themselves or to distinguish themselves from others (Vigneron & Johnson, 2004). Social value refers to people's desire to possess luxury fashion brands that may serve as symbolic markers of group membership (Kim et al., 2010; Vigneron & Johnson, 2004). Consumers may use luxury fashion brands to conform to their professional position or to demonstrate their social status (Arghavan & Zaichkowsky, 2000). Luxury fashion brand consumption can serve as an effective method of conforming to the conventions of a specific social class. Emotional value refers to an essential characteristic of the perceived utility acquired from luxury products (Dubois & Gilles, 1994). Previous studies have identified emotional responses to the consumption of luxury brands, such as pleasure or excitement (Choi & Kim, 2003; Kim et al., 2010; Vigeneron & Johnson, 2004). Recent research supports the theory that perceived brand value relates to consumer purchasing behavior. Kim et al., (2010) examines the brand value perceptions of South Korean consumers regarding foreign luxury fashion brands and the influence of brand value on brand loyalty. Brand value positively affects a consumer's willingness to pay premium prices (Keller, 1993). The quality of a product significantly increases purchase motivation and thus affects consumer's purchasing decisions (Park & Park, 2003). Researchers associate perceived quality with willingness to pay, brand purchase intentions and brand choices (Netemeyer et al., 2004). Researchers support the view that the unique aspects of a brand affect both consumer preferences and the willingness to pay a higher price for the brand (Kalra & Goodstein, 1998; Netemeyer et al., 2004). Researchers observe that hedonistic consumers are more interested in their own thoughts and feelings and are more willing to pay higher prices for luxury brands (Vigneron & Johnson, 2004). In addition, consumers who purchase luxury brands show more positive emotions than those who have never purchased luxury brands (Kim et al., 2010). The consumption of luxury goods appears to have a strong social function. Therefore, the social dimension of luxury value perception refers to the perceived utility that individuals acquire by consuming products or services recognized within their own social group(s); such goods may confer conspicuousness and prestige value, which may significantly affect a customer's evaluation of and propensity to purchase or consume luxury brands (Wiedman et al., 2007). These results demonstrate the importance of the perceived value of luxury brands with respect to potential purchasing decisions. H2. A consumer's perceived brand value has a positive influence on his/her willingness to pay for luxury fashion brands. 2.5. Past purchasing experience One's past experience influences one's future actions. Specifically, purchasing experiences show a strong effect on future purchase intentions (Atwal & Williams, 2009; Kwong, Yau, Lee, Sin, & Tse, 2003; Shim, Eastlick, & Lotz, 2001). For example, Shim et al. (2001) conducted a study on the effect of prior Internet purchasing experience on intentions to make Internet purchases in the future. Kwong et al. (2003) indicates that customers with a strong intention to buy pirated CDs are likely to have had past purchasing experience buying pirated CDs. Atwal and Williams (2009) identify that achieving the greatest level of long-term success for luxury brands greatly depends on connecting with luxury consumers through brand-related experiences. According to Bian and Veloutsou (2007), prior counterfeit purchasing experience among consumers of luxury fashion brands may be common. Counterfeiting refers to the act of producing or selling a product containing an intentional and calculated reproduction of a genuine trademark (McCarthy, 2004). “Fake” brands are available in every category of fashion brands; thus, consumers must be convinced that the “real” brand has a distinctive value if they are to purchase luxury brands (Bruce & Kratz, 2007). Counterfeiting poses a significant challenge to fashion brand firms aiming to protect their revenues. Some researchers interpret counterfeiting activity in negative terms because it damages brand reputation and profits (Wee, Tan, & Cheok, 1995). However, Yoo and Lee (2009) suggest that counterfeit consumers may become buyers of the genuine items because the counterfeit items may effectively function as promotional tool for the genuine items. The willingness of consumers to pay more for noncounterfeit products directly increases in relation to their knowledge of and attitudes toward counterfeit apparel goods (Sara & Mack, 2009). Customers are divisible into different groups according to past purchasing experience, such as genuine luxury fashion brand consumers, counterfeit luxury fashion brand consumers, and potential luxury fashion brand consumers. These groups may share different fashion lifestyles, perceived brand values, and willingness to pay for luxury fashion brands. H3. The influence of a consumer's fashion lifestyle on his/her willingness to pay for luxury fashion brands is different among customers with different prior purchasing experiences. H4. The influence of a consumer's perceived brand value on his/her willingness to pay for luxury fashion brands is different among customers with different prior purchasing experiences. This study proposes a research model that examines the influence of fashion lifestyle and perceived brand value on willingness to pay and that tests how relationships between fashion lifestyle, perceived brand value, and willingness to pay vary based on different prior purchasing experiences. 3. Methods 3.1. Measurement instrument The purpose of measuring fashion lifestyles is to determine comprehensive characteristics that connect consumers to their consumption behavior with respect to luxury fashion brands. The methods used to measure fashion lifestyles were based on a previous study by Ko et al. (2007). This study modified the measurement of perceived brand value from Kim et al.'s study (2010). The item “Cheaper than other similar luxury fashion brands” changed to “I purchase luxury fashion brands only when they are on sale.” The study used a seven-point Likert scale to solicit the responses. In this study, Chinese consumers were divided into four groups according to their previous purchasing behavior. These four groups were as follows: people who had bought only counterfeit luxury fashion brands (group 1), people who had never bought genuine and counterfeit luxury fashion brands (group 2), people who had bought both genuine and counterfeit luxury fashion brands (group 3), and people who had bought only genuine luxury fashion brands (group 1518 G. Li et al. / Journal of Business Research 65 (2012) 1516–1522

4). The top luxury fashion apparel brands included Gucci, Hermes, Chanel, Prada, Versace, LV, Dior, DKNY, VG and Armani. 3.2. Data collection The study collected data using both online and offline surveys. This study uploaded the questionnaire to the Qualtrics.com website. To solicit respondents, the authors registered on a social networking site for business people, visited a fashion and beauty online forum, and obtained the email list of all members of an online tourism community. The authors sent invitations to request that potential respondents visit the Qualtrics.com website and fill out the online survey. To implement the offline survey, the study used a convenience sampling method to collect the data. The sample is from the employees of one company, one bank and one university. In addition, the authors sent some questionnaires to shopping mall customers in the Chinese cities of Harbin and Beijing. During the data collection period, 295 people answered the online questionnaire, and 290 respondents completed the offline questionnaire. After removing incomplete responses from the sample, the study obtained a valid sample of 480 respondents, representing a rate of 82.1% for valid responses. 3.3. Characteristics of the sample To deeply study the consumers based on their prior luxury fashion brand purchasing experience, this study divided Chinese consumers into four groups (as above). Table 1 reports the summary statistics of our survey data for the full sample and for the four groups. For the full sample, approximately 22.7% of the sample was male, and 45.9% of the sample was married. Approximately half of the sample was between 20 and 29 years old, and one-fourth of the sample was between 30 and 39 years old. Approximately 7.7% of the sample was between 40 and 49 years old. Most of the respondents have university education (47.2%), followed by postgraduate education (21.8%) and college education (20.7%). The represented occupations in the sample included company employees (31.6%), university students (26.4%), university teachers and government officers (8.3%), white-collar employees of foreign companies (15.5%), CEOs and private company managers (7.8%), and other occupations (10.4%). Monthly income was less than 2000 yuan RMB for 33.3% of respondents, between 2000 and 5000 yuan RMB for 33.3% of respondents, between 5000 and 8000 yuan RMB for 15.1% of respondents, between 8000 and 15,000 yuan RMB for 16.1% of respondents, and over 15,000 yuan RMB for 2.1% of respondents. 4. Data analysis and results The study conducted factor analyses of both fashion lifestyles and perceived brand value to examine the validity of our measurement. With principal component analysis and varimax rotation, fourteen items of the fashion lifestyle construct resulted in four factors with eigenvalues of 1.0 or higher; these factors accounted for 64.9% of the total variance in fashion lifestyles. The two items “When selecting clothes, I consider their function rather than their color or design” and “I am the first to try new fashions” were deleted because they demonstrate cross-loading on different factors. Thus, the fashion lifestyle construct consisted of four factors: personality, information, practicality and brand prestige. The factor loadings ranged from 0.533 to 0.849, and the corresponding values of Cronbach's a ranged from 0.726 to 0.795 (see Table 2). For the perceived brand value construct, the study identified three factors from fourteen items: social/emotional value, utilitarian value, and economic value. These factors accounted for 58.5% of the total variance in perceived brand values. The item “Easy to mix and match with other products” was deleted due to low factor loading (see Table 3). Next, a multiple linear regression model examined whether fashion lifestyles (H1) and perceived brand value (H2) affect the willingness of consumers to pay for luxury fashion brands. The study used the following multiple linear regression equation: Y ¼ α0 þ α1x1 þ α2x2 þ α3x3 þ α4x4 þ α5x5 þ α6x6 þα7x7 þ α8x3x7 þ α9x4x5 þ ε: Y is the willingness to pay for luxury fashion brands, x1 denotes a personality lifestyle, x2 denotes an information lifestyle, x3 denotes a practicality lifestyle, x4 denotes a brand prestige lifestyle, x5 is perceived social/emotional value, x6 is perceived utilitarian value, and x7 is perceived economic value. Considering possible existing interaction effects of independent variables, we also tested the multiplication of practicality lifestyle (x3) and perceived economic value (x7) and the multiplication of brand prestige lifestyle (x4) and perceived social value (x5). Table 4 shows the multiple linear regression model results for the dependent variables. For fashion lifestyle variables, only the variable for practicality fashion lifestyle (β= 0.189, P= 0.035) indicated a significant effect on willingness to pay for luxury fashion brands. Other variables of fashion lifestyle had no Table 1 Demographic characteristics of survey respondents. Characteristic Full sample (%) Group 1 (%) Group 2 (%) Group 3 (%) Group 4 (%) n= 480 n1= 123 n2= 150 n3= 63 n4= 145 Gender Male 22.7 16.7 35.0 4.2 24.1 Female 77.3 83.3 65.0 95.8 75.9 Age (years) b20 6.8 8.1 8.5 4.1 5.4 20–29 50.1 49.2 52.6 83.4 67.6 30–39 25.4 36.6 23.7 12.5 18.0 40–49 7.7 4.1 13.5 0 9.0 >50 1.0 2.0 1.7 0 0 Monthly income (RMB) Rb2000 33.3 18.4 25.9 50.0 47.4 2000≤Rb5000 33.3 36.7 31.0 20.8 36.8 5000≤Rb8000 15.1 28.6 13.8 16.7 5.3 8000≤Rb15,000 16.1 14.3 24.1 12.5 10.5 15,000≤Rb50,000 1.6 2.0 3.4 0 0 R≥50,000 0.5 0 1.7 0 0 Married or not Yes 45.9 53.3 51.8 21.7 43.9 No 54.1 46.7 48.2 78.3 56.1 Education Middle school and below 0.5 2.1 0 0 0 High school 9.8 12.5 15.0 12.5 0 College 20.7 16.7 18.3 20.8 24.6 University 47.2 54.2 45.0 37.5 49.1 Post-graduate and above 21.8 14.6 21.7 29.2 26.3 Occupation CEOs and private company managers 7.8 8.1 10.2 18.6 0 University teachers a nd government officers 8.3 14.3 6.8 4.2 7.0 Company employees 31.6 28.6 33.9 20.8 38.6 White-collar employees in foreign companies 15.5 20.4 18.6 12.5 10.5 Students 26.4 14.3 22.0 33.3 38.6 Other 10.4 14.3 8.5 12.5 5.3 Note: group 1: only counterfeit luxury fashion brand; group 2: neither genuine nor counterfeit luxury fashion brand consumers; group 3: both genuine and counterfeit luxury fashion brand consumers; group 4: only genuine luxury fashion brand consumers. G. Li et al. / Journal of Business Research 65 (2012) 1516–1522 1519

significant influence. H1 is partially supported for the full sample. For perceived brand value, all variables showed significant influence on willingness to pay for luxury fashion brands. Specifically, perceived economic value (β=−0.270, P= 0.003), perceived social/emotional value (β= 0.248, P= 0.018), and perceived utilitarian value (β= 0.439, P= 0.000) each contributed significantly to the predictive capacity of the model. Therefore, H2 is supported for the full sample. To test H3 and H4, the study compared the results of multiple linear regression models among four groups with different prior purchasing experience. Willingness to pay of consumers who only have counterfeit luxury fashion brands experience (in group 1) was more related to brand prestige lifestyle (β= 0.658, pb0.05), perceived social/emotional value (β= 0.398, pb0.05), and their interaction (β= 0.204, pb0.05). Consumers who have neither genuine nor counterfeit luxury fashion brand experience (in group 2) are more willing to pay when they have a practicality fashion lifestyle (β= 0.433, pb0.001) and perceive more economic value in fashion luxury brands (β=−0.321, pb0.01). Although consumers in these two groups did not have any genuine luxury fashion brand purchasing experience, their perceived utilitarian value had significant influence on their willingness to pay for genuine fashion products (β= 0.443, pb0.01; β= 0.603, pb0.001). Consumers who have only genuine luxury fashion brand experience (in group 4) are more likely to pay for luxury fashion brands because they perceive more social/emotional value (β= 0.353, pb0.01) and utilitarian value (β= 0.358, pb0.01). Surprisingly, for those both genuine and counterfeit luxury fashion brand consumers (in group 3), we did not find any influencing effect of fashion lifestyle and perceived value on willingness to pay (see Table 4). 5. Discussion and conclusions Focusing on China, this study assesses how the willingness of consumers to pay for luxury fashion brands relates to their fashion lifestyles and perceived value and examines how previous purchasing experience of genuine and counterfeit luxury fashion brands affects this relationship. This study validates the measurements for fashion lifestyles and perceived value for luxury fashion brands in China. This study shows that the dimensions of fashion lifestyles in China are similar to those in other countries (Ko et al., 2007). Furthermore, in this study, we found three dimensions for perceived value: social/emotional, utilitarian, and economic value. These results differ from the results of Kim et al.'s (2010) study. The results of this study also suggest that the variables for practicality fashion lifestyle, perceived social/emotional value, perceived utilitarian value, and perceived economic value significantly influence the willingness of Chinese consumers to pay for luxury fashion brands. This finding indicates that consumers who assign a higher priority to the practical aspects of fashion, such as comfort and necessity, demonstrate a greater willingness to pay for luxury fashion brands. At the same time, Chinese consumers display value consumption in their luxury fashion brand purchase behavior. They perceive and pursue comprehensive values associated with luxury fashion brands. In comparing the effects of fashion lifestyles and perceived value on the willingness to pay, we found significant differences among the four groups of consumers with different prior purchasing experiences. The perceived social/emotional and utilitarian values of Table 2 Factor analysis: fashion lifestyle. Factors Items Loading Eigenvalues % of Variance Cronbach's α Fashion lifestyle Personality I have good taste in coordinating colors and clothing designs. 0.849 3.230 22.045 0.726 I have a good eye for selecting clothes. 0.806 I tend to consider overall clothing coordination. 0.648 Information I go shopping to get ideas, even though I have no intention to buy. 0.846 1.839 17.186 0.753 I often read fashion-related books and magazines. 0.640 I like to shop in many different stores. 0.563 Brand prestige Accessories must be expensive to have class. 0.600 1.585 12.917 0.770 I believe that wearing famous brands helps me be acknowledged by others. 0.789 I believe that expensive clothes are good. 0.834 Practicality I usually wear clothes that are comfortable. 0.801 1.319 12.795 0.795 I am unlikely to purchase clothes that are difficult to maintain, even though I prefer these clothes. 0.654 I make purchases only when necessary. 0.533 KMO= 0.794; Bartlett's test= 572.058 Sig.= 0.000. Table 3 Factor analysis: perceived brand value. Factors Questions Loading Eigenvalues % of Variances Cronbach's α Perceived value Social/emotional value A person who wears luxury fashion brands appears financially successful. 0.518 5.3 27 0.779 A person who wears luxury fashion brands appears sophisticated. 0.778 A person who wears luxury fashion brands appears to be a member of a high social class. 0.752 A person who wears luxury fashion brands appears to be respected by others. 0.713 Looks the brand good/fit on the person. 0.668 I tend to release stress by shopping for luxury fashion brands. 0.512 Purchasing luxury fashion brands is refreshing. 0.731 Affiliate with social others. 0.720 Utilitarian value Luxury brands have a consistent quality. 0.751 1.6 19 0.757 Luxury brands are easy to refund and exchange. 0.797 Luxury brands are easy to maintain and wash. 0.527 Economic value I purchase luxury fashion brands only when they are on sale. 0.735 1.3 11 0.712 I purchase luxury fashion brands only when they are reasonably priced. 0.831 KMO= 0.811; Bartlett's test= 697.610; Sig.= 0.000. 1520 G. Li et al. / Journal of Business Research 65 (2012) 1516–1522

consumers who only have genuine luxury fashion brand experiences have a significant influence on willingness to pay for luxury fashion brands. In this sense, retail sellers should consider emphasizing the intrinsic brand value to current Chinese customers in their marketing strategies. For consumers who only have counterfeit luxury fashion brands experience, their brand prestige lifestyle, perceived social/ emotional value, and interaction between brand prestige lifestyle and perceived social/emotional value influence the willingness to pay for genuine luxury fashion brands. This result suggests that luxury brands may publicize social and emotional functions in which consumers could appear successful, sophisticated, and be respected when they wear genuine luxury fashion brands. For consumers who have no genuine or counterfeit luxury fashion brand experience, the perceived economic value of luxury fashion brands has a significant influence on their willingness to pay. The retailers may consider more sales promotions to increase the number of genuine luxury fashion brand purchasing experiences and, further, to cultivate potential genuine luxury fashion brand markets. Marketing strategies should aim to help this group cease their purchases of counterfeit products and to improve their value perception of genuine luxury fashion brands. According to our study, utilitarian value significantly influences future purchase intentions for those consumers of only genuine luxury fashion brands. Thus, perceived brand quality and customer service are important factors that affect the willingness of current luxury consumers to continue to pay for luxury fashion brands in the future. This result suggests that international fashion manufacturing retailers must satisfy Chinese consumers by consistently providing high quality luxury fashion brands. This result also suggests that further research should determine the factors that may affect the willingness to pay for those people who are both genuine and counterfeit luxury fashion brand customers. For consumers who bought both genuine and counterfeit luxury fashion brands, there were no factors that had significant effects on their willingness to pay. This paper makes a significant academic contribution. The study broadens the theoretical research pertaining to fashion lifestyles proposed by Ko et al. (2007) by exploring the Chinese context. From a practical perspective, the study's results provide suggestions for competitive marketing strategies for luxury fashion companies in the Chinese market. Gaining a better understanding of the behavior of Chinese consumers of luxury fashion brands should benefit the fashion industry by improving customer relationships in potentially the largest consumer market. References Aaker DA. Managing brand equity. New York: The Free Press; 1991. Agarwal MK, Rao V. An empirical comparison of consumer-based measures of brand equity. Marketing Letters 1996;7(3):237–47. Arghavan N, Zaichkowsky JL. Do counterfeits devalue the ownership of luxury brands? Journal of Product & Brand Management 2000;9(7):485–97. Atwal G, Williams A. Luxury brand marketing — the experience is everything! Journal of Brand Management 2009;16:338–46. Bian X, Veloutsou C. Consumers' attitudes regarding non-deceptive counterfeit brands in the UK and China. Journal of Brand Management 2007;14:211–22. Bruce M, Kratz PC. Competitive marketing strategies of luxury fashion companies. Fashion marketing: contemporary issues. 2nd ed. UK: Elsevier Ltd.; 2007. Brun A, Caniato F, Caridi M, Castelli C, Miragliotta G, Ronchi S, Sianesi A, Spina G. Logistics and supply chain management in luxury fashion retail: Empirical investigation of Italian firms. International Journal of Production Economics 2008;114(2): 554–70. Burmann C, Jost-Benz M, Riley N. Towards an identity-based brand equity model. Journal of Business Research 2009;62:390–7. Choi E, Kim M. Comparison of consumers' apparel purchasing behavior in the Internet retail, shopping mall, and cable TV home shopping. Clothing Culture Study 2003;11(6):808–25. Coreil J, Levin JS, Jaco EG. Lifestyle — an emergent concept in the sociomedical sciences. Culture, Medicine and Psychiatry 1985;9(4):423–37. Danziger PN. Let them eat the cake: marketing luxury to the masses as well as the classes. Chicago: Dearborn Trade Publishing; 2005. Dubois B, Gilles L. Attitudes toward the concept of luxury: an exploratory analysis. Asia-Pacific Advances in Consumer Research 1994;1(2):273–8. Fullerton S, Dodge HR. A reassessment of life style and benefits-based segmentation strategies. Journal of Marketing Management 1993;2(2):42–6. Green GT, Cordell HK, Bertz CJ, Distefano C. Construction and validation of the national survey on recreation and the enviornment's lifestyle scale. Journal of Leisure Research 2006;38(4):514–7. Henriksen M. Luxury fever in china: an analysis of Chinese luxury consumption. http:// studenttheses.cbs.dk/xmlui/handle/10417/4592009. Horley J. A longitudinal examination of lifestyles. Social Indicators Research 1992;26: 205–19. Jackson T. A comparative analysis of global luxury brands. In: Bruce M, Moore C, Birtwistle G, editors. International retail marketing. Oxford: Elsevier Butterworth-Heinemann; 2004. Jih WJK, Lee SF. An exploratory analysis of relationships between cellular phone users' shopping motivators and lifestyle indicators. Journal of Computer Information Systems 2004;44(2):65–73. Kalra A, Goodstein RC. The impact of advertising positioning strategies on consumer price sensitivity. Journal of Marketing Research 1998;35:210–24. Keller KL. Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing 1993;57:1-22. Kim YK, Lee J. Benefit segmentation of catalog shoppers among professionals. Clothing and Textiles Research Journal 2000;18(2):111–20. Kim M, Kim S, Lee Y. The effect of distribution channel diversification of foreign luxury fashion brand on consumers' brand value and loyalty in Korean market. Journal of Retailing and Consumer Services 2010;17(4):286–93. Ko E, Kim KH, Kwon JH. Impact of fashion online community characteristics on brand loyalty: comparisons among lifestyle groups. Journal of Korean Academy of Marketing Science 2006;16(3):87-106. Ko E, Kim E, Taylor CR, Kim KH, Kang IJ. Cross-national market segmentation in the fashion industry: a study of European, Korean, and US consumers. International Marketing Review 2007;24(5):629–51. Kwong KK, Yau OHM, Lee JSY, Sin LYM, Tse ACB. The effects of attitudinal and demographic factors on intention to buy pirated CDs: the case of Chinese consumers. Journal of Business Ethics 2003;47:223–35. McCarthy JT. McCarthy's desk encyclopedia of intellectual property. 3rd ed. Washington, DC: Bureau of National Affairs; 2004. Table 4 Multiple regression results for groups with different prior purchasing experience. Dependent variable Willingness to pay for luxury fashion brands Full sample Group 1 Group 2 Group 3 Group 4 Constant 0.033 0.037 0.035 0.364⁎ −0.137 Control variables Gender 0.042 0.103 0.046 −0.577 0.140 Age 0.026 0.012 0.007 0.054 0.008 Income 0.045 −00.048 0.067 −0.021 0.012 Education −0.004 0.072 0.046 −0.145 0.065 Marriage −0.132 −0.266 0.032 −0.666 0.186 Occupation 0.013 −0.0102 0.076 −0.024 0.031 Independent variables: fashion lifestyle Personality (x1) 0.020 −0.021 0.158 0.153 −0.020 Information (x2) 0.074 −0.117 0.267 −0.033 0.047 Practicality (x3) 0.189⁎ 0.099 0.433⁎⁎⁎ 0.258 0.052 Brand prestige (x4) 0.065 0.658⁎ 0.097 0.321 −0.190 Independent variables: perceived brand value Social/emotional value (x5) 0.248⁎ 0.398⁎ 0.206 0.407 0.353⁎⁎ Utilitarian value (x6) 0.439⁎⁎⁎ 0.443⁎⁎ 0.603⁎⁎⁎ 0.213 0.358⁎⁎ Economic value (x7) −0.270⁎ −0.233 −0.321* −0.230 −0.166 Interacted independent variables Practicality∗economic value (x3 ∗x7) −0.104 0.025 0.012 −0.234 −0.274⁎⁎ Brand prestige∗ social/ emotional value (x4 ∗x5) −0.012 0.204⁎ 0.010 −0.183 −0.067 Group 1: only counterfeit luxury fashion brand; group 2: neither genuine nor counterfeit luxury fashion brand consumers; group 3: both genuine and counterfeit luxury fashion brand consumers; group 4: only genuine luxury fashion brand consumers. ⁎ pb0.05. ⁎⁎⁎ pb0.001. ⁎⁎ pb0.01. G. Li et al. / Journal of Business Research 65 (2012) 1516–1522 1521

Netemeyer RG, Krishnan B, Pullig C, Wang G, Yagci M, Dean D, Ricks J, Wirth F. Developing and validating measures of facets of customer-based brand equity. Journal of Business Research 2004;57:209–24. O'Cass A, McEwen H. Exploring consumer status and conspicuous consumption. Journal of Consumer Behavior 2004;4(1):25–39. Park H, Park J. Mature consumers' purchasing motivation for imported and domestic suites. Clothing Culture Study 2003;11(1):1-10. Phau I, Prendergast G. Consuming luxury brands: the relevance of the 'Rarity Principle'. Journal of Brand Management 2000;8(2):122–38. Plummer JT. The concept and application of lifestyle segmentation. Journal of Marketing 1974;38(1):33. Potts J. Fashionomics. Policy 2007;23(4):10–6. Sara BM, Mack CS. Consumer concern, knowledge and attitude towards counterfeit apparel products. International Journal of Consumer Studies 2009;33(3):327–37. Shim S, Bickle ML. Benefit segments of the female apparel market: psychographics, shopping orientations, demographics. Clothing and Textiles Research Journal 1994;12(2):1-12. Shim S, Eastlick MA, Lotz SL. An online prepurchase intentions model: the role of intention to search. Journal of Retailing 2001;77:397–416. Tynan C, McKechnie S, Chhuon C. Co-creating value for luxury brands. Journal of Business Research 2009;63(11):1156–63. Vickers JS, Renand F. The marketing of luxury goods: An exploratory study - three conceptual dimensions. The Marketing Review 2003;3(4):459–78. Vigneron F, Johnson LW. A review and a conceptual framework of prestige-seeking consumer behavior. Academy of Marketing Science Review 1999;1:1-14. Vigneron F, Johnson LW. Measuring perceptions of brand luxury. Journal of Brand Management 2004;11(6):484–503. Wee CH, Tan SJ, Cheok KH. Non-price determinants of intention to purchase counterfeits goods. International Marketing Review 1995;12(6):19–46. Wells WD. Life Style and Psychographics. Chicago: American Marketing Association; 1974. Wiedman KP, Hennigs N, Siebels A. Measuring consumers' luxury value perception: a cross-cultural framework. Academy of Marketing Science Review 2007;7:1-21. Yoo B, Donthu N. Developing and validating a multidimensional consumer-based brand equity scale. Journal of Business Research 2001;52:1-14. Yoo B, Lee SH. Buy genuine luxury fashion products or counterfeits? Advances in Consumer Research 2009;36:280–6. Zaichkowsky JL. The psychology behind trademark infringement and counterfeiting. New Jersey: Lawrence Erlbaum Assoicates, Inc., Publishers; 2006. Zeithaml VA. Consumer perceptions of price, quality, and value: a means model and synthesis of evidence. Journal of Marketing 1988;52:2-22. 1522 G. Li et al. / Journal of Business Research 65 (2012) 1516–1522

The standardization-localization dilemma of brand communications for luxury fashion retailers' internationalization into China☆ Sindy Liu a, ⁎, Patsy Perry b , Christopher Moore c , Gary Warnaby b a International University of Monaco, 2 Avenue Albert II, Monaco b The University of Manchester, Oxford Road, Manchester M13 9PL, UK c Glasgow Caledonian University, 40 Fashion Street, Spitalfields E1 6PX, UK article info abstract Article history: Received 1 September 2014 Received in revised form 1 December 2014 Accepted 1 January 2015 Available online 28 August 2015 Keywords: Luxury communication International retailing International standardization Localization China Regional differences This paper considers the standardization–localization debate within the context of foreign luxury fashion retailers' internationalization into the emerging Chinese marketplace. Luxury retailers must balance a global– local dilemma, given the challenging trading conditions of a complex marketplace with low brand awareness and loyalty, alongside the need to maintain exclusivity and standardization of brand image across all markets. Qualitative data from 22 luxury fashion retailers currently active in the Chinese market provide rich insights that reveal the decision-making process for marketing strategies that support entry into China. Foreign luxury retailers balance the ‘global–local dilemma’ in China firstly by locating operational management control within the strategic hub of Hong Kong, and secondly by implementing far more adaptive and enterprising marketing communications than seen in other mature markets. At the same time, foreign luxury retailers retain tight strategic control of key branding dimensions at head offices in their home markets, as part of a successful long-term luxury brand management strategy. As Chinese luxury consumers increasingly feel part of a global elite, overlocalization may cause confusion over brand identity and country of origin (COO). The insights may be strategically useful for luxury retailers entering or expanding in China and also provide indications of future trends of luxury retailing in China and other emerging markets. © 2015 Elsevier Inc. All rights reserved. 1. Introduction The luxury fashion sector has undergone considerable structural change, as couture houses expanded in scale and scope, with many luxury fashion retailers internationalizing extensively through market segmentation and the development of diffusion brands to target a global segment of cosmopolitan luxury consumers. Consumer desire for luxury goods transcends national and regional cultural boundaries, forming a global luxury culture (Wiedman, Hennigs, and Siebel, 2007), and horizontal segmentation strategies adopted recognize these developments (Cleveland, Papadopoulos, & Laroche, 2011). Key constructs of luxury brands contributing to consumer motivation to purchase include heritage/history, craftsmanship/product integrity and global reputation (Fionda & Moore, 2009). China's emerging luxury market provides luxury retailers with immense opportunities (Bain & Co., 2011; KPMG, 2013) along with considerable marketing challenges resulting from its high psychic distance from countries of brand origin (O'Grady & Lane, 1996). Retail challenges include the vast size of the country and significant regional differences, censorship of international social media, unreliable printed media and low brand awareness (Chevalier & Lu, 2010; Kapferer & Bastein, 2009). Consequently, brands must ‘localize’ marketing communication strategies (Chevalier & Lu, 2010). This paper offers deeper insights into the decision-making processes of industry experts responsible for negotiating these challenges, by posing the following research question. How do foreign luxury fashion retailers balance the need for adaptation of marketing communications to overcome challenging local market conditions in China, alongside the need for standardization of marketing communications, to maintain exclusivity of brand image globally? The following sections review the literature on Chinese consumer behavior, luxury retail internationalization and standardization/adaption theory. After setting out the qualitative methodological approach, the study presents research findings. The final section offers a conclusion and discussion of managerial implications. 2. Characteristics of the Chinese luxury consumer Luxury retailers' global expansion involves targeting wealthy consumers and elite levels of society (Kapferer, 2014). As cosmopolitan consumers gain extensive exposure to other cultures via traveling or Journal of Business Research 69 (2016) 357–364 ☆ Sindy Liu was with University of the Arts (London College of Fashion) when this article was submitted and accepted. ⁎ Corresponding author. E-mail addresses: [email protected] (S. Liu), [email protected] (P. Perry), [email protected] (C. Moore), [email protected] (G. Warnaby). http://dx.doi.org/10.1016/j.jbusres.2015.08.008 0148-2963/© 2015 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Journal of Business Research

global media consumption, they develop more international and less provincially-oriented self-perceptions (Alden, Steenkamp, & Batra, 2006; Cleveland et al., 2011; Yeĝenoĝlu, 2005). Promulgated by the mass media, certain product categories become symbols of modernity and cosmopolitanism within a global consumer culture (Alden, Steenkamp, & Batra, 1999) and luxury goods may be used to signal social and personal identity (Hennigs et al., 2012; Zhang & Kim, 2013). However, despite the rapid development in Chinese luxury consumers' taste and sophistication in first-tier cities (i.e. Beijing, Shanghai, Guangzhou, Shenzhen), to a level where Chinese consumer behavior converges with that of consumers in mature markets (Bain & Co., 2012; Mintel, 2012), evidence suggests Chinese consumers still endorse both modern and traditional values (Dong and Tian, 2009; Hung, Li, & Belk, 2007). Chinese consumer luxury purchasing behavior is significantly different to other nationalities, with the lack of brand awareness and loyalty, beyond the most recognizable brands, representing key barriers to retailer success in China (Bain & Co., 2011; Chevalier & Lu, 2010; Lu, 2011, 2012; Zhan & He, 2012). Chinese luxury consumers are younger (KPMG, 2013; Lu & Pras, 2011) and more conspicuous than those in developed Western markets, and their motives for purchasing luxury goods are to show off their wealth and social status (Chadha & Husband, 2006; Kapferer, 2014; Li, Li & Kambele, 2012; Lu, 2011; Lu & Pras, 2011). In recent years, however, Chinese consumer behavior has shifted to a greater focus on intrinsic value, rarity and exclusivity (Harjani, 2013; Kapferer, 2014; KPMG, 2013; Red Luxury, 2012). Chinese consumers' behavior is also influenced by a brand's social meaning and the sense of belonging to a desirable social group, as opposed to the focus upon self-expression and individualism typically found in Western markets (Kapferer, 2014; Shukla & Purani, 2011; Zhan & He, 2012). Furthermore, Chinese luxury consumers are not only concentrated within the four Tier 1 cities, but increasingly fragmented across Tier 2, 3 and 4 cities, that could provide future investment opportunities for foreign retailers (Deloitte China, 2011). 3. Luxury retailing internationalization: standardization and adaptation Fashion retailers tend to adopt a global strategy that replicates domestic retail formulas across geographical markets in order to control brand image (Alexander & Doherty, 2010; Salmon & Tordjman, 1989). Success depends on creating an exclusive brand image through globally standardized advertising campaigns, store format and product offerings (Moore, Fernie, & Burt, 2000). Luxury brand communication strategies aim to not only sell the product, but also create a dream and reinforce brand values (Dubois & Paternault, 1995; Kapferer & Bastein, 2009). Standardizing brand positioning and advertising is therefore essential in maintaining consistency. Global positioning confers greater credibility, authority, power and value, ultimately leading to stronger brand equity (Aaker 1991; Kapferer 1992). This differs from other consumer goods categories, where advertising standardization/adaptation decisions tend to be based on cost efficiency considerations (Cleveland et al., 2011; Fastoso & Whitelock, 2007). Luxury retailers also tend to adopt centralized management, aiming for a high degree of control over brand perception and development of operations in foreign markets (Kapferer & Bastein, 2009). To succeed in international expansion, luxury retailers must maximize brand power, but with the rapid growth of emerging markets with different characteristics, it becomes crucial to adapt to local market needs (Chevalier & Lu, 2010). Marketing of luxury goods must also balance satisfaction of increasing consumer demand and protection of brand status and exclusivity (Dubois & Paternault, 1995; Kapferer, 2014). Hence, retailers' international strategies are largely based on a mix of both global and multinational approaches (Goldman, 2001). Brand communications seeking cultural congruity with consumer preferences in a particular region may incorporate local cultural values, nationalistic appeals, colors, symbols, artifacts, and myths (Westjohn, Singh, & Magnusson, 2012). Advertising examples include simple linguistic and visual adaptations, or more extensive adaptations involving model ethnicity, sentiments, usage situations or advertising appeals (Hung et al., 2007; Zhou & Belk, 2004). China's complex luxury retail market necessitates adaptation of marketing communication strategies to include PR tactics such as lavish flagship store events, fashion shows and use of Chinese celebrities as brand ambassadors to increase brand awareness (Bain & Co., 2012; Chevalier & Lu, 2010; McKinsey, 2012; Mintel, 2012). 4. Method Using a phenomenological approach (Goulding, 2005) to generate understanding of a real-world issue in a meaningful way (Patton, 2002), the first author conducted expert interviews during 2009–2014 with senior marketing managers in 22 luxury fashion retail brands operating in China. Previous studies focusing on luxury fashion retailers have successfully utilized such qualitative approaches (Dion & Arnauld, 2011; Moore et al., 2000; Tynan, McKechnie, & Chhuon, 2010). Following Zhang and Kim (2013), the term ‘luxury fashion’ includes all items that can be worn on the person including apparel, watches, jewelry and leather goods. The sample came from a wider population of luxury fashion brands present in the Chinese marketplace in 2008, which were identified using the luxury associations of four countries that produce luxury fashion goods (Fondazonie Altagamma, Italy; Comite Colbert, France; Walpole, UK; CFDA, USA), luxury market reports (Mintel, Bain & Co., McKinsey), online databases (WGSN, Business of Fashion) and top fashion magazines (Vogue China, Elle China). The authors contacted 57 foreign luxury fashion brands, identified from the above sources, and ultimately a final self-selected sample of 22 companies (representative in terms of size, country of origin and scope of product range), agreed to participate (see Table 1). Brand identity in all cases is anonymized at the request of respondents. Adopting a purposive sampling strategy within each company to access senior managers with relevant expertise and depth of involvement allowed for the creation of information rich cases (Patton, 2002), as shown in Table 1. The authors contacted the CEO of each company, who then provided access to the most relevant employee. Subsequently, a snowball technique provided access to further relevant employees in each company. The first author conducted an average of two semistructured face-to-face interviews, lasting an average of 90 min in total with each brand, with at least one interview at Director level. Given the fast-moving nature of the business environment and the Chinese market slowdown since end of 2012 (due to the Chinese government's clamp-down on corruption), eight companies agreed to participate in follow-up interviews in 2014 (companies 1, 2, 3, 4, 6, 10,17, and 20). Each interview covered a number of topics, as appearing in Table 2. Respondents gave permission for interviews to be recorded, which enabled subsequent transcription. As a form of thematic analysis, template analysis (King, 2012) facilitated data organization and reduction, enabling systematic comparison of themes across a large body of data to provide a rich and detailed account (Braun & Clarke, 2006). From the literature review, the authors generated a priori codes to create the initial template for analysis and the flexibility of the template analysis method allowed the incorporation, where appropriate, of additional codes as the interpretive process proceeded (King, 2012). The study used established qualitative data analysis principles to identify crossconnections and relationships within the data (Miles & Huberman, 1994). The authors analyzed the data deductively, by coding the interview transcripts according to key themes identified in the literature, and inductively, with themes emerging from the data itself (Braun & Clarke, 2006). The authors then revisited the literature alongside the preliminary analysis, which led to re-coding and refinement of themes 358 S. Liu et al. / Journal of Business Research 69 (2016) 357–364

and categories to create a final template for structuring the final analysis (King, 2012), as shown in Table 3 below. 5. Results and discussion Foreign luxury retailers balance the ‘global–local dilemma’ in China by adapting to challenging market conditions, while maintaining a standardized luxury image through two key areas of strategic adaptation. The first of these is the use of Hong Kong as a bridgehead to entry into the Chinese mainland. Secondly, highly adaptive and enterprising brand communications are employed to raise brand awareness and generate loyalty, to an extent not seen in retailers' home markets. Findings also reveal nuances in consumer sophistication between different tier cities and the dangers to brand image and country of origin perceptions that may occur as a result of the over-localization of marketing communications. 5.1. Use of Hong Kong as cultural and managerial bridgehead Strong evidence supports movement towards decentralization of operational management control through the use of Hong Kong as an intermediate management place. As most luxury retailers have long histories of success in Hong Kong, the location provides luxury retailers with the existing resources and knowledge to successfully expand into mainland China. Retail internationalization patterns are influenced by geographic or psychic proximity (Moore et al., 2000; O'Grady & Lane, 1996; Salmon & Tordjman, 1989), and in this case, Hong Kong's colonial history and geographic location provides cultural proximity to the West, as well as cultural and geographical proximity to the Chinese mainland. As the most popular tourist destination of Chinese ‘mainlanders’ (KPMG, 2013; Yu, 2012) and with greater luxury expertise, Hong Kong is crucial in increasing brand awareness and success in China: For brands that already had huge success in Hong Kong, our company could already see the potential of China before the mainland was opened up to foreign investors…Plus the distributors of luxury brands are all from Hong Kong, which helped many of our brands' start-up in mainland China. (Group Senior Executive for Asia Pacific, Company 3) A follow-up interview in 2014 confirmed the continued adoption of this “near neighbor approach” (Moore et al., 2000, p.921): Hong Kong is the window of mainland China. The rest of Asia also use Hong Kong as a reference in terms of luxury, lifestyle, fashion, pop culture, business center etc. And doing business is much easier in Hong Kong, plus the Hong Kong retail market is one of the most competitive in the world, so that if you're strong in Hong Kong, people respect you in Asia. (Head of Franchising & Licensing, Company 17) Hong Kong's connection to both China and the luxury retailers' home markets provides luxury retailers with a perfect intermediate location, and there was evidence of how Hong Kong's importance could increase incrementally over time. The first stage in moving away from centralized control in the retailer's home market is decentralized operational control in Hong Kong, with a regional management team wholly responsible for marketing communication strategies, regional retail operations management and regional warehouse and distribution. Experience over time could lead to the next stage of decentralized operational Table 1 Sample breakdown. Company/producta Country of origin Year and city of entry into China Respondent(s) Number of interviews 1. Leather goods, accessories & RTW France 1992 (Beijing) Vice President Marketing Manager (Greater China Region) CRM Manager 4 2. Men's bespoke & RTW UK 1992 (Beijing) CEO Managing Director 3 3. 3 key brands of conglomerate: jewelry, watches & menswearb Switzerland 1993 (Beijing) Group Senior Executive for Asia Pacific International Marketing Director Marketing Manager, China 5 4. Leather goods France 1997 (Beijing) Commercial Area Manager, China 2 5. RTW Italy 1997 (Beijing) International Group Marketing Director 2 6. RTW UK 1997 (Shanghai) Head of Licensing 3 7. RTW UK 1999 (Shanghai) Head of Wholesale 3 8. RTW Germany 1999 (Shanghai) President, Greater China Region Director (Accessories) 2 9. Leather goods & RTW Italy 1999 (Shanghai) Regional Marketing Director (Asia Pacific) 1 10. Jewelry US 2001 (Shanghai) Marketing Manager (Greater China Region) European Marketing Manager (responsible for Chinese tourists) 4 11. Jewelry UK 2002 (Shanghai) President, China 2 12. Leather goods & accessories US 2003 (Shanghai) Head of International Public Relations 3 13. RTW Italy 2004 (Hangzhou) Brand Manager (Greater China Region) 1 14. RTW UK 2004 (Shanghai) International Marketing Director 3 15. RTW France 2004 (Shanghai) CEO Director of International Marketing 2 16. RTW & couture France 2005 (Beijing) Marketing Director (Asia Pacific) International Marketing Manager 2 17. RTW France 2006 (Shanghai) Head of Franchising & Licensing 3 18. Men's bespoke UK 2008 (Beijing) Director/Owner 2 19. RTW Italy 2009 (Beijing) CEO 1 20. Footwear UK 2009 (Beijing) Brand Manager (Greater China Region) 2 a RTW refers to ‘ready-to-wear’ (pret-à-porter). b Company 5 is comprised of three interviewees within the same conglomerate,who requested their data be grouped together. The Group Senior Executive for Asia Pacific did not wish to reveal his actual job title. S. Liu et al. / Journal of Business Research 69 (2016) 357–364 359

control in China, but integrated with the regional Hong Kong operation. This could take the form of a China subsidiary managed by the Hong Kong subsidiary and the head office in the home market, or alternatively a China operation which is jointly managed by the Hong Kong subsidiary and a local partner. Research findings suggest no significant variation in this pattern, according to the size or the product orientation of companies. In all cases but one, Hong Kong is an instrumental step for foreign luxury retailers to expand into mainland China, and acts as a strategic gateway to the mainland Chinese market. 5.2. Use of flagship stores for brand tangibility and relationship building Due to the low brand awareness and loyalty of Chinese consumers, retail success is largely dependent on the level of brand recognition. Findings suggest that foreign luxury retailers incorporate this by undertaking heavy marketing communication strategies to build up brand awareness and increase brand loyalty. The flagship store was reaffirmed as a highly effective marketing strategy to increase brand awareness in China, as 20 of the 22 companies set up either directly-owned flagship stores, or sizable standalone boutiques managed by their partners. For larger foreign luxury retailers, the flagship store is highly associated with a luxury brand's prestige and status in China. Flagship stores offer luxury retailers the perfect marketing tool for tangibly establishing brand image and for organizing events, affirming previous studies on the role of flagship stores in marketing communication of luxury retailers' brand image (Joy, Wang, Chan, Sherry, & Cui, 2014; Moore, Doherty, & Doyle, 2007). PR events are a very popular marketing communication tactic – for example, store openings, special fashion shows, exhibitions that showcase the brand's heritage and craftsmanship, and charity events. Flagship stores were utilized to foster brand loyalty by providing a high level of personalized service, offering special collections of products exclusive to China, and educating consumers on brand heritage and quality through in-store workshops and small exhibitions. The current study extends previous literature by emphasizing the importance of setting up flagship stores in China not only at national, but also regional level, given the size of the market. Setting up regional flagship stores as a key strategy for success in China was affirmed by all respondents, including a leading British luxury retailer (Company 2), whose CEO spoke of the PR benefits of store opening events: local media would clamor to report the presence of Chinese celebrity guests at launch parties in second or third tier cities, which promoted the establishment of brand image in regional locations. The Vice President of Company 1 highlighted the importance of regional flagship stores in such a vast marketplace where wealth is not solely concentrated in first tier cities: In such a big country with regional differences, a lot of messages get lost if you only communicate in Shanghai or Beijing, so there's no better way to communicate your brand image via a flagship store. Regional flagship stores are very important in China, especially since wealth is not only concentrated in first tier cities like in most markets. Additionally, flagship stores were found to assist smaller luxury brands to upwardly position their brand, thereby increasing brand Table 2 Interview topic guide. Initial interviews Follow-up interviews Market entry methods Marketing strategies When did your brand enter China? What were your motives for entering China and your expansion process in China? What were your market entry method(s) in China? Advantages and disadvantages of market entry method(s)? Did you have problems with the Chinese government's restrictions on foreign investment? If so, how? Is this your ideal market entry method? If not, what is? Do you plan to change it in the foreseeable future? How different is China compared to your other markets in terms of market entry method selection? Do you have a flagship store in China? If so, how does this benefit your brand in China? Do you have a Chinese website? If so, how does this benefit your business in China? Which specific marketing strategies do you adopt in China? How do they benefit your business in China? In terms of management, do you hire local Chinese or transfer from your home country? What are the advantages and disadvantages of this approach? Do you manage your Chinese operation directly from your head office, or does it operate alone as an independent market, or is it managed from your head office in Asia? What are the advantages and disadvantages of this approach? Has your business performance in China achieved your expectation so far? Based on your company's experience, what are the elements for achieving success in the Chinese luxury market? Based on your company's experience, what are the differences between second tier cities (e.g. Hangzhou, Nanjing) and major cities (e.g. Shanghai, Bejing)? What problems are you facing/have you faced in China? How does your company deal with the Chinese business culture? How different is managing your Chinese operation compared to other markets? What is your company's future outlook in China? Any changes in market entry method? Reasons? Benefits of such change in China? Downside of such change? Future change? Other marketing strategy changes Adaptive vs. standardized marketing activities Examples of recent significant marketing activities in China Further development of the Chinese luxury market New opportunities? New obstacles? Have previous obstacles improved? Has the global economy downturn and the rapid change of the Chinese market affected your business in terms of marketing strategies? From a worldwide market perspective and from the Chinese market perspective? Additional comments on future development of luxury fashion retailing/marketing in China? Table 3 Final template for data analysis. Key challenges in the ‘global–local’ dilemma Development pattern Specific strategies 1. Brand management 1.1 Combine centralized and decentralized approach 1.1.1 Hong Kong subsidiary 1.1.2 China subsidiary 1.1.3 Head office control 2. Market entry methods 2.1 Differences between sizes of brands 2.1.1 Wholesale 2.1.2 Franchise 2.1.3 Direct retail as future trend 2.1.4 Use of hotels 2.2 Flagship store 2.2.1 Flagship for image and status 2.2.2 Flagship for CRM 2.3 Mixed entry method 2.3.1 Benefits 2.3.2 Challenges 3. Marketing strategies 3.1 Global strategy 3.1.1. Standardize ad campaign 3.2 Differences between city tiers 3.2.1 Level of adaptation 3.2.2 Volume of events 3.2.3 Country and provincial level flagship stores 3.2.4 Need to adopt Chinese social media channels 3.3 Creativity and innovation 3.3.1 Hgh levels of innovation 360 S. Liu et al. / Journal of Business Research 69 (2016) 357–364

status, especially if located in premium surroundings, such as the same shopping malls as large luxury retailers: With a directly-owned flagship store, we can offer everything to our customers at the same level as our store in Bond Street, which cannot be achieved through wholesaling; if you're not Vuitton or Chanel, it's even more important to look ‘luxury’ as part of the luxury brand identity, to compete with other brands as well as to engage with customers at a deeper level. (Head of Licensing, Company 6)In order to establish a luxury brand's image and to become highly visible in China, one must set up a flagship store … it's also an opportunity for smaller brands to build up their image quickly, especially in less sophisticated cities. (President, Greater China Region, Company 11) Such flagship stores and regional flagship stores can, therefore, act as a platform to represent the luxury brands' image and a location from which to conduct marketing communications in order to drive brand awareness and increase brand loyalty. Smaller luxury retailers can thus benefit from the lack of luxury awareness in the Chinese marketplace and build a reputation more quickly than might be possible in a mature, competitive marketplace with high levels of brand awareness. 5.3. Highly enterprising marketing communications Respondents confirmed the unique characteristics of Chinese luxury consumers that necessitated an adaptive approach to brand communications, consistent with previous literature (Dong and Tian, 2009; Hung et al., 2007). Furthermore, in a follow-up interview in 2014, the lack of cross-generational luxury experience was highlighted as a barrier to brand awareness and loyalty: Most wealthy clientele in China are first generation, therefore their behavior is very different to the West where a lot of wealthy clientele had generations of luxury lifestyle, so it's highly important to show the Chinese the true value of the luxury brands and keep educating them. (Group Senior Executive for Asia Pacific, Company 3) Linking to the discussion above, flagship stores also enable larger retailers to experiment with entreprising communication strategies. In the case of a key brand from companies 3 and 18 (both very small in size internationally compared to their counterparts, but highly recognizable in China), both indicated that their flagship stores enabled them to experiment with more effective customer relationship management in China, such as opening up private clubs within their flagship stores. Many luxury retailers are entreprising in communications in the Chinese market to an extent not seen in other markets, for example by launching occasional special collections or products for China, through active social media involvement, PR events and TV advertising. Innovation in marketing communication attracts consumer attention and PR benefits, thereby building brand awareness. In the case of established retailers, such activities maintain market leader position by showing artistic innovation and special concern for Chinese consumers, as a means of increasing brand loyalty: Chinese are very demanding shoppers… we're always trying our best to offer consumers better shopping experience than others… We'll soon improve our aftersales service by offering free repair service for customers who purchased our products outside, because the Chinese buy a lot abroad. (Vice President, Company 1)We had a big fundraising event for families who suffered in the Sichuan earthquake…Chinese people very much appreciated that, it showed the company cares about the people in China. It's very innovative because corporate responsibility was a new thing back in 2008 in China. (International Group Marketing Director, Company 5)We are very discreet in other markets, but our partners insist on the need to adapt in China that we must conduct more events to communicate with Chinese consumers. (Managing Director, Company 2) Educating the consumer by way of exhibitions that highlight the artisanal techniques and craftsmanship underpinning luxury products, as well as the legendary roots of the brand, is an effective way of building brand awareness and communicating the symbolic authority and intangible benefits (Kapferer, 2014). This also underpins the shift from the luxury market's traditional conspicuous consumption model focused on the product, to an experiential sensibility defined by the consumer (KPMG, 2013; Atwal & Williams, 2009; Wiedman, Hennigs, and Siebels, 2007). Within this shift, retailers create a highly personalized service for the most important clients of the firm, which goes far beyond products and retail service but instead aims to closely integrate the brand within the consumer's lifestyle, as indicated in one of the follow-up interviews in 2014: With very important clients, our aim is to make them ‘dream’, providing not just goods and services that can bought in-store, but an emotional connection … We also propose a program for them when they're traveling …if for holiday in Paris for example, we'll entertain them, open doors for them that they won't be able to by themselves, such as some very private leisure activities. (CRM Manager, Company 1) Despite extensive evidence of such highly adaptive tactics, there was a keen awareness of the need to retain a degree of consistency in overall global positioning. In the following subsections, the overarching need for standardization of brand positioning is discussed, with particular reference to the nuances that exist between different city tiers. 5.4. Standardization and the dangers of over-localization Strategies for brand communications in the Chinese marketplace require careful adaptation of global consumer culture positioning that acknowledges the traditional values of Chinese luxury consumers. Respondents identified the danger of over-localization, in terms of compromising the standardized international luxury brand image. Over-localization could result in inadvertently patronizing consumers or causing consumer confusion as regards brand identity and country of origin. The creation of special collections for the Chinese marketplace should be carefully considered, in the words of one respondent: Companies really must consider China as part of the global community; it's silly to become the ‘clichéd Chinese’ because Chinese hate gimmicks…putting a dragon on an Aston Martin or Ralph Lauren doing a whole cheongsam dress collection is considered as patronizing by Chinese! (President of Greater China Region, Company 11) The Commercial Area Manager in Company 4 praised Cartier's successful adaptation of its series of exhibitions across China, noting how the exhibition was educational as well as culturally influential. It was adapted to Chinese regional heterogeneity by being curated in different ways in different provinces, and then combining all of them to create a complete history of the brand that underpins its heritage and craftsmanship. In contrast, the efforts of Emporio Armani in creating a separate seasonal ad campaign for the Greater China region, using exclusively Chinese celebrities, were strongly criticized: It's a terrible mistake to over-localize. We did some consumer research about Emporio Armani after the company used two Hong Kong celebrities to front their campaign in the Greater China Region, and the mainland Chinese thought it's a completely different brand to the Armani brand, and thought the brand is of Hong Kong origin…So it's clear that the brand image must be consistent internationally to avoid confusion over brand identity. (International Marketing Director, Company 14) The need for global consistency was confirmed by the respondent from Company 4, who explained that over-localization could S. Liu et al. / Journal of Business Research 69 (2016) 357–364 361

compromise brand image, especially for well-traveled Chinese consumers with greater brand knowledge (cf. KPMG, 2013): We do very strong business in China, but we also do very strong business with Chinese people who travel abroad; so it's very important we have a centralised strategy that can affect brand image, like advertising campaign and merchandising strategy. (Commercial Area Manager, Company 4) While Hung et al. (2007) indicate that a local (Asian) look enhances consumer identification among Chinese female consumers, this study's findings suggest that Chinese luxury consumers prefer to be considered part of a global elite, rather than targeted as a discrete sub-set. This supports the existence of a global luxury culture whose members identify with similar people around the world to a greater extent than with those in their own community (Westjohn et al., 2012; Zhang & Khare, 2009). This sense of belonging to a global luxury elite overrides the influence of cultural congruity on perceived value, which is usually seen in consumers from countries where national identity is high (Steenkamp & Geyskens, 2006). Recent evidence suggests Chinese consumers are moving further towards consumerism, increasing demand for a more Western look in TV ads (Morris, 2014). The 2014 follow-up data emphasized that the sophisticated Chinese luxury consumers in first tier cities are truly ‘global’ consumers, thus the need to include China as part of the global marketing strategy is inevitable. Such globally-oriented consumers are more likely to hold positive attitudes towards global brands rather than local brands (Alden et al., 2006; Zhang & Khare, 2009). With their facets of heritage and history, luxury brands also benefit from a strong connection to country of brand origin (COO). As COO is a key contributor of consumer value perception for luxury brands, the danger of causing confusion by over-localization is very real. Hence, marketing strategies that can affect brand image, such as seasonal ad campaigns and bi-annual key collections, should be strictly centralized in all luxury companies as part of a successful long term brand management strategy to maintain luxury image and positioning (Kapferer & Bastein, 2009). Company 1 highlighted the decision to include a famous Chinese actress as part of their global ad campaign, rather than in a separate campaign for China. However, the strategic implication of the global luxury culture in China is that brand managers must look to wider trends than those within the Chinese marketplace: High net worth individuals in China are the true global luxury consumers, they travel abroad often, especially those in their 20s and 30s, have much more choice and more information received from digital media, they're willing to experience more with everything. So the brands need to react very fast, be constantly in tune with the new trends globally rather than only focusing on China. Nevertheless, it's about the balance, brands shouldn't let consumers dictate strategies, but should use strategies to enforce products and brand image. (Group Senior Executive for Asia Pacific, Company 3) However, caution is expressed about moving towards a model of cocreation of brand experience that is too consumer-led, which may dilute the luxury status of the brand (cf. Kapferer & Bastein, 2009; Tynan et al., 2010). 5.5. Regional nuances between city tiers Regarding retail store expansion, the follow-up interviews in 2014 acknowledged the dangers of brand over-exposure, which subsequently led to label fatigue, especially as Chinese consumers are not brand loyal. Thus, established brands in China should focus on the development of existing flagship stores in first tier cities and prominent second tier cities only, to maintain status and exclusivity. Respondents from the 2014 interviews explained the nuances between first tier and lower tier consumer behaviors: We've noticed a significant change of behavior in Chinese consumers since 2012: the Chinese from first tier cities are very sophisticated, comparable to the West; but most second tier cities and third tiers are still highly conspicuous. So marketing strategies must be more targeted than before. (European Marketing Manager (responsible for Chinese tourists), Company 10)A few superbrands made the mistake of opening too many shops in Shanghai, thinking they must catch up with the demand of the market without realizing the rapid development of Chinese consumers' level of sophistication, so now they got tied up to some new shopping malls and cannot get out. Store opening must be carefully planned, some brands are suitable for heavy expansion but some aren't, especially the ones who are at the highest position. (Commercial Area Manager of Greater China, Company 4) Respondents from other companies expressed similar views, and the interviewees from smaller companies emphasized how the rapid development of consumer sophistication provides smaller luxury brands with greater opportunities in first tier cities. Therefore, the already established luxury brands in China must seek out more innovative strategies to increase brand loyalty and maintain consumer interest. Data from 2009 to 2011 identified the need for a high level of adaptation in marketing communication, and other than the use of flagship stores to support brand image building, PR events are the most significant marketing communication strategy in China. Examples of entreprising marketing communications include conducting special projects for the Beijing Olympics by one company, and enclosing brochures within top fashion magazines by another to enhance direct communication with consumers. Neither of these tactics were conducted in either company's established markets. The follow-up interviews in 2014 indicated the need for an even higher level of innovation in first tier cities. For example, respondents emphasized the importance of exhibitions and PR events aiming at brand education and promotion of heritage and craftsmanship. Citing examples of innovative marketing communications events such as Burberry's 3D holographic store opening show and digitally enhanced flagship stores, respondents spoke of the viral marketing and reputation building benefits of activities that emphasized brand heritage and storytelling, hence increase brand awareness and loyalty among Chinese consumers: Our research found that Chinese consumers from first tier cities are more interested in story and quality than logos. (Brand Manager of Greater China Region, Company 20)We are constantly thinking about the innovative ways to excite our Chinese consumers… events are not only exciting and a great tool for product promotion, they also intrigued and educated Chinese consumers about the Western luxury heritage that they missed out on during that era. (Commercial Area Manager of Greater China Region, Company 4) Other respondents shared similar views that being innovative is particularly important in first tier cities. Similarly, Kapferer (2014) notes how luxury brands' associations with high art has longer-term benefits than the comparatively superficial tactic of celebrity endorsement. Innovating with art cements the luxury brand qualities of “timelessness, substance, and cultural and temporal substance” (p.376). The followup interviews in 2014 also confirmed that using innovative marketing communications is highly useful in building luxury brand status for smaller companies, which is otherwise difficult, and can subsequently upward position smaller brands and enable them to to compete with large brands in first tier Chinese cities: The consumers in first tier cities are very sophisticated now, so it creates opportunities for smaller brands if you can come up with unique 362 S. Liu et al. / Journal of Business Research 69 (2016) 357–364

and interesting strategies to promote the brand. For instance, we created a lively … event … and it really enhanced our luxury image and enabled us to compete with the larger menswear brands in China. (CEO, Company 2) Table 4 summarizes the findings. 6. Conclusions and implications The importance of balancing the ‘global-local dilemma’ is significant for foreign luxury retailers in China. Adaptation must take place within the overarching constraint of a globally standardized brand image. The study affirms Chevalier and Lu's (2010) view on the importance of adapting marketing communication strategy for entry into China and highlights in particular the significance of incorporating Hong Kong as an adaptive management operational control strategy. Additionally, it indicates the pattern of future luxury retailing in China, as an increasingly decentralized operational management strategy in mainland China, and a centralized strategy in the home market to maintain a standardized global luxury brand image. While seasonal ad campaigns should adhere to a global consumer culture positioning (Alden et al., 1999) in order to maintain the global luxury brand image, other marketing communications tactics can be highly adaptive, in particular setting up flagship stores at provincial level to maximize effective marketing communication impact across the vast Chinese market, and becoming more entreprising in PR tactics to further enhance brand awareness and loyalty. Although the importance of adapting marketing strategies to enhance brand image has been acknowledged in previous studies on luxury retailing (Chevalier & Lu, 2010; Kapferer & Bastein, 2009), the high level of adaptation of communication strategies was not previously seen in other markets and can thus be concluded as being distinctive to China, especially considering that many of these innovative strategies were rarely conducted even in the brands' home markets. Such high innovation can aid smaller brands' positioning to create business opportunities in the sophisticated first tier cities. The categorization of the Chinese cities according to tier level provides a valuable overview to understanding the characteristics of the current Chinese luxury market, and the consequent need for some degree of adaptation. The study identifies a number of strategic implications. Firstly, luxury retailers should aim to secure top tier luxury consumers' loyalty, as opinion leaders and influencers, through effective customer relationship management. Secondly, innovative PR events such as exhibitions are the most subtle and effective way of marketing communication, incorporating leisure activity, entertainment, education and promotion. The luxury flagship store remains the most important strategic tool for luxury brands because it provides the highest level of direct consumer engagement and the best location for showcasing the brands' heritage, story and creativity. Nevertheless, being innovative in store design and service are key in order to stand out from the competition (cf. Le Fort, 2014). Finally, Chinese consumers, especially in the most sophisticated first tier cities, want to be included as part of the global community, therefore it is important to incorporate China into the global marketing communications strategy. At the same time, luxury companies must recognize the significant differences emerging within the Chinese market, and subsequently should design targeted marketing communication programs for different tiers of cities and different regions. Based on these findings, luxury strategies in China may be categorized into three separate markets. In first tier cities (especially Beijing and Shanghai), sizeable flagship stores showcase the brand image, with entreprizing marketing communication strategies focused upon brand education and innovation. Brand development is regarded at the same level as mature markets, hence these cities are incorporated into the brand's long term management (e.g. innovative communication strategies in China that benefit an international audience). In second tier cities (especially the most prominent, such as Hangzhou and Chengdu), regional flagship stores are combined with heavy marketing communications (albeit quite traditional) strategies such as store events and celebrity endorsement are still relevant. Greater levels of conspicuous consumption compared to first tier cities necessitate a separate communication strategy. Third tier cities are still relatively underdeveloped and may be unsuitable for the most exclusive luxury brands. For future research, the insights drawn from this study could be empirically tested and theoretically extended. In-depth case studies could be conducted on luxury retailers that achieved success (or failure) in China. Additionally, conducting comparison studies across different product categories could contribute to the understanding of the different sectors within the luxury industry. Finally, testing in other emerging markets, especially the Asian emerging markets surrounding China that share similar cultures and consumer behaviors, would potentially further identify the nuances that exist in global markets. A further aspect worth investigating would be the differences in strategy between Western luxury brands and those originating from emerging nations. References Aaker, D. A. (1991). Managing Brand Equity. New York: The Free Press. Alden, D. L., Steenkamp, J. -B. E. M., & Batra, R. (1999). Brand positioning through advertising in Asia, North America, and Europe: The role of global consumer culture. Journal of Marketing, 63, 75–87. Alden, D. L., Steenkamp, J. -B. E. M., & Batra, R. (2006). Consumer attitudes toward marketplace globalization: Structure, antecedents and consequences. International Journal of Research in Marketing, 23, 227–239. Alexander, N., & Doherty, A. M. (2010). International retail research: Focus, methodology and conceptual development. International Journal of Retail & Distribution Management, 38(11/12), 928–942. Atwal, G., & Williams, A. (2009). Luxury brand marketing: The experience is everything. Journal of Brand Management, 16(5/6), 338–346. Bain & Co. (2011). China Luxury Market Study. Retrieved from http://www.bain.com/ publications/articles/2011-china-luxury-market-study.aspx (accessed 1.2.14) Bain & Co. (2012). Luxury Good China Market Study. Retrieved from http://www. bain.com/publications/articles/luxury-goods-china-market-study-2012.aspx (accessed 1.2.14) Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. Chadha, R., & Husband, P. (2006). The cult of the luxury brand: Inside Asia's love affair with luxury. London: Nicholas Brealey International. Chevalier, M., & Lu, P. (2010). Luxury China: Market opportunities and potential. Hoboken: John Wiley & Sons (Asia). Table 4 Summary of results. Key themes Summary of results Use of Hong Kong as managerial bridgehead Location acts as a two-way window to China and the West Availability of talented human resources and efficient distribution systems Use of flagship stores for brand tangibility and relationship building Create brand awareness by conducting PR activities within the store Increase brand loyalty through educational activities and highly personalized levels of customer service Provincial level flagship stores maximize communications impact across a vast market Highly enterprising marketing communications Varied activities which are not always conducted in brand's home market Key to educating first generation luxury consumers Standardization and the dangers of over-localization Key elements of marketing communications must be standardized globally Overlocalization could inadvertently patronize consumers who perceive themselves as part of a global elite Overlocalization could cause confusion over brand COO Regional nuances between city tiers High levels of innovation are needed to communicate with highly sophisticated consumers in first tier cities More cautious approach in lower tier cities to avoid over-exposure of brand S. Liu et al. / Journal of Business Research 69 (2016) 357–364 363

Cleveland, M., Papadopoulos, N., & Laroche, M. (2011). Identity, demographics, and consumer behaviors: International market segmentation across product categories. International Marketing Review, 28(3), 244–266. Deloitte China (2011). China Powers of Retailing 2011. Retrieved from http://www. deloitte.com/assets/Dcom-China/Local%20Assets/Documents/Industries/Consumer% 20business%20and%20transportation/cn_cbt_ChinaPowersRetailing2011_181011.pdf (accessed 1.2.14) Dion, D., & Arnauld, E. (2011). Retail luxury strategy: Assembling charisma through art and magic. Journal of Retailing, 87, 502–520. Dong, L., & Tian, K. (2009). The Use of Western Brands in Asserting Chinese National Identity. Journal of Consumer Research, 36(3), 504–523. Dubois, B., & Paternault, C. (1995). Understanding the world of international luxury brands: The “dream formula”. Journal of Advertising Research, 35(4), 69–76. Fastoso, F., & Whitelock, J. (2007). International advertising strategy: The standardization question in manager studies. International Marketing Review, 24(5), 591–605. Fionda, A. M., & Moore, C. M. (2009). The anatomy of the luxury fashion brand. Journal of Brand Management, 16(5/6), 347–363. Goldman, A. (2001). The transfer of retail formats into developing economies: The example of China. Journal of Retailing, 77, 221–242. Goulding, C. (2005). Grounded theory, ethnography and phenomenology: A comparative analysis of three qualitative strategies for marketing research. European Journal of Marketing, 39(3/4), 294–308. Harjani, A. (2013). Logo fatigue? Chinese now want understated luxury. Retrieved from http://www.cnbc.com/id/101216688 (accessed 1.9.14) Hennigs, N., Wiedmann, K. P., Klarmann, C., Strehlau, S., Godey, B., Pederzoli, D., et al. (2012). What is the value of luxury? A cross-cultural consumer perspective. Psychology and Marketing, 29(12), 1018–1034. Hung, K. H., Li, S. Y., & Belk, R. W. (2007). Glocal understandings: Female readers' perceptions of the new woman in Chinese advertising. Journal of International Business Studies, 38(6), 1034–1105. Joy, A., Wang, J. J., Chan, T. -S., Sherry, J. F., Jr., & Cui, G. (2014). M(Art)worlds: Consumer perceptions of how luxury brand stores become art institutions. Journal of Retailing, 90(3), 347–364. Kapferer, J. -N. (1992). Strategic Brand Management: New Approaches to Creating and Evaluating Brand Equity. New York: The Free Press. Kapferer, J. -N. (2014). The artification of luxury: From artisans to artists. Business Horizons, 57, 371–380. Kapferer, J. -N., & Bastein, V. (2009). The specificity of luxury management: Turning marketing upside down. Journal of Brand Management, 16(5/6), 311–322. King, N. (2012). Doing template analysis. In G. Symon, & C. Cassell (Eds.), Qualitative organizational research: Core methods and current challenges. London: Sage. KPMG (2013). Global reach of China luxury. Retrieved from http://www.kpmg.com/DE/ de/Documents/The_Global_Reach_Of_China_Luxury.pdf (accessed 1.12.14) Le Fort, C. (2014). In China, Hermès paces itself. Retrieved from http://www. businessoffashion.com/2014/09/china-hermes-paces.html (accessed 19.9.14) Li, G., Li, G., & Kambele, Z. (2012). Luxury fashion brand consumers in China: Perceived value, fashion lifestyle, and willingness to pay. Journal of Business Research, 65, 1516–1522. Lu, X. (2011). Termination of the war of strong luxury brands in China affluent consumer market. Luxury special sectionHarvard Business Review. (pp. 53–58), 53–58. Lu, X. (2012). Market entry strategies of international luxury brands in China. Finance & Trade Economics, 8, 105–112. Lu, X., & Pras, B. (2011). Profiling mass affluent luxury goods consumers in China: A psychographic approach. Thunderbird International Business Review, 53(4), 435–456. McKinsey (2012). Luxury without borders: China's new class of shoppers take on the world. Retrieved from http://solutions.mckinsey.com/insightschina (accessed 1. 2.14) Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. Thousand Oaks: Sage. Mintel (2012). Global Luxury Goods Retailing. London: Mintel. Moore, C. M., Doherty, A., & Doyle, S. (2007). Flagship stores as a market entry method: The perspective of luxury fashion retailing. European Journal of Marketing, 44(1/2), 139–161. Moore, C., Fernie, J., & Burt, S. (2000). Brands without boundaries — The internationalisation of the designer retailer's brand. European Journal of Marketing, 34(8), 919–937. Morris, R. (2014). The Western stars of Chinese adverts. Retrieved from http://www.bbc. co.uk/news/magazine-29183095 (accessed 19.9.14) O'Grady, S., & Lane, H. W. (1996). The psychic distance paradox. Journal of International Business Studies, 27(2), 309–333. Patton, M. Q. (2002). Qualitative evaluation and research methods. London: Sage. Red Luxury (2012). China's boredom with bling? Retrieved at http://red-luxury.com/ brands-retail/chinas-boredom-with-bling (accessed 1.2.14) Salmon, W. J., & Tordjman, A. (1989). The internationalisation of retailing. International Journal of Retail & Distribution Management, 4(2), 3–16. Shukla, P., & Purani, K. (2011). Comparing the importance of luxury value perceptions in cross-national contexts. Journal of Business Research, 65(10), 14–17. Steenkamp, J. -B. E. M., & Geyskens, I. (2006). How country characteristics affect the perceived value of web sites. Journal of Marketing, 70(3), 136–150. Tynan, C., McKechnie, S., & Chhuon, C. (2010). Co-creating value for luxury brands. Journal of Business Research, 63, 1156–1163. Westjohn, S. A., Singh, N., & Magnusson, P. (2012). Responsiveness to global and local consumer culture positioning: A personality and collective identity perspective. Journal of International Marketing, 20(1), 58–73. Wiedman, K. P., Hennigs, N., & Siebels, A. (2007). Measuring consumers' luxury value perception: a cross-cultural framework. Academy of Marketing Science Review, 7, 1–21. Yeĝenoĝlu, M. (2005). Cosmopolitanism and nationalism in a globalized world. Ethnic and Racial Studies, 28(1), 103–131. Yu, J. (2012). Hong Kong still No 1 for mainland Chinese tourists. Retrieved from http:// www.scmp.com/news/hong-kong/article/1081729/hk-still-no-1-mainland-tourists (accessed 20.11.14) Zhan, L., & He, Y. (2012). Understanding luxury consumption in China: Consumer perceptions of best known brands. Journal of Business Research, 65(10), 1452–1460. Zhang, Y., & Khare, A. (2009). The impact of accessible identities on the evaluation of global versus local products. Journal of Consumer Research, 36(3), 524–537. Zhang, B., & Kim, J. -H. (2013). Luxury fashion consumption in China: Factors affecting attitude and purchase intent. Journal of Retailing and Consumer Services, 20, 68–79. Zhou, N., & Belk, R. W. (2004). Chinese consumer readings of global and local advertising appeals. Journal of Advertising, 33(3), 63–76. 364 S. Liu et al. / Journal of Business Research 69 (2016) 357–364

Explaining and predicting purchase intentions following luxury-fashion brand value co-creation encounters Eunha Choi a , Eunju Ko a, ⁎, Angella J. Kim b a Department of Clothing & Textiles, Yonsei University, Samsung Hall 313, Yonsei University 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, Republic of Korea b Department of Apparel Merchandising & Management, California State Polytechnic University Pomona, 3801 West Temple Avenue, Pomona, CA 91768, USA article info abstract Article history: Received 1 July 2015 Received in revised form 1 February 2016 Accepted 1 April 2016 Available online 7 June 2016 This study investigates luxury brand value co-creation processes. The authors use a mixed method approach to examine the relationships among encounter attributes, consumer value, brand value, and purchase intentions. Indepth interviews with Chanel customers reveal their reactions to their digital marketing and art marketing value co-creation experiences. Structural equation modeling analysis shows that the experiences influence customer value, customer value then influences brand value, and brand value then influences purchase intentions. Findings suggest that marketers should emphasize emotional and experiential aspects during customer–brand interaction encounters. © 2016 Elsevier Inc. All rights reserved. Keywords: Art marketing Digital marketing Encounter Luxury Value co-creation 1. Introduction In modern marketing, value is sometimes one of the most important elements of brand success. Value comes from consumers' perceptions regarding whether the brand has provided positive shopping experiences (Schechter, 1984) and has fulfilled their personal expectations. When customers perceive that a brand offers what they need and want, value is created (Gilbreath, 2010). In the past, companies have created value by providing goods and services that meet customers' expectations and meet company goals. The consumer role was to passively accept and then evaluate company offerings. Now the trend is toward emphasizing customers in planning and executing business activities. Consequently, value is created via business-to-customer (B2C) interactions. Brand activities involving customer participation allow customers to engage with brands to assure that the brands improve their lives (Mun, 2011). The most important element of the value co-creation process in which customers actively participate in brand-related activities, is an encounter, where learning and communication interactions between brands and customers take place (Koo & Rha, 2012; Woodside & Ko, 2013). Various brand-related activities such as marketing events allow interactions between brands and customers for creating value (Koo & Rha, 2012) and are important value co-creation encounters. The luxury fashion market has been particularly attracted to value co-creation. The most challenging task for the long-lasting success of luxury fashion brands is to communicate value by providing novel value and building strong customer bonds (Choi, Ko, Kim, & Mattila, 2015; Kim & Ko, 2010a, 2012) through unique interactions differentiating them from the competition. To create novel values, luxury fashion brands have focused their marketing efforts on digital communication and art encounters (Heine & Berghaus, 2014; Tynan, Mckechnie, & Chhuon, 2010), using Internet, social media, and digital technologies as marketing tools (Dhaoui, 2014). Customers can learn about brands for making smarter consumption decisions through B2C interactions via various online platforms. In addition to using digital technologies, luxury fashion brands collaborate with famous artists to launch limited editions as a way of integrating art into various brand-related activities (Kim, Ko & Lee, 2012). For example, the Murakami Takashi Louis Vuitton collection and Chanel's sponsorship of artistic charity events for the Foundation of AIDS Research were innovative marketing strategies combining art and luxury to enhance brand images and satisfy customers' emotional needs. The focus on digital technologies and art provides opportunities for customers to participate in brand-related activities and create additional value. Although many companies encourage customers to participate in value creation activities, empirical research is lacking to show how B2C interactions co-create values. The process of value co-creation and its impact on customers is particularly critical for the success of luxury fashion brands. Given the limited empirical study in the area of luxury brand value co-creation, this study attempts to explore luxury brand value co-creation using a mixed method approach. Study 1, a qualitative investigation focusing on luxury fashion brand marketing activities as Journal of Business Research 69 (2016) 5827–5832 ⁎ Corresponding author. E-mail addresses: [email protected] (E. Choi), [email protected] (E. Ko), [email protected] (A.J. Kim). http://dx.doi.org/10.1016/j.jbusres.2016.04.180 0148-2963/© 2016 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Journal of Business Research

encounters of value co-creation, explores value co-creation encounters and their attributes. Study 2, a follow-up, quantitatively examines the relationships among encounter attributes, customer value, brand value, and purchase intention to explain the process and influence of value co-creation encounters. 2. Value co-creation Some researchers view the concept of value co-creation from various angles, but the common understanding is that customers actively participate in creating brand values (Payne, Storbacka, Frow & Knox, 2009; Prahalad & Ramaswamy, 2004; Vargo & Robert, 2004) by interacting through marketing encounters such as product or brand-related activities. Koo and Rha (2012) explain that interactions between brands and customers are independent but recursively interrelated value co-creation processes. That is, customers experience the brand and then form perceptions of brand values. Brand managers then learn about the perceptions and selectively apply them to develop better products or services. When customers accept newly provided products or services, they ultimately contribute to value co-creation. Payne, Storbacka, and Frow (2008) emphasized the importance of creating experiential relationships between brands and customers. Value co-creation means that customer emotional, cognitive, and behavioral experiences are the basis of the value, impressions, recognition, and internalization they accord to the brand. 2.1. Encounters of value co-creation An encounter is the point of interaction, touch point, or contact with a brand that then determines the value co-creation. Products or services customers purchase from the brand are the basis of value co-creation encounters (Koo & Rha, 2012) in addition to physical experiences in retail venues, promotional events, or online websites where brands and customers communicate and interact. Value co-creation encounters include both cognitive and emotional experiences (Anderson, 1995; Gentile, Spiller, & Noci, 2007; Payne et al., 2008; Schmitt & Simmonson, 1997). A brand's social media platform is an example of a value co-creation encounter where the brand and its customers interact cognitively to learn about and share brand/productrelated information; they also interact emotionally in sharing their common interests. 2.2. Luxury fashion brand value co-creation In the luxury market, competition means that brands must distinguish their unique characteristics through distinctive brand-related marketing (Kim, Ko, Xu & Han, 2012; Ko & Megehee, 2012). Luxury fashion brands widely adopt value co-creation encounters to provide novel interactions and innovative experiences (Lee, Ko, & Megehee, 2015). Tynan et al. (2010) qualitatively examined corporate marketing activities in the luxury brand market. They found that important corporate art-related activities that contribute to value co-creation include art-inspired products, artistically constructed store facilities, artist collaborations, arts and exhibitions (e.g., photography, movies), and artist support. Important brand networking/communication activities include digital communications (e.g., online communities, blogs, SNSs, mobile applications), oneto-one sales representative–customer interactions, and customer relationship management (Park, Song, & Ko, 2011). 3. Customer value and brand value Brand–customer interactions yield both customer value and brand value (Gentile et al., 2007; Kim, Ko, Xu et al., 2012). When customers interact with brands, they gain both utilitarian and hedonic values (Addis & Holbrook, 2001). In luxury fashion brand consumption, they particularly derive social, hedonic, conspicuous, and self-expressive values in addition to practical values (Tynan et al., 2010; Vigneron & Johnson, 1999; Lee et al., 2015; Woodside & Ko, 2013). Brand–customer interactions also yield performance-related outcomes including sales and market share, positive brand images, and brand loyalty (Gentile et al., 2007). Brand assets are the intangible and ultimate outcome brands expect to gain in return for their business activities. Yoo and Donthu (2001) identified three dimensions of brand assets: brand loyalty, quality recognition, and brand recognition. They focused on marketing activities as encounters of value co-creation that created brand loyalty and brand recognition. 4. Study 1 Study 1 was a qualitative exploration of luxury fashion brand value co-creation to identify the attributes of luxury fashion brand value co-creation encounters. Luxury fashion brands are defined as brands that have global reputations, inspire customer desires, and are associated with tradition and innovation (Tynan et al., 2010). Hermes, Chanel, and Louis Vuitton were considered because they are commonly ranked in luxury brand indexes (i.e., L2 Think Tank's Gen Y Prestige Brand Ranking's, 2010; Millward Brown's Top 100 Most Valuable Global Brands's, 2012). Chanel was finally selected to represent luxury fashion brands in the research because it best met the selection criteria: 1) the brand is a leader in the luxury fashion industry; 2) the brand has a successfully established cultural tradition; 3) many people can share the brand's culture but the brand will retain its exclusiveness as a luxury brand. 4.1. Sampling and data collection Using purposive sampling, snowball sampling techniques, and in-depth interviews, the researchers recruited ten Chanel customers, in their 20s and 30s, from the Seoul metropolitan area. All participants were active, loyal Chanel customers who were fully aware of the brand culture. They owned at least two leather bags from the brand's main product category and had shoes, apparel, and cosmetics from other categories. They had also participated in Chanel's brand-related special events (see Table 1 for the participant characteristics). Following Patton's (1990) interview guide approach, semi-structured in-depth interviews were conducted regarding encounters of value cocreation and their attributes. The study focused on marketing encounters of value co-creation, so the interviewers asked about customer experiences with the marketing activities. We first asked about Chanel's recent encounters allowing customers to interact with the brand and then asked about interaction experiences. The participants were asked to speak freely about various experiences related to the brand. Interviews were audiorecorded and lasted 1–2 h depending on how much experience the interviewees had to report. 4.2. Results To identify participants' encounter experiences, we asked, “Does Chanel use remarkable attempts to deliver customer value?” “What brand-related activities do you and Chanel use to communicate/interact?” “How do you share your experiences with Chanel?” Participants emphasized Chanel's digital marketing and art marketing activities, as expected. 4.2.1. Digital marketing Participants said that their recent interaction with Chanel took place through online communication value co-creation encounters. Participants' experience communicating with or about Chanel online through digital communication encounters were classified as brand-driven platforms (i.e., official home pages, official blog pages, and official social network services sites; SNSs) and customer-led online communities and personal blogs (Park et al., 2011). Participants visited both brand-driven and customer-led online communication venues almost daily and participated in brand-related communications. Their activities included 5828 E. Choi et al. / Journal of Business Research 69 (2016) 5827–5832

viewing contents posted by the brand or other users and uploading personal photos of their brand experiences or writing comments. Participants reported that they constantly see the brand via their SNSs, and consequently feel more connected to the brand and find it interesting to communicate online with others who share similar interests. For example, Chanel is my Facebook friend. Because I see Chanel's new posts nearly every day through my timeline, I feel connected to Chanel all the time. When people write comments on my posts on Chanel's blog, I reply and talk to them. It's fun! 4.2.2. Art marketing Participants most frequently recalled Chanel's art director Karl Lagerfeld's Photo Exhibition held in Seoul as one of the remarkable brand-related activities that enabled customers to interact with the brand. Eight of the ten participants attended the exhibition and said they felt closer to the brand after attending and connecting through the Lagerfeld photographs. Participants also mentioned Chanel's fashion shows as another remarkable brand-related activity that showed Chanel's artistry and luxury. For example, Karl Lagerfeld is a designer who uses modern and sophisticated methods to mix arts and culture with his own brand and fashion. Chanel's fashion shows always remind me that Chanel is an artistic and luxurious brand. The stage is so beautiful and diverse that it is an art object itself. In both cases, art was the key element allowing the brand and customers to connect and interact. Luxury fashion brands have historically used art to attract customers, satisfy aesthetic desires of elite customers, provide customers' opportunities to express their taste, communicate their brand philosophy and culture, and provide information about the brand. 4.2.3. Value co-creation encounter and its attributes During the interviews, participants shared their experiences interacting with Chanel. As expected, they mentioned Chanel's recent digital marketing and art marketing encounter activities and categorized their experiences as having either cognitive or emotional effects, as seen in previous research (Payne, Storbacka et al., 2009; Ko & Megehee, 2012). Interviewees reported that their cognitive needs were met when brand-related information was shared in both digital marketing and art marketing activities. In terms of information content, art marketing cases focused both on art/culture-related knowledge and brand/ product-related information. Two encounters with digital and art marketing also included emotional attributes that stimulated positive emotions and bonding with the brand. Table 2 is a summary of Study 1 results. 5. Study 2 Study 2 was undertaken to quantitatively analyze luxury brand value co-creation by examining the relationships among customer value, brand value, and purchase intention attributes of the value co-creation encounter. 5.1. Hypotheses development Building on the findings of the qualitative Study 1 and a review of extant literature concerning value co-creation, we developed and analyzed a model demonstrating the proposed relationships among the constructs of luxury brand value co-creation. As in Study 1, Study 2 examined cognitive and emotional attributes in digital marketing and art marketing encounters. The final model includes cognitive and emotional attributes as luxury brand value co-creation encounters; social/conspicuous, aesthetic/self-expressive, experiential/hedonic, and quality value as consumer value; and brand loyalty and brand recognition as brand value. Purchase intention was treated as the consequence of the value co-creation (see Fig. 1 for the proposed conceptual model). Brand–customer interactions yield both customer and brand value (Addis & Holbrook, 2001; Gentile et al., 2007). The value creation process involves experiential interactions with various brand-initiated encounters; encounter attributes influence customer experience (Gentile et al., 2007). Cognitive and emotional attributes are the most emphasized encounter factors influencing customer satisfaction, attitudes, judgment, and behavior in relation to the brand (Payne, Storbacka et al., 2009; Kim, Ko, Xu et al., 2012), which leads to the first and second hypotheses. H1: The higher the evaluation of the encounter attribute, the higher the consumer value. H2: The higher the evaluation of the encounter attribute, the higher the brand value. As mentioned, customer value and brand value are created in value co-creation processes (Gentile et al., 2007). When customers evaluate their brand–customer interactions, they will perceive worth in relation to their experiences, which then results in brand value, including brand loyalty and brand awareness (Schechter, 1984). Yang (2003) did not directly examine the influence of consumers' perceived consumption value on brand loyalty, but found a significant relationship in the value co-creation context. Park's (2009) research on luxury fashion brand attributes and customer-brand relationship revealed that show-off tendency, exclusiveness, product quality, hedonic experience, and self-expansion via consumption significantly influenced brand evaluation. Thus, consider the third hypothesis. H3: The higher the evaluation of consumer value, the higher the brand value. When consumers have positive brand interactions, their satisfaction further leads to desires to purchase and use the brand (Assael, 2005). Similarly, highly satisfied customers are expected to continue consuming the brand, which leads to the final hypotheses. H4: The higher the evaluation of consumer value, the higher the purchase intention. H5: The higher the evaluation of brand value, the higher the purchase intention. Table 1 Characteristics of interview participants. Participants Age Occupation Number of Chanel bags owned Other products owned Case 1 32 Teacher 4 Golf club, shoes, accessories, sunglasses, cosmetics Case 2 32 Secretary 3 Apparel, shoes, accessories, sunglasses, watch, cosmetics Case 3 27 Teacher 2 Shoes, accessories, sunglasses, cosmetics, perfume Case 4 28 Office job 2 Shoes, accessories, sunglasses, cosmetics, perfume Case 5 28 Office job 3 Shoes, accessories, sunglasses, watch, cosmetics, perfume Case 6 27 Office job 5 Shoes, accessories, sunglasses, watch, cosmetics Case 7 30 Lawyer 2 Shoes, accessories, sunglasses, watch, cosmetics, perfume Case 8 28 Accountant 5 Golf club, apparel, shoes, watch, cosmetics Case 9 31 Composer 5 Golf club, apparel, shoes, watch, cosmetics, perfume Case 10 38 Lecturer 5 Apparel, shoes, watch, accessories, sunglasses, cosmetics E. Choi et al. / Journal of Business Research 69 (2016) 5827–5832 5829

5.2. Measures The first section of the questionnaire presented questions concerning participants' experience with Chanel's Facebook page and Karl Lagerfeld's photo exhibition. Participants were randomly assigned to view one of the two stimuli and asked to respond to the questions measuring cognitive and emotional attributes. The encounter attributes measure included a three-item measure of cognitive attribute (Payne, Storbacka et al., 2009; α = 0.86) and a four-item measure of emotional attribute (Payne, Storbacka et al., 2009; α = 0.92). Items to measure cognitive attribute were: “I can gain information about the brand while participating in this event; I can experience various brand aspects while participating in this event; I can share knowledge about the brand with others while participating in this event.” Items to measure emotional attribute were: “I can better understand emotions conveyed by the brand after participating in this event; This event is emotionally interesting; This event is fun; This event stimulates my emotions.” In the second section, participants were asked to respond to the measures of variables under investigation. We modified scales of customer value from previous research (Park, 2009; Vigneron & Johnson, 1999; Yang, 2003) to measure the four constructs identified for customer value. The 15-item measure included four items for social/ conspicuous value, three items for aesthetic/self-expressive value, four items for experiential/hedonic value, and four items for quality value. Reported reliabilities of the measures were α = 0.88, α = 0.90, α = 0.85, and α = 0.88 respectively. Brand value was assessed by items used in Yoo and Donthu (2001) and included a two-item measure of brand loyalty (α = 0.89) and a three-item measure of brand recognition (α = 0.80). Purchase intention was assessed by items used in Kim and Ko (2010a, b; α=.92). Encounter attributes, customer value, brand value, and purchase intention were measured on seven-point Likert scales (1 = not at all,7= very much). The last section of the questionnaire assessed luxury purchase behavior and demographic information. Because all multi-item measures were reliable, we averaged items to form a composite measure of each construct. 5.3. Sampling and data collection Recruited to participate in Study 2 through convenience sampling method were women consumers of luxury fashions, especially Chanel, in their 20s and 30s, who live in the Seoul metropolitan area. Participants completed a self-administered survey about reactions to Chanel's Facebook page and Karl Lagerfeld's photo exhibition as encounters of value co-creation. Those two visual stimuli were selected because participants in Study 1 mentioned both as the most impressive brand activities they had experienced. The survey questionnaire was presented in Korean and was distributed both online and offline. 5.4. Participant characteristics The convenience sample consisted of 418 participants, equally distributed in that 209 (50%) were in their 20s and 209 were in their 30s; 63.4% were never married/single; 36.6% held office jobs; 23.3% were students; 14.6% were housewives; 13.8% had professional jobs; 51.4% had purchased luxury fashion brands in the past two years; 48.6% of those had purchased Chanel in particular. Table 2 Results of Study 1. Encounter Attributes Cognitive Emotional Art marketing • Fashion shows, exhibitions, advertising campaigns • Designer activities (i.e., music, photography, architecture) • Exterior and interior store designs • Provides art-related knowledge and cultural information about the brand. • Stimulates and satisfies desires to know about the brand. • Stimulates positive emotions. • Stimulates brand experiences and creates emotional bonds. Digital marketing • Website, SNSs • Blogs, online communities, and SNSs • Mobile applications • Provides brand-related information via various digital platforms. • Allows customers to share brand-related information. • Stimulates sharing of positive emotions through postings and digital platforms. Fig. 1. Proposed conceptual model. 5830 E. Choi et al. / Journal of Business Research 69 (2016) 5827–5832

5.5. Hypotheses test A structural model indicating the hypothesized relationships among encounter attributes, customer value, brand value, and purchase intention was analyzed. The goodness of fit of the model was χ2 = 304.23, df = 23, CFI = 0.93, GFI = 0.93, NFI = 0.92, IFI = 0.93, RMSEA = 0.50, indicating sufficient explanations for the relationships between constructs in the model. Detailed analysis of the path coefficients of the structure model found that higher assessment of encounter attributes consisting of cognitive and emotional attributes had more positive effects on consumer value (β = 0.59, p b 0.001), supporting hypothesis 1. However, the hypothesis regarding relationships between encounter attributes and brand value was rejected. Meanwhile, consumer value generated through value co-creation consisting of social/conspicuous value, aesthetic/selfexpressive value, experiential/hedonic value, and quality value was significantly related to brand values (β = 0.90, p b 0.001). Thus, hypothesis 3 was supported. Regarding the relationships between consumer value, brand value, and purchase intention, only brand value was significantly related to purchase intention, indicating that a higher assessment of brand value positively affected purchase intention (β = 0.41, p b 0.001). The results of hypotheses analysis revealed that attributes of the value co-creation encounter influenced customer value, brand value, and purchase intention. The significant paths in the model indicated that encounter attributes directly influenced customer value, and customer value then transferred to brand value and purchase intention. Fig. 2 provides a visual summary of the results of hypotheses testing. 5.6. Additional analyses Although the structural model examined the proposed relationships among the variables under investigation, it did not capture the influence of specific elements within the constructs examined in the model. To provide more detailed understanding of the influences of value cocreation encounter attributes and the role of each sub-construct (i.e., social/conspicuous, aesthetic/self-expression, experiential/hedonic, quality, brand loyalty, brand recognition), we performed separate regression analyses between the sub-constructs of variables. Regarding the influence of encounter attributes on customer value, emotional attribute had significant influences on every sub-construct of consumer value examined (social/conspicuous: β = 0.43, p b 0.001; aesthetic/selfexpressive: β = 0.46, p b 0.001; experiential/hedonic: β = 0.38, p b 0.001; quality: β = 0.39, p b 0.001). However, cognitive attribute had no significant on customer value sub-constructs, emphasizing the importance of the emotional attribute of customer–brand interaction. Among the sub-constructs of customer value, experiential/hedonic value had a significant influence on brand loyalty (β = 0.51, p b 0.001) while aesthetic/self-expressive (β = 0.18, p b 0.001) and quality (β = 0.52, p b 0.001) had significant influences on brand recognition. In line with the result of hypotheses 5 testing, both brand loyalty (β = 0.36, p b 0.001) and brand recognition (β = 0.53, p b 0.001) had significant influences on purchase intention. See Table 3 for the results of the additional analyses. 6. General discussion and conclusion This study was an investigation of luxury fashion brand value cocreation processes and the relationships among encounter attributes, consumer value, brand value, and purchase intentions. This examination contributes to expanding knowledge on value co-creation specific to luxury fashion brands. In-depth interviews with ten Chanel customers revealed that they interacted with the brand and other consumers through digital marketing and art marketing encounters of value cocreation. They were able to exchange cognitive information about the brand through digital marketing and to emotionally bond with the brands Fig. 2. Results of testing the hypothesized relationships. Table 3 Results of additional analyses on the sub-constructs of variables. Independent ➔ dependent Estimate S.E. C.R. Encounter attributes ➔ customer value Emotional ➔ social/conspicuous 0.43 0.048 9.035⁎⁎⁎ Emotional ➔ aesthetic/self-expressive 0.46 0.049 9.415⁎⁎⁎ Emotional ➔ experiential/hedonic 0.38 0.037 10.190⁎⁎⁎ Emotional ➔ quality 0.39 0.044 9.015⁎⁎⁎ Customer value ➔ brand value Experiential/hedonic ➔ brand loyalty 0.51 0.103 14.606⁎⁎⁎ Aesthetic/self-expressive ➔ brand recognition 0.18 0.038 4.691⁎⁎⁎ Quality ➔ brand recognition 0.52 0.048 10.800⁎⁎⁎ Brand value ➔ purchase intention Brand loyalty ➔ purchase intention 0.36 0.025 14.382⁎⁎⁎ Brand recognition ➔ purchase intention 0.53 0.038 13.822⁎⁎⁎ ⁎⁎⁎ p b 0.001. E. Choi et al. / Journal of Business Research 69 (2016) 5827–5832 5831

through art marketing. Study 1, a qualitative study, aligns with previous research on luxury brand value co-creation to show that art and digital technology-based brand activities are platforms for luxury customers to interact with brands and create value (Tynan et al., 2010; Hwang, Ko, & Megehee, 2014). The findings confirm that cognitive and emotional attributes of value co-creation encounters are key attributes of customer– brand interaction (Gentile et al., 2007; Payne, Kaj & Pennie, 2009). Study 2, a quantitative examination, tested the proposed luxury value co-creation model and showed that value co-creation encounter attributes influence consumer value. Although cognitive and emotional encounter attributes did not directly influence brand value, customer experience at encounters conveyed consumer value which then transferred to brand value, perhaps because brand loyalty and brand recognition are not direct outcomes of customer–brand interactions. Instead, brand loyalty and recognition develop over time as consumers experience numerous encounters. Value co-creation encounters impacted brand value and consequently, purchase intentions. Thus, luxury fashion brands should provide a variety of customer interactions for building brand values. To assure that emotional attributes transfer to consumer values, artistic elements should be embedded into digital communication or VIP events by collaborating with artists to develop products, supporting artists, and planning exhibitions. Additional analyses on the sub-constructs of variables showed that hedonic/experiential value influences brand loyalty. The result emphasizes the importance of experience-based interactions. Findings also show that brands are easily recognized when aesthetic and quality values are emphasized in brand–customer interactions. Of the two encounter attributes, only emotional attributes significantly influenced customer values. The study participants know the Chanel brand very well, so providing brand-related information may have been less important for satisfying their emotional needs during the interactions. The result aligns with previous research (Panigyrakis & Zarkada, 2014; Payne et al., 2008) showing the importance of creating experience-based interactions for successful value co-creation. The study has theoretical and practical implications for luxury fashion brands by providing empirical support for the positive effect of digital marketing and art marketing activities on value creation and purchase intentions. First, the study takes a mixed-method approach to clarify the concept of value co-creation, encounter attributes, consumer value, and brand value. The results show that brand-related digital marketing and art marketing encounters contribute to value co-creation, but influences on brand value depend on the types of customer value. The findings also suggest that brands should strategically develop encounters to maximize positive experiences by emphasizing emotional elements. This study has some limitations. As an exploration of the luxury fashion brand value co-creation process, the focus was on digital marketing and art marketing as value co-creation encounters. In reality, brand– customer interactions take place in a variety of ways. Consumer responses were collected from digital marketing and art marketing encounters to analyze hypothesized relationships. However, different encounters could have different attributes and varying influences. Future study could re-examine the relationships among the constructs in the value co-creation model in different contexts. Comparing the results would have meaningful implications for luxury fashion brand marketers. Also, because consumers worldwide are increasingly adopting social media, customer–customer interactions in brands' online communication venues are increasing. Future study should examine the potential influence of customer–customer interaction as another type of value cocreation. References Addis, M., & Holbrook, M. B. (2001). On the conceptual link between mass customization and experiential consumption: An explosion of subjectivity. Journal of Consumer Behaviour, 1(1), 50–66. Anderson, J. (1995). Cognitive psychology and its implication. New York W.H: Freeman. Assael, H. (2005). Consumer behavior a strategic approach. Dreamtech Press. Choi, H., Ko, E., Kim, E. Y., & Mattila, P. (2015). The role of fashion brand authenticity in product management: A holistic marketing approach. Journal of Product Innovation Management, 32(2), 233–242. Dhaoui, C. (2014). An empirical study of luxury brand marketing effectiveness and its impact on consumer engagement on Facebook. Journal of Global Fashion Marketing, 5(3), 209–222. Gentile, C., Spiller, N., & Noci, G. (2007). How to sustain the customer experience: An overview of experience components that co-create value with the customer. European Management Journal, 25(5), 395–410. Gilbreath, B. (2010). The next evolution of marketing: Connect with your customers by marketing with meaning. New York: McGraw-Hill. Heine, K., & Berghaus, B. (2014). Luxury goes digital: How to tackle the digital luxury brand– consumer touchpoints. Journal of Global Fashion Marketing, 5(3), 223–234. Hwang, Y., Ko, E., & Megehee, C. M. (2014). When higher prices increase sales: How chronic and manipulated desires for conspicuousness and rarity moderate price's impact on choice of luxury brands. Journal of Business Research, 67(9), 1912–1920. Kim, A. J., & Ko, E. (2010a). Impacts of luxury fashion brand's social media marketing on customer relationship and purchase intention. Journal of Global Fashion Marketing, 1(3), 164–171. Kim, A. J., & Ko, E. (2010b). The impact of design characteristics on brand attitude and purchase intention: Focus on luxury fashion brands. Korean Society of Clothing and Textiles, 34(2), 252–265. Kim, A. J., & Ko, E. (2012). Do social media marketing activities enhance customer equity? An empirical study of luxury fashion brand. Journal of Business Research, 65(10), 1480–1486. Kim, K., Ko, E., & Lee, Y. I. (2012a). Art infusion in fashion product: The influence of visual art on product evaluation and purchase intention of consumers. Journal of Global Fashion Marketing, 3(4), 180–186. Kim, K. H., Ko, E., Xu, B., & Han, Y. (2012b). Increasing customer equity of luxury fashion brands through nurturing consumer attitude. Journal of Business Research, 65(10), 1495–1499. Ko, E., & Megehee, C. (2012). Fashion marketing of luxury brands: Recent research issues and contributions. Journal of Business Research, 65, 1395–1398. Koo, H. G., & Rha, J. (2012). The conceptualization and the practical application of consumer business co-creation of value. Korean Society of Consumer Studies, 23(1), 193–227. Lee, J., Ko, E., & Megehee, C. M. (2015). Social benefits of brand logos in presentation of self in cross and same gender influence contexts. Journal of Business Research, 68(6), 1341–1349. MillwardBrown (2012). Top 100 Most Valuable Global Brands. (Retrieved from) http:// www.millwardbrown.com/mb-global/brand-strategy/brand-equity/brandz/top-globalbrands/2012 Mun, D. (2011). Value based marketing strategy in the social media age. Journal of Korean Cultural Industries, 11(2), 161–178. Panigyrakis, G., & Zarkada, A. (2014). A philosophical investigation of the transition from integrated marketing communications to metamodern meaning co-creation. Journal of Global Scholars of Marketing Science, 24(3), 262–278. Park, A. (2009). Effects of the characteristics of luxury brands on the consumer-brand relationship: Focused on fashion luxury goods. Unpublished master's thesis, Yonsei University, Seoul, Korea. Park, J., Song, H., & Ko, E. (2011). The effect of the lifestyles of social networking service users on luxury brand loyalty. Journal of Global Scholars of Marketing Science, 21(4), 182–192. Patton, M. Q. (1990). Qualitative evaluation methods. Sage publications. Payne, A. F., Kaj, S., & Pennie, F. (2009a). Co-creating brands: Diagnosing and designing the relationship experience. Journal of Business Research, 62(3), 379–389. Payne, A. F., Storbacka, K., & Frow, P. (2008). Managing the co-creation of value. Journal of the Academy of Marketing Science, 36(1), 83–96. Payne, A., Storbacka, K., Frow, P., & Knox, S. (2009b). Co-creating brands: Diagnosing and designing the relationship experience. Journal of Business Research, 62(3), 379–389. Prahalad, C. K., & Ramaswamy, V. (2004). Co-creation experience: The next practices in value creation. Journal of Interactive Marketing, 18(3), 5–14. Schechter, L. (1984). A normative conception of value. Progressive Grocer, 2, 12–14. Schmitt, B. H., & Simmonson, A. (1997). Marketing aesthetics: The strategic management of brands, identity, and image. Free Press. Think Tank (2010). Gen Y Prestige Brand Ranking. (Retrieved from) http://issuu.com/l2- thinktank/docs/gen_y_report_may2010 Tynan, C., Mckechnie, S., & Chhuon, C. (2010). Co-creating value for luxury brands. Journal of Business Research, 63(11), 1156–1163. Vargo, S. L., & Robert, F. L. (2004). Evolving to new dominant logic for marketing. Journal of Marketing, 68(January), 1–17. Vigneron, F., & Johnson, L. W. (1999). A review and a conceptual framework of prestige-seeking consumer behavior. Academy of Marketing Science Review, 1999(1), 484–506. Woodside, A. G., & Ko, E. (2013). Luxury fashion theory, culture, and brand marketing strategy: UK, Emerald. Yang, J. (2003). Relationships between benefit seeking, brand loyalty, and fashion leadership of fashion luxury goods. Unpublished master's thesis, Konkuk University, Seoul, Korea. Yoo, B., & Donthu, N. (2001). Developing and validating multidimensional consumerbased brand equity scale. Journal of Business Research, 52(1), 1–14. 5832 E. Choi et al. / Journal of Business Research 69 (2016) 5827–5832

Contents lists available at ScienceDirect Journal of Business Research journal homepage: www.elsevier.com/locate/jbusres Brand loyalties in designer luxury and fast fashion co-branding alliances☆ Bin Shena , Tsan-Ming Choib,⁎ , Pui-Sze Chowc a Glorious Sun School of Business and Management, Donghua University, 200051 Shanghai, China b Business Division, Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong c School of Management, Centennial College, 3 Wah Lam Path, Pokfulam, Hong Kong ARTICLE INFO Keywords: Luxury fashion Brand loyalty Co-branding Brand alliance management ABSTRACT Fast fashion brands, such as H & M, have co-branding projects with designer luxury brands. However, how the brand loyalties of the associated brands theoretically affect the co-branding's performance is largely unknown. Motivated by the observed industrial practices, we build a formal analytical model to examine the impacts of brand loyalty on revenues in luxury and fast fashion co-branding. The commonly adopted schemes in industry such as the profit sharing scheme, fixed-royalty scheme and mergers scheme are examined to investigate the brand performance. It is analytically found that the associated brands would perform best under the mergers scheme. This implies that the internal cooperation within a big group is the most desirable strategy for cobranding. Moreover, we provide the analytical evidence that fast fashion brands should work with well-known luxury fashion brands for brand alliance. 1. Introduction Co-branding is a brand alliance strategy in which two or more brands collaborate and launch their co-brand (Blackett & Board, 1999). In markets across the globe, we are now witnessing a growing interest in establishing strategic partnership in co-branding (Bengtsson & Servais, 2005). For example, the fast fashion co-branding strategy, which is formed between a designer-label fashion brand and a fast fashion brand, such as H & M, is quite popular. H & M first started its co-branded collection with the designer luxury fashion brand Chanel's chief designer Karl Lagerfeld in 2004. It appears that H & M has shown a unique formula of selecting an appropriate luxury fashion partner and co-branding cooperation is beneficial to both involved brands (Labbrand, 2011). The history of H & M collaboration is shown in Table 1. As we can see from Table 1, most of the participating designer luxury fashion brands are famous brands. This observation motivates our study. As we will prove theoretically later on, this is in fact a wise and “optimal” mode of co-branding alliance. The purpose of launching co-brands is to respond to the fast changing marketplace and increasing customers' brand loyalty towards the involved brands (Voss & Mohan, 2016). The benefits of co-branding partnership in terms of brand loyalty in the designer luxury and fast fashion co-branding have been widely documented in both academic literature (Ahn, Kim, & Forney, 2010; Jang, 2006; Oeppen & Jamal, 2014; Shen, Jung, Chow, & Wong, 2014) and industrial reports (as shown in Table 2). In co-branding, companies can work with other companies to integrate resources and leverage individual core competencies, or they can use current resources within one company to promote multiple products. There are many different forms of co-branding adopted in the fashion industry. They include ingredient co-branding, joint venture co-branding and same-company co-branding. Ingredient co-branding is formed when the ingredient brand is contained within the manufacturer brand. The well-known example of ingredient cobranding in fashion is the use of YKK zippers. Joint venture co-branding is established when two or more brands/companies enter a partnership to launch their co-brand. Joint venture co-branding includes many cases such as the fashion co-branding we mentioned above. In this paper, we call the co-branding between two brands of the same company/enterprise the “intra-company co-branding”. This is relatively rarely seen in the fashion industry. However, it is do-able as the big fast fashion corporations, such as Inditex, own many fashion brands of different kinds and even tiers. For example, the Uniqlo Group owns the brands Theory (a designer luxury fashion brand) and Uniqlo (fast fashion). Thus, a co-branding alliance can be formed between Uniqlo and Theory if the Uniqlo Group wishes. A natural question hence arises: Is intra-company co-branding a wise co-branding strategy? This http://dx.doi.org/10.1016/j.jbusres.2017.06.017 Received 29 June 2016; Received in revised form 26 May 2017; Accepted 19 June 2017 ☆ This paper was presented in The Mystique of Luxury Brands Conference in Shanghai, 12th–13th May, 2016. The authors sincerely thank the three anonymous referees who carefully reviewed an initial and short version of this paper, submitted to The Mystique of Luxury Brands Conference. The authors would also like to thank the editors and reviewers who provided useful comments and insights which improve the quality of the paper. This paper is supported by National Natural Science Foundation of China (71401029) and The Hong Kong Polytechnic University (A/C: G-UA1Q). ⁎ Corresponding author. E-mail address: [email protected] (T.-M. Choi). Journal of Business Research 81 (2017) 173–180 Available online 05 July 2017 0148-2963/ © 2017 Elsevier Inc. All rights reserved. MARK

question will be answered later on in this paper. According to the observed market practices, each specific cobranding project of the designer luxury brand and the fast fashion brand is usually a one-shot project, i.e., for the limited edition products. However, from a fast fashion brand's perspective (e.g., H & M), it does launch co-branding projects fr uently and how it selects the designer luxury label partner is an interesting question and important issue. As the brand loyalty is a critical element in the co-branding projects (Kim, Lee, & Lee, 2007), in this paper, we examine the impacts brought by “brand loyalties” on the revenues of the participating brands in cobranding. Our approach of incorporating brand loyalty into the market demand function is analogous to previous works (Raju, Srinivasan, & Lal, 1990; Villas-Boas, 2004; Wernerfelt, 1991). We study the brand alliance performance by using the commonly adopted cooperation schemes such as profit sharing, fixed royalties, and mergers in the co-branding partnership. With the neat analytical results, this model allows marketers to gain a better understanding of the effects of cobranding on business performance. The rest of the paper is organized as follows. Section 2 shows a concise literature review. Section 3 discusses the basic model. Section 4 explores the strategic alliance schemes in co-branding. Section 5 concludes with a discussion of the remarkable insights and managerial implications. To improve presentation technical proofs are presented in the appendix. 2. Literature review In the following review, we first examine some related co-branding studies. The spillover effect in co-branding has been greatly examined by empirical approaches. For example, Simonin and Ruth (1998) indicate that consumer attitudes towards co-branding can positively influence the subsequent consumers' attitudes towards the individual brands that comprise the alliance. Blackett and Board (1999) find that co-branding is a way to mutually enhance the associated brands. Desai and Keller (2002) compare different ingredient branding strategies. The authors show that co-branded ingredients facilitate the initial acceptance of expansions. Washbrun et al. (2004) establish a direct link between brand equity and co-branded products, and show that the high brand equity of the partner brands improves the perceived brand equity of the co-branded product and thereby generates positive spill-over effects. However, if the partnership is not appropriate in co-branding, it will have a negative effect on the associated brands. Helmig, Huber, and Leeflang (2008) argue that the co-branding strategy might have negative effects if either the combination of the two brands does not fit or the negative value perceptions about one brand hurts the partner brand. Observe that many studies in the empirical literature have explored the positive or negative spillover effect on co-branding. Dall'Olmo Riley, Pina, and Bravo (2013) discuss the value of cobranding on the position of luxury brand by conducting experiments. Mazodier and Merunka (2014) examine the impact of self-congruity and need for uniqueness on symbolic co-branding purchases. They use the co-brands of mobile phones and fashion brands as experiments. They identify that managers should pay more attention to self-congruity, rather than the attitude towards the secondary brand, when selecting a strategic partner in co-branding alliance. Voss and Mohan (2016) empirically verify the role of corporate brand as a parent of its product brands in brand alliance by experiments. They find that the corporate brand is more diagnostic for customer evaluation of a cobranded product if the brand portfolio is more consistent. The work of Voss and Mohan (2016) provides important insights on the relationship between the parent brand and the participated brands in co-branding. In this paper, we examine the co-branding between fast fashion and luxury fashion. As a matter of fact, fast fashion brands such as H & M are more likely to be a parent of the co-brand while luxury fashion brands Table 1 Collaborated designers/designer luxury labels with H & M. Designer Collaborative brand Mother company (merging year) Co-branding year Co-brand name Karl Lagerfeld Karl Lagerfeld APAX partners (2006) 2004 Karl Lagerfeld × H & M Stella McCartney Stella McCartney Stella McCartney (Kering 2001) 2005 Stella McCartney × H & M Viktor & Rolf Viktor & Rolf DIESEL (2008) 2006 Viktor & Rolf × H & M Roberto Cavalli Roberto Cavalli Clessidra S.P.A. (2015) 2007 Roberto Cavalli × H & M Rei Kawakubo Comme des Garcons Comme des Garcons 2008 Comme des Garcons × H & M Matthew Williamson Matthew Williamson Matthew Williamson 2009 Matthew Williamson for H & M Jimmy Choo Jimmy Choo Labelux (2013) 2009 Jimmy Choo × H & M Sonia Rykiel Sonia Rykiel Li & Fung Group (2012) 2009 Sonia Rykiel × H & M Alber Elbaz (2001–2015) Lanvin Lanvin 2010 Lanvin for H & M Donatella Versace (1997–now) Versace Gianni Versace S.p.A 2011 Versace for H & M Consuelo Castiglioni Marni OTB S.P.A (2015) 2012 Marni × H & M Maison Martin Margiela Maison Martin Margiela OTB S.P.A (2002) 2012 Maison Martin Margiela × H & M Isabel Marant Isabel Marant Isabel Marant 2013 Isabel Marant POUR × H & M Alexander Wang Alexander Wang Alexander Wang 2014 Alexander Wang × H & M Pierre Balmain BALMAIN BALMAIN 2015 BALMAIN × H & M Table 2 Public news on H & M's co-branding. Author Evidence Rivkin (2009) “As its aesthetic evolved he found himself ripe for a mass-market collaboration. H & M is a line in the sand for me in that it is an opportunity to collate and gather everything that everyone thinks of when they think of Matthew Williamson.” Labbrand (2011) “Co-branding helped consumers to become more knowledgeable about and familiar with luxury brands.” Hutzler (2011) “Luxury consumers won't be confused by the brand or think less of a limited-time lower priced line extension being offered by H & M.” Kong (2013) “H & M have been trying their best to provide the greatest joy of high-end fashion for their loyal consumers by collaborating with famous designers' brand.” Paul (2014) “When the average customer of today could become the luxury consumer of tomorrow, that's just no way to build brand loyalty.” Edmonds (2014) “Alexander Wang is my absolute favourite designer. It got really cold about 3 am. But I was brave and didn't get any warm drinks or anything as I didn't want to lose my spot.” Carreon (2014) “This will be a great way for a wider audience to experience elements of the Alexander Wang brand and lifestyle.” Machube (2015) “This makes us love H & M even, more and I suspect this was all part of their plan to yet again increase our brand loyalty.” Tronquet (2015) “For designers, they can reach a new audience and seize an opportunity to build loyalty with shoppers at an early age.” Teather (2015) “The collection has caused an online frenzy due to the popularity of some of Balmain's biggest fans.” B. Shen et al. Journal of Business Research 81 (2017) 173–180 174

are the participated one (Labbrand, 2011). The market size of fast fashion is larger than luxury fashion as the target consumers of former brand is mass consumers whereas the latter one is the rich group of consumers. Newmeyer et al. (2014) classify the structure of co-branding strategies into three dimensions: co-branding integration, co-branding exclusivity, and co-branding duration. They find that the above three dimensions significantly affect the value of focal brands in partner selection. In this paper, we examine the designer luxury and fast fashion co-branding, which is characterized by having a high integration level, a high exclusivity level, and short duration. Recently, Oeppen and Jamal (2014) examine the co-branding strategies in the fashion industry. They argue that the collaboration with limited availability in the fast fashion co-branding protects the brand from dilution or cannibalization of sales for the partner brand, and generates consumers' interest in a new market through the mass-market retailer. Observe that the above reviewed studies are all empirical in nature. The use of an analytical modeling approach to study co-branding is relatively rare. One example is the work by Geylani, Inman, and Ter Hofstede (2008) which reports an analytical modeling study on how co-branding strategies affect the brand image reinforcement and reveal that co-branding reinforces the partners' images. Brand performance is largely determined by brand loyalty (from consumers) which is a fundamental concept in strategic marketing (Mazodier & Merunka, 2012; Colicev, O'Connor, & Vinzi, 2016). Much of the related literature has defined brand loyalty. Aaker (1991) defines brand loyalty as the attachment that a customer has. Later, the definition is updated by Oliver (1999) who states that brand loyalty is a deeply held commitment to repurchase product or service consistently in the future by consumers. In this paper, brand loyalty refers to consumers' repeated purchasing behaviors and consumer satisfaction (Kim et al., 2007). As a behavior, customer loyalty has been measured as the long-term choice probability for a brand (Colombo & Morrison, 1989; Wernerfelt, 1991). Kim et al. (2007) indicate that brand loyalty has emerged as a significant marketing concept for many consumer driven businesses and consumers with a high level of loyalty would spend more money on the products or services. Thus the level of loyalty is closely related to the consumers' purchase behaviors. In an early book from Jacoby and Chestnut (1978), the authors develop models to measure the degree of consumer's brand loyalty. Danaher, Wilson, and Davis (2003) consider the stochastic loyalty by using the online and offline sales data for over 100 brands in 19 grocery product categories. They yield an interesting finding that consumers have a high (low) degree of brand loyalty for the brand with a high (low) market share online. In this paper, we consider the situation in which the degree of brand loyalty can be measured and optimized. He et al. (2012) examine brand loyalty from a social identity perspective by collecting the survey data via mall intercept interviews. The authors integrate brand identity and identification with value, trust and satisfaction in predicting brand loyalty. Moreover, brand loyalty has been examined by analytical models. Raju et al. (1990) analytically investigate the impact of brand loyalty and find that brands with larger brand loyalty promote less often. Agrawal (1996) examines the impact of advertising and price promotions on brand loyalty by a game theoretic analysis. Notice that one important finding in the literature on brand loyalty is: The brand loyal consumers are willing to pay higher prices and are less pricesensitive (Villas-Boas, 2004). Based on the above literature, we build a formal analytical model to conduct analysis on fashion co-branding, between a designer luxury fashion brand and a fast fashion brand, with the consideration of brand loyalty. Notice that there is still much controversy over how brand loyalty can be enhanced as it may be compensated and potentially neutralized by its rival (Shugan, 2005). In this paper, we consider the degree of brand loyalty significantly affects market demand and our focal point is to explore how the degree of brand loyalty for the participated brands in fashion co-branding influence firms' performance. 3. Basic model In this section, we develop an analytical model to examine the strategy of designer luxury and fast fashion co-branding. We denote the fast fashion brand as “brand A” (e.g. H & M), and the designer luxury fashion brand as “brand B” (e.g., Jimmy Choo, Lavin, Alexander Wang), their co-brand as “brand C” (e.g., Jimmy Choo × H & M, Lavin for H & M; Alexander Wang × H & M). We denote li as the level of brand loyalty of customers towards brand i, where i∈[A,B]. According to the extant literature, brand loyalty refers to the consumers' repeated purchase behavior, i.e. a higher degree of brand loyalty implies the consumers purchase the products more repeatedly (Kim et al., 2007; Oliver, 1999). Thus, we consider the situation when the brand loyalties of both collaborated brands may affect their co-brand, but we do not consider the brand loyalty of brand C as this co-brand project is usually a one shot project which vanishes after the co-branding products are sold out during a short duration of time. We have the following assumption to construct our model. Assumption 1. Co-branding has a spillover effect, and the consumer purchase is influenced by the brand loyalty levels of both cooperated brands. Under Assumption 1, we consider the situation when the co-brand C's demand is related to the brand loyalty of its associated brands A and B. We consider the demand function of co-brand C as follows. DC=a+b(lA+μlB), where a represents the basic market demand,1 b is the coefficient of the difference between lA and lB, where b > 0, and μ is the coefficient of lB with respect to lA, where μ≠0. We consider: (i) b is positive because many industrial reports show that the final consumers are the actual consumers of fast fashion and may purchase fast fashion products repeatedly (Kong, 2013; Machube, 2015; Paul, 2014); (ii) μ is the frequency level of purchasing designer luxury fashion brands. A positive and larger μ means the brand loyalty of luxury fashion positively affects the market demand of co-brand C and the group of luxury fashion consumers would purchase the luxury fashion brands more frequently, whereas a negative and smaller μ means the brand loyalty of luxury fashion negatively influences the market demand of co-brand C and the group of luxury fashion consumers purchases the luxury fashion brands less frequently. This assumption is consistent with the conspicuous luxury consumption pattern under which there are bandwagon and snob consumption of luxury products (Kastanakis & Balabanis, 2014; Shen, Qian, Chen, & Jochen, 2015; Shen, Qian, & Choi, 2017). We denote p as the retail price and c as the production cost. Thus, the profit function for co-brand C is πC(lA,lB)=DC(p−c)=(a+b(lA+μlB)) (p−c). In this paper, we consider the case when the retail price and the production cost are exogenously given. The brand i's profit is defined as fi(li). After launching the co-brand C, the profits of the associated brands, brands A and B, might be changed compared to the case without co-branding. This profit change is denoted by ϖi(li). As a notation, we denote π l i i ( ) as the profit for brand i excluding the profit of co-brand C, namely, πl fl ϖl i i () () () = + i i ii . Moreover, we denote πi(li) as the profit for brand i including the profit of co-brand C, namely, πi(li) =fi(li)+ϖi(li)+πC(lA,lB). To avoid trivial cases, we have: π l i i () 0 ≥ . To have analytically tractable results, we consider that π l i i ( ) is increasingly concave in li. Notice that this assumption is mild and it might be true that when the cost of brand loyalty enhancement has an exponential or a quadratic growth in the degree of brand loyalty. In this case, when the degree of brand loyalty is relatively small, enhancing brand loyalty can increase the firm's profit, whereas when the degree of brand loyalty is 1 We sincerely thank the reviewer for mentioning the impact of brand awareness. Here the basic market demand a can somehow reflect consumers' brand awareness. A high brand awareness towards the co-brand will have a high value of a. B. Shen et al. Journal of Business Research 81 (2017) 173–180 175

sufficiently large, further enhancing brand loyalty hurts firm's net profit as the cost of brand loyalty improvement is too high; in this case, the brand will stop enhancing brand loyalty. Thus, we consider the case in which enhancing brand loyalty can increase the firm's profit, i.e. π l i i ( ) is strictly increasing in li. There are several scenarios for launching co-brand C on the associated brand i: Scenario 1)ϖi(li) > 0, andπl fl ϖl () () () 0 =+ > ; i i i i ii Scenario 2)ϖi(li)= 0, and πl fl ϖl () () () 0 =+ > ; i i i i ii Scenario 3)ϖi(li) < 0, and πl fl ϖl () () () 0 =+ > . i i i i ii Notice that the scenarios “ϖi(li) > 0 and fi(li)+ϖi(li)<0” and “ϖi(li)= 0 and fi(li)+ϖi(li)<0” will not happen because fi(li) is sufficiently large.2 Here, ϖi(li) is defined as the spillover effect of launching co-brand C for the corresponding brand. Thus, ϖi(li) > 0 implies that launching co-brand C is beneficial to the corresponding brand, ϖi(li) < 0 refers to the scenario under which launching co-brand C hurts the corresponding brand, and ϖi(li)= 0 means that there is no cobrand partnership or launching co-brand C has no effect on the participated brands. Without loss of generality, we consider ϖi_n(li_n) =δi_nli_n, where n∈(1, 2, 3), δi_1 > 0, δi_2= 0 and δi_3 < 0 (i.e.δi_1 > δi_2 > δi_3). We define that a larger ϖi(li) implies the spillover effect of launching co-brand C for the corresponding brand is larger. Lemma 1. When there is no partnership between brands A and B, namely, ϖi(li)= 0, the optimal brand loyalty of brand i is = ′ = ∗ l fl arg { ( ) 0} i l i i i . Lemma 1 gives a simple expression for finding the unique optimal brand loyalty for each parent brand in the absence of the co-branding alliance. 4. Strategic alliance schemes in fashion co-branding In this section, we investigate the effectiveness of strategic alliance schemes (profit sharing scheme, fixed royalties scheme,3 and mergers scheme) between the designer luxury fashion and the fast fashion brands. We propose the mergers scheme in the partnership of cobranding as it is widely observed in the real world and also being examined in the literature. 4.1. Profit sharing scheme We first examine the profit sharing (PS) scheme between brand A and brand B, under which brand A obtains (1−λ) and brand B takes the remainingλ of the co-brand profit, where 0 < λ < 1. The profit maximization problems of brands A and B are shown below. max ( ) ( ) (1 ) ( , π l π l λπ l l = +− ) l An An An An C An Bn A n (1) and max ( ) ( ) ( , ) π l π l λπ l l = + l Bn Bn Bn Bn C An Bn A n (2) s.t. li_n > 0 and 0 < λ < 1. We now derive the optimal brand loyalty for Eqs. (1) and (2), respectively. We denote the optimal solution by li_n ∗ , where i∈(A,B). We have Proposition 1. Proposition 1. (i) The optimal brand loyalties for brands A and B exist and are unique under all the scenarios; (ii) lA_PS_1 ∗ > lA_PS_2 ∗ > lA_PS_3 ∗ and lB_PS_1 ∗ > lB_PS_2 ∗ > lB_PS_3 ∗ . Proposition 1 (i) indicates that the optimal brand loyalty for both fast fashion and designer fashion brands can be found. Proposition 1(ii) implies that the high levels of brand loyalty towards the collaborated brands in a co-brand have a significant spillover effect, i.e. a high level of brand loyalty towards the collaborated brand will have a higher impact on the brand's profit after co-branding collaboration. These results are some important managerial findings for luxury and fast fashion co-branding. Further, we consider the situation when brand A, brand B and their co-brand C constitute an alliance system. We use the subscript SC to denote the alliance cases. The corresponding profit of the alliance is defined by πSC_n(lA_n,lB_n). We denote the optimal brand loyalty of alliance for brand i as li , SC_n ∗ , and it maximizes the alliance profit. When the brand loyalties of brands A and B achieve lA, SC_n ∗=lA_n ∗ and lB , SC_n ∗=lB_n ∗ simultaneously, both brands A and B maximize not only the respective individual brand profits but also the centralized brand alliance profit. Thus, we have the following definition: Definition 1. When the brand loyalties for brands A and B simultaneously satisfy lA , SC_n∗ =lA_n∗ and lB , SC_n∗ =lB_n∗ , the alliance coordination is said to be achieved. Since the alliance profit includes the profits from brands A and B as well as the profit from co-brand C, its function can be expressed below = +++ =+ ++ + π ll π l π l f l πl l f l ϖ l f l ϖ l πl l ( ) () () () ( , ) ( ) ( ) ( ) ( ) ( , ). SC n A n B n A n A n B n B n A n An C An Bn A n An An An B n Bn Bn Bn C An Bn (3) To optimize the alliance profit πSC_n(lA_n,lB_n), we can derive the first and second order derivatives of πSC_n(lA_n,lB_n) with respect to lA_n and lB_n, and find Proposition 2 by using the first order conditions. Proposition 2. (i) The alliance coordination cannot be achieved under the PS scheme; (ii) lA_SC_n ∗ > lA_PS_n ∗ and lB_SC_n ∗ > lB_PS_n ∗ . Proposition 2 indicates that the PS scheme cannot help coordinate the alliance. This is a surprising result as the PS scheme is known to be very useful in coordinating alliance systems. In addition, the centralized optimal brand loyalty for brand A is higher than its decentralized counterpart under Scenarios 1, 2, and 3. The value of μ affects the optimal brand loyalty for brand B (i.e. the luxury brand). Proposition 3. lB_PS_n ∗ and lB_SC_n ∗ are increasing in μ. Proposition 3 reveals the impacts of μ on the brand loyalty. Recall that μ is the coefficient of luxury fashion brand loyalty on co-branding market demand. A positive and larger μ means the level of brand loyalty for the designer luxury fashion brand positively affects the market demand of co-brand C and the group of luxury fashion consumers would purchase the luxury fashion brands more frequently, whereas a negative and smaller μ means the brand loyalty of luxury fashion consumers negatively influences the market demand of co-brand C and the group of luxury fashion consumers would purchase the luxury fashion brands infrequently. The co-brand would be more successful (i.e. μ is positive and larger) when the brand loyalty of designer luxury fashion is strong (i.e. consumers have a high frequency of repeated purchase). Notice that this spillover effect is common in co-branding (Desai & Keller, 2002; Helmig et al., 2008). It is an important result for collaboration partner selection. If the partnership is not appropriate in co-branding, a brand might have a negative impact on the market demand of its cobrand. 2 We sincerely thank the reviewer for mentioning this assumption. We avoid the case of fi(li)+ϖi(li) < 0 because it is not analytically sound. A sufficiently large fi(li)is consistent with the real practices where the successful fashion brands (e.g. H & M, Alexander Wang, Lanvin) collaborate and they actually can gain a high profit by their own without forming any co-branding partnership (Oeppen & Jamal, 2014; Shen et al., 2014). 3 One of the informants from H & M mentioned that both the profit sharing scheme and the fixed royalties scheme are commonly adopted in co-branding partnership between designer luxury fashion brands and H & M. Designers may involve in the co-branding collection with the profit sharing scheme by the title “brand B × brand A”, and with the fixed royalties scheme by the title “brand B for brand A”. B. Shen et al. Journal of Business Research 81 (2017) 173–180 176

4.2. Fixed royalties scheme After exploring the PS scheme, we now examine the fixed-royalty (FR) scheme under which the profit gained from co-brand C is not shared between brands A and B; instead, one party (e.g. brand A) is responsible for running co-brand C and collecting its revenue, and then a fixed royalty w is paid to the other party (e.g. brand B). Here, we separate this scheme into 2 cases, where Case 1 considers the situation when brand A plays the role of the leader and pays w to brand B, whereas the brand B plays the role of the leader and pays w to brand A in Case 2. The profit functions of brand A, brand B and the co-brand alliance under the fixed-royalty scheme are shown below. Case 1 (Brand A is the leader): π l π l πl l w An An An An C An Bn () () (, ) =+ − , (4) πl πl w Bn Bn Bn Bn () () , = + (5) and π l l π l π l πl l SC n A n B n A n A n B n B n C A n B n ( , ) () () ( , ) =++ . (6) Case 2 (Brand B is the leader): πl πl w An An An An () () = + , (7) π l π l πl l w Bn Bn Bn Bn C An Bn () () ( , ) =+ − , (8) and π l l π l π l πl l SC n A n B n A n A n B n B n C A n B n ( , ) () () ( , ) =++ . (9) We use the subscript L to represent the leader and subscript F to denote the follower in the alliance. The leader implies the party who is the host, i.e. who runs and operates the co-brand. The follower then receives the fixed royalty fee. By maximizing the various parties' profits, we have two propositions. Proposition 4. (i) The optimal brand loyalties for brands A and B are unique under the FR scheme; the optimal brand loyalties satisfy lA_FR_1 ∗ > lA_FR_2 ∗ > lA_FR_3 ∗ and lB_FR_1 ∗ > lB_FR_2 ∗ > lB_FR_3 ∗ . Proposition 5. The alliance coordination cannot be achieved under the FR scheme. Proposition 4 indicates the optimal brand loyalties for both luxury and fast fashion brands after launching their co-brand under the FR scheme. The explicit analytical relationships are found. One important insight is: The brand leadership in establishing the co-brand significantly affects the relationships among the optimal brand loyalties. This is a critical issue because it relates to the proper matching of the fast fashion and designer luxury brands. For different desirable situations, different leaderships should be adopted. Proposition 5 further shows that the alliance cannot be coordinated under the FR scheme. 4.3. Mergers scheme As discussed above, both the PS and the FR schemes fail to achieve alliance coordination. To develop a proper coordination measure, we now examine the mergers (MG) scheme under which brands A and B share the profit from the whole alliance with the launching of co-brand C. To be specific, brand B obtains η of the whole alliance profit and brand A gets the remaining(1−η), where 0 < η < 1. In this case, the profit functions of brands A and B are shown below. π l ηπ l l An An ( ) (1 ) ( , ) = − SC n A n B n , (10) π l ηπ l l Bn Bn () ( , ) = , SC n An Bn (11) π l l π l π l πl l SC n A n B n A n A n B n B n C A n B n ( , ) () () ( , ) =++ . (12) By maximizing various parties' profits, we have Propositions 6 and 7. Proposition 6. (i) The optimal brand loyalties for brands A and B uniquely exist under the MG scheme; (ii) lA_MG_1 ∗ > lA_MG_2 ∗ > lA_MG_3 ∗ and lB_MG_1 ∗ > lB_MG_2 ∗ > lB_MG_3 ∗ , (iii) When μ is positive, lB_MG_i ∗ is increasing in μ. Proposition 7. The alliance coordination can be achieved under the MG scheme. Proposition 6 shows the relationship of optimal brand loyalties under different scenarios for the fast fashion brand and the designer luxury fashion brand under the MG scheme. Under the MG scheme, the brand loyalty for designer luxury fashion brand is similar to that under the PS scheme. The co-brand would be successful (i.e. μ is positive and large) when the brand loyalty for designer luxury fashion is large (e.g., consumers have a high frequency of repeated purchase). Proposition 7 is interesting and important. It implies that under the MG scheme, both fast and luxury fashion brands would simultaneously build their levels of brand loyalty equal to the centralized optimal brand loyalty for profit maximization. Thus, the alliance can achieve the best outcome with respect to profit. As a result, the MG scheme is a useful measure to yield an optimal alliance. 5. Conclusion and managerial implications In this study, we have proposed a general model to examine the impacts of brand loyalty on revenue management in co-branding of designer luxury fashion and fast fashion. We have considered a cobrand alliance which is consisted of two associated brands (i.e. a designer luxury fashion brand and a fast fashion brand) and their cobrand. As we have proven analytically in this paper, such an alliance is beneficial to both parties: both fast fashion and designer fashion brands can increase customers' brand loyalties and expand their consumer base with profit maximization. We have explored the commonly adopted schemes in the industry such as the PS scheme, the FR scheme and the MG scheme and reveal whether the “brand alliance” can be coordinated. We have identified the optimal brand loyalty levels with respect to the different financial outcomes of launching a co-brand and different types of brand alliance. Based on the results we derived from our analytical models, we have yielded the following managerial implications. 5.1. A. Impacts brought by brand loyalties of participating brands. Among the three explored schemes, if the fast fashion wants to increase its profit after launching the co-brand, the brand loyalty for fast fashion should be enhanced; if the designer luxury fashion brands want to increase their profit after launching the co-brand, when μ is positive (negative), the brand loyalty for luxury fashion should be improved (reduced). Moreover, we find that if the co-brand is successful, the brand loyalty for the fast fashion brand should be high. This insight is consistent with the industrial observation that most consumers purchasing the co-brands are the customers interested in fast fashion brands and they are only the potential customers of designer luxury fashion (e.g. Carreon, 2014; Kong, 2013; Paul, 2014; Rivkin, 2009; Tronquet, 2015). 5.2. B. Brand alliance coordination. We find that the alliance can only be coordinated by the MG scheme, but not the PS and FR schemes. Brand alliance coordination guarantees the attainment of the optimal outcome among the parent brands and their co-brand. In other words, the total profit of the fast fashion brand, the designer luxury fashion brand, and the co-brand are maximized. Our results imply that both the PS and FR schemes are not able to achieve alliance coordination. In other words, many current cooperation strategies adopted in practice are not yet the most efficient (e.g. H & M when working with the designer luxury fashion brands). B. Shen et al. Journal of Business Research 81 (2017) 173–180 177

5.3. C. Intra-company co-branding. Our analytical results show that the brand alliance can only be coordinated by the MG scheme. In other words, adopting the MG scheme is the most efficient partnership strategy for launching co-brands among the three strategies. The MG scheme can be adopted for the brands which belong to the same company. This shows that internal cooperation within the group is the most efficient scheme for co-branding. This also means, e.g., the big fashion groups, such as Uniqlo Corporation and Inditex Group (which have multiple fashion brands) can organize some fashion brands within themselves to launch some co-brands and apply the MG scheme for overall profit maximization. Moreover, this finding further suggests that fast fashion brands like H & M may consider merging with some designer luxury fashion brands for co-branding and adopt the MG scheme. 5.4. D. Co-branding partner selection for fast fashion brand (e.g. H & M). From our analytical result under the PS, FR, and MG schemes, the co-brand would be more successful when μ is positive and large, i.e., when the brand loyalty for the designer luxury fashion brand is strong (i.e. consumers have a high frequency of repeated purchase). This result implies that fast fashion brands such as H & M should cooperate with the designer luxury fashion brand which possesses a high degree of brand loyalty. This result is consistent with the real industrial practices according to the summary of H & M co-brands, as shown in Table 1. H & M works with the well-known fashion brands which have a large group of consumers with a strong level of brand loyalty. This insight could be interpreted from the other side: If the designer fashion brands are willing to work with the fast fashion brands for co-branding, they should work hard to attract a larger group of consumers with a strong level of brand loyalty. Therefore, selecting the right partners is important to join hands in strength, realize the win-win cooperation, and achieve both firms' business sustainability in the co-branding venture. Our study is subject to two main limitations that point towards fruitful directions for future research. First, to focus on brand loyalties in luxury and fast fashion, we simplify the analytical model and consider the case when the retail price is given. In future research, it would be interesting to examine optimal pricing issues and consider how the pricing issues of the participating brands and co-brand would affect the optimal brand loyalties and firms' performance. Second, the brand loyalty towards the participated brands in the co-brand partnership may be changed after launching the co-brand. Thus it would be important to evaluate the brand loyalty changes in the short and long terms after launching the co-brand by a longitudinal study (Dawes, Meyer-Waardenb, & Driesener, 2015; Fatma & Rahman, 2017). This may lead to a future extension of the analytical model. Last but not least, information asymmetry (Yue, Mukhopadhyay, & Zhu, 2006) is a critical issue which can also be explored in the analytical model in the future. Appendix A. all proofs and tables A.1. Proof of Lemma 1: We take the first and the second order derivatives of πi with respect to li when ϖi(li)= 0. As fi(li) is defined as the concave function of li, we can find that when ϖi(li)= 0, the optimal brand loyalty of brand i is = ′ = ∗ l fl arg { ( ) 0}. i l i i i (Q.E.D.) A.2. Proof of Proposition 1 Based on Eqs. (1) and (2), in order to obtain the optimal brand loyalties for brands A and B, we take the first and the second order derivatives of πi with respect to li. For brand A, ∂ ∂ = ′ + + ∂ − = ′ + +− − π l fl δ λπ l l dl ( ) f l δ λ p cb (1 ) ( , ) ( ) (1 )( ) , A n A n A n An An C An Bn A n A n An An (A1) and ∂ πl f l A n ∂ = A n ′′ () 0 < . A n A n 2 2 (A2) For brand B, ∂ ∂ = ′ + + ∂ − = ′ ++ − π l fl δ λπ l l dl ( ) f l δ λμ p c b (1 ) ( , ) () ( ) , B n B n B n Bn Bn C An Bn B n B n Bn Bn (A3) and ∂ πl f l B n ∂ = B n ′′ ( ) 0. < B n B n 2 2 (A4) As the second order derivatives of πi with respect to li are all negative, we can obtain the result of Proposition 1(i) that the optimal brand loyalties exist and are unique. Then, we can compare the optimal brand loyalties. For Proposition 1(ii), we let the Eqs. (A1) and (A3) be zero and obtain the optimal brand loyalty level. As we assumeδi_1 > δi_2 > δi_3 > 0, we have lA_PS_1 ∗ > lA_PS_2 ∗ > lA_PS_3 ∗ and lB_PS_1 ∗ > lB_PS_2 ∗ > lB_PS_3 ∗ . (Q.E.D.) A.3. Proof of Proposition 2 Based on Eq. (3), in order to obtain the optimal brand loyalties for alliance with respect to brand A or B, we take the first and the second order derivatives of πSC with respect to li, respectively. Then we can find. = ′ =− − − ∗ l f l δ p cb A SC n arg { ( ) ( ) } l A n An An A n (A5) and B. Shen et al. Journal of Business Research 81 (2017) 173–180 178

= ′ =− − − ∗ l fl δ b arg { ( ) μp c ( ) } B SC n l B n Bn Bn B n (A6) We can easily find that li_SC_n ∗≠li_PS_n ∗ . As a result, the alliance does not achieve coordination as shown in Proposition 2(i). Moreover, we compare the optimal brand loyalty of centralized with the decentralized one, we can obtain Proposition 2(ii) and (iii). (Q.E.D.) A.4. Proof of Proposition 3 According to the results of Propositions 1 and 2, we can find that lA_SC_n ∗ > lA_PS_n ∗ andlB_SC_n ∗ > lB_PS_n ∗ .(Q.E.D.) A.5. Proofs of Propositions 4 and 5 To obtain the optimal brand loyalties for the collaborated brands A and B as well as the alliance, we take the first and second order derivatives of the profit functions with respect to the corresponding brand loyalty, respectively. When brand A is the leader, ∂ ∂ = ′ + ′ + ∂ ∂ = ′ + +− π l fl ϖl πl l l () () f l δ p cb (, ) () ( ) A n A n A n A n A n A n C An Bn A n A n An An (A7) and ∂ ∂ = ′′ < π l f l() 0 A n A n A n A n 2 2 (A8) ∂ ∂ = ′ + π l fl δ ( ) B n B n B n Bn Bn (A9) and ∂ ∂ = ′′ < π l f l() 0. B n B n B n B n 2 2 (A10) When brand B is the leader, ∂ ∂ = ′ + ′ = ′ + π l fl ϖl fl δ () () () A n A n A n A n A n A n A n An An (A11) and ∂ ∂ = ′′ < π l f l() 0. A n A n A n A n 2 2 (A12) ∂ ∂ = ′ + ′ + ∂ ∂ = ′ ++ − π l fl ϖl πl l l () () f l δ μp c b (, ) () ( ) B n B n B n B n B n B n C An Bn B n A n An Bn (A13) and ∂ ∂ = ′′ < π l f l() 0. B n B n B n B n 2 2 (A14) We find that the second order derivatives with li are all negative. Thus, by letting the first order derivatives of profit functions with respect to the corresponding brand loyalty be zero (i.e., by the first order conditions in Eqs. (A7), (A9), (A11), and (A13)), we can obtain the optimal brand loyalties lA_FR_n ∗ and lB_FR_n ∗ . Then we compare the results of brand loyalties, we obtain Proposition 4. Moreover, comparing the results of optimal brand loyalties in the centralized system with the ones in the decentralized system (from Eqs. (A5) and (A6)), we can easily see that the brand loyalties of fast fashion and designer luxury fashion in the decentralized case are not equal to the ones in the centralized system simultaneously, i.e., the alliance coordination cannot be achieved. Then we obtain Proposition 5. (Q.E.D.) A.6. Proofs of Proposition 6 To obtain the optimal brand loyalties for the collaborated brands A and B, we take the first and second order derivatives of the profit functions with respect to the corresponding brand loyalty, respectively. ∂ ∂ = − ∂ ∂ + − ∂ ∂ π l l ηπ l l ηπl l l ( ) (1 ) ( ) (1 ) ( , ) . An An A n SC n A n A n C An Bn A n (A15) ∂ ∂ = − ∂ ∂ < π l l ηπ l l ( ) (1 ) ( ) 0. An An A n SC n A n A n 2 2 2 2 (A16) ∂ ∂ = ∂ ∂ + ∂ ∂ π l l ηπ l l ηπ l l l ( ) () (, ) . Bn Bn B n SC n B n B n C An Bn B n (A17) ∂ ∂ = ∂ ∂ < π l l ηπ l l ( ) ( ) 0. Bn Bn B n SC n B n B n 2 2 2 2 (A18) Let the first order derivatives of the profit functions (i.e. Eqs. (A15) and (A17)) with respect to the corresponding brand loyalty be zero, we can B. Shen et al. Journal of Business Research 81 (2017) 173–180 179

obtain the optimal brand loyalties. Then by comparing the results of brand loyalties, we can obtain Proposition 6. (Q.E.D.) A.7. Proofs of Proposition 7 Moreover, comparing the result of optimal brand loyalties in the centralized system with the ones in the decentralized system, we can find that when calculating the optimal brand loyalty in brand A and brand B, (1−η) in Eq. (A15) and η in Eq. (A17) are cancelled out. In other words, we have lA, SC_n ∗=lMG_n ∗ and lB , SC_n ∗=lB_MG_n ∗ . (Q.E.D.) References Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New York: The Free Press. Agrawal, D. (1996). Effects of brand loyalty on advertising and trade promotions: A game theoretic analysis with empirical evidence. Marketing Science, 15(1), 86–108. Ahn, S., Kim, H., & Forney, J. A. (2010). Fashion collaboration or collision? Examining the match-up effect in co-marketing alliances. Journal of Fashion Marketing and Management, 14(1), 6–20. Bengtsson, A., & Servais, P. (2005). Co-branding on industrial markets. Industrial Marketing Management, 34(7), 706–713. Blackett, T., & Board, B. (1999). Co-branding: The science of alliance. London: Macmillan Press. Carreon, B. (2014). H & M teams up with Alexander Wang for next collaboration. Forbes. April 13. Accessed at http://www.forbes.com/sites/bluecarreon/2014/04/13/hmteams-up-with-alexander-wang-for-next-collaboration/#26ac23a5f428. Colicev, A., O'Connor, P., & Vinzi, V. E. (2016). Is investing in social media really worth it? How brand actions and user actions influence brand value. Service Science, 8(2), 152–168. Colombo, R. A., & Morrison, D. G. (1989). A brand switching model with implications for marketing strategies. Marketing Science, 8(1), 89–99. Dall'Olmo Riley, F., Pina, J. M., & Bravo, R. (2013). Downscale extensions: Consumer evaluation and feedback effects. Journal of Business Research, 66(2), 196–206. Danaher, P. J., Wilson, I. W., & Davis, R. (2003). A comparison of online and offline consumer brand loyalty. Marketing Science, 22(1), 461–476. Dawes, J., Meyer-Waardenb, L., & Driesener, C. (2015). Has brand loyalty declined? A longitudinal analysis of repeat purchase behavior in the UK and the USA. Journal of Business Research, 68(2), 425–432. Desai, K. K., & Keller, K. L. (2002). The effects of ingredient branding strategies on host brand extendibility. Journal of Marketing, 66(1), 73–93. Edmonds, L. (2014). Fashion frenzy: loyal fans queue in freezing temperatures to get their hands on Alexander Wang × H & M, as Ellie Goulding and Jamie Winstone have sneak peak at the launch. Evening Standard. November 6. Accessed at http://www. standard.co.uk/fashion/news/fashion-frenzy-loyal-fashion-fans-queue-in-freezingtemperatures-to-get-their-hands-on-alexander-9843559.html. Fatma, M., & Rahman, Z. (2017). An integrated framework to understand how consumerperceived ethicality influences consumer hotel brand loyalty. Service Science, 9(2), 136–146. Geylani, T., Inman, J. J., & Ter Hofstede, F. (2008). Image reinforcement or impairment: The effects of co-branding on attribute uncertainty. Marketing Science, 27(4), 730–744. Helmig, B., Huber, J. A., & Leeflang, P. S. H. (2008). Co-branding: The state of the art. Schmalenbach Business Review, 60, 359–377. Hutzler, K. (2011). Versace, H & M collaboration: Brand erosion or harsh reality? Luxury Daily. June 22. Accessed at http://www.luxurydaily.com/versace-and-hmcollaborate-on-fall-2011-line/. Jacoby, J. W., & Chestnut, R. W. (1978). Brand loyalty measurement and management. New York: Wiley. Jang, E. (2006). An analysis on cases of fashion collaboration strategy. Journal of Fashion Business, 10(6), 110–121. Kastanakis, M. N., & Balabanis, G. (2014). Explaining variation in conspicuous luxury consumption: An individual differences' perspective. Journal of Business Research, 67(10), 2147–2154. Kim, W. G., Lee, S., & Lee, H. Y. (2007). Co-branding and brand loyalty. Journal of Quality Assurance in Hospitality & Tourism, 8(2), 1–23. Kong, J. (2013). Isabel Marant × H & M is coming!. Maxmaya. October 31. Accessed at http://www.maxmayo.com/lookbooks/isabel-marant-x-hm-is-coming/. Labbrand (2011). Luxury brands and co-branding: Versace for H & M, Labbrand. Accessed at http://www.labbrand.com/brand-source/luxury-brands-and-co-brandingversace-hm. Machube, R.. Review: H & M X BALMAIN, concrete-onine. (2015, March 11). http://www. concrete-online.co.uk/review-hm-x-balmain/ Accessed at. Mazodier, M., & Merunka, D. (2012). Achieving brand loyalty through sponsorship: The role of fit and self-congruity. Journal of the Academy of Marketing Science, 40(6), 807–820. Mazodier, M., & Merunka, D. (2014). Beyond brand attitude: Individual drivers of purchase for symbolic cobranded products. Journal of Business Research, 67, 1552–1558. Oeppen, J., & Jamal, A. (2014). Collaborating for success: Managerial perspectives on cobranding strategies in the fashion industry. Journal of Marketing Management, 30(9–10), 925–948. Oliver, R. (1999). Whence consumer loyalty? Journal of Marketing, 63, 33–44. Paul, K. J. (2014). When high meets low: Luxury brands and retail chain collaborations. Squarespace. April 22. Accessed at http://static1.squarespace.com/static/ 56509a6ce4b02e24589f6576/t/56d3812527d4bd3f75e72655/1456701748841/ when-high-meets-low-luxury-brands-and-retail-chain-collaborations-april-22-2014. pdf. Raju, J. S., Srinivasan, V., & Lal, R. (1990). The effects of brand loyalty on competitive price promotional strategies. Management Science, 36(3), 276–304. Rivkin, A. (2009). Matthew Williamson's technicolour dream. Evening Standard(May 14), Accessed at http://www.standard.co.uk/fashion/matthew-williamsons-technicolourdream-6711070.html. Shen, B., Jung, J., Chow, P. S., & Wong, S. (2014). Chapter 7: Co-branding in fast fashion: The impact of consumers' need for uniqueness on purchase perception. Fashion branding and consumer behaviors: Scientific ModelsSpringer (International Series on Consumer Science). Shen, B., Qian, R., Chen, L., & Jochen, S. (2015). Optimal pricing and online retail service for luxury fashion with social influence. 12th International Conference on service systems and service management. Guangzhou, China June 22–24 (pp. 2015–). . Shen, B., Qian, R. R., & Choi, T. M. (2017). Selling luxury fashion online with social influences: Demand changes and supply chain coordination. International Journal of Production Economics, 185, 89–99. Shugan, S. M. (2005). Brand loyalty programs: Are they shams? Marketing Science, 24(2), 185–193. Simonin, B. L., & Ruth, J. A. (1998). Is a company known by the company it keeps? Assessing the spillover effects of brand alliances on consumer brand attitudes. Journal of Marketing Research, 35(1), 30–42. Teather, C. (2015). H & M Balmain: Oxford Street store has to close after thousands of shoppers turn up for collection's launch. Evening standard. November 5. Accessed at http://www.standard.co.uk/fashion/news/hm-on-oxford-street-closes-for-healthand-safety-as-number-of-shoppers-for-balmain-collaboration-a3107796.html. Tronquet, M. (2015). Improving the retail customer experience for haute couture collaborations. Lenati. December 9. Accessed at http://www.lenati.com/blog/2015/12/ improving-the-retail-customer-experience-for-haute-couture-collaborations/. Villas-Boas, J. M. (2004). Consumer learning, brand loyalty, and competition. Marketing Science, 23(1), 134–145. Voss, K. E., & Mohan, M. (2016). Corporate brand effects in brand alliances. Journal of Business Research, 69, 4177–4184. Wernerfelt, B. (1991). Brand loyalty and market equilibrium. Marketing Science, 10, 229–245. Yue, X., Mukhopadhyay, S. K., & Zhu, X. (2006). A Bertrand model of pricing of complementary goods under information asymmetry. Journal of Business Research, 59(10−11), 1182–1192. B. Shen et al. Journal of Business Research 81 (2017) 173–180 180

Contents lists available at ScienceDirect Journal of Business Research journal homepage: www.elsevier.com/locate/jbusres “You are too friendly!” The negative effects of social media marketing on value perceptions of luxury fashion brands Minjung Parka,⁎ , Hyunjoo Imb , Hye-Young Kimb a Sigmund Weis School of Business, Susquehanna University, 514 University Ave., Selinsgrove, PA 17870, United States of America b Retail Merchandising, University of Minnesota, 240 McNeal Hall, 1985 Buford Avenue, St. Paul, MN 55108, United States of America ARTICLE INFO Keywords: Luxury brands Social media marketing Psychological distance Consumer engagement Value perceptions Brand dilution ABSTRACT In light of the growing concern about brand dilution of luxury brands on social media, the purpose of this research was to examine the impact of brand-consumer engagement on value perceptions of luxury fashion brands within the context of social media marketing. The result of Study 1 demonstrated that luxury brands were inherently psychologically distant than mainstream brands. The results of Study 2 and Study 3 showed that a luxury brand with a high level (vs. low level) of brand-consumer engagement resulted lower value perceptions (i.e., social, uniqueness, and quality value perceptions) of the brand, and such relationships were mediated by decreased psychological distance. This research provides important implications for luxury brand managers and scholars that luxury fashion brands should maintain psychological distance on social media to protect the core value perceptions of the brands. 1. Introduction Social media refers to Internet-based platforms which aim to enable user interactions such as creating and sharing information and discussing ideas (Kaplan & Haenlein, 2010). Social media has become a powerful marketing tool for brand managers because of its interactive nature. Brands engage with consumers on social media by responding to consumers' posts and encouraging user-participation. According to a report by Schneider (2015), leading brands respond to about 60% of consumers' tweets on Twitter. Luxury brands also have increasingly utilized social media to engage in two-way communications with consumers (Kim & Ko, 2012). Among luxury brands, Coach and Karen Millen encourage consumers to upload photos of their products with hashtag on social media and feature consumer photos on their websites. Also, Cartier actively responds to consumers' questions and engages in conversations with consumers on its Facebook brand page. A growing number of general brand studies have documented positive outcomes of brand-consumer engagement on social media. For example, Schivinski and Dabrowski (2016) found that user-generated social media brand communications positively influenced brand loyalty and perceived brand quality. Similarly, Labrecque (2014) found that brand-user engagement increased loyalty intentions and willingness to provide information to the brand. Despite the positive outcomes of brand-consumer engagement on social media documented in the literature, an important question still remains: Is a high level of brand-consumer engagement always beneficial to luxury brands? It is clear that the fundamental concepts of social media and luxury contradict each other: social media is inclusive, interactive, accessible, and designed for the masses, while luxury is exclusive, controlled, and intended for a selected group of wealthy consumers (Desai, 2016; Reed, 2015). Therefore, social media, characterized by interactivity and accessibility, may damage the core meaning of exclusivity inherently embedded in a luxury brand. In line with this perspective, previous researchers have raised concern about the risks of brand dilution of luxury brands on social media (BlascoArcas, Holmqvist, & Vignolles, 2016; Tungate, 2009). However, most of empirical studies have focused on the positive effects of social media on luxury brands (e.g., Chu, Kamal, & Kim, 2013; Kim & Ko, 2012), and the critical issue of how social media may backfire in the context of luxury brands has received little attention. Considering the possible long-term impact of brand dilution, it is imperative to investigate the possible negative impact of social media marketing on value perceptions of luxury brands and the underlying mechanism of the effect. This information could deepen our understanding of the factors that influence luxury brands on social media and generate strategic guidelines for luxury brand managers to protect their brand reputation while taking advantage of social media marketing. The current study builds on the contention that active brandhttps://doi.org/10.1016/j.jbusres.2018.07.026 Received 10 November 2017; Received in revised form 13 July 2018; Accepted 15 July 2018 ⁎ Corresponding author. E-mail addresses: [email protected] (M. Park), [email protected] (H. Im), [email protected] (H.-Y. Kim). Journal of Business Research 117 (2020) 529–542 Available online 24 July 2018 0148-2963/ © 2018 Elsevier Inc. All rights reserved. T

consumer engagement on social media may damage the core value perceptions (i.e., social, uniqueness, and quality value perceptions) of luxury brands because the brands feel too close and accessible to the general consumers. As mentioned earlier, luxury brands are intended to cater to only a privileged class of consumers, and they should maintain distance from the masses to stay desirable and valuable (Fuchs, Prandelli, Schreier, & Dahl, 2013; Kapferer & Bastien, 2012). Drawing upon construal level theory of psychological distance, this study aims to illustrate maintaining psychological distance to the masses on social media is essential for luxury brands to protect their important value perceptions. In the current research, psychological distance is defined as consumers' subjective perception about the distance between a luxury brand and the mass market consumers. 2. Literature review and hypotheses 2.1. Construal level theory of psychological distance Construal level theory of psychological distance (CLT) (Liberman & Trope, 2008) is a theory that explains the relationship between a person's subjective experience of distance and the person's way of thinking. CLT posits that the more distant (vs. closer) an object is from the self, the more abstractly (vs. concretely) the object is construed, leading to high-level construals (vs. low-level construals). Therefore, CLT posits that as the distance between an object and the self increases (vs. decreases), people perceive the object at high-level construals (vs. lowlevel construals). According to CLT, psychological distance can be determined by four dimensions: 1) temporal distance (i.e., the perceived distance in time between the perceiver's present time and the event); 2) spatial distance (i.e., the perceived distance in physical space between a person's location and the object); 3) social distance (i.e., the extent to which a target person or object is related to the self); and 4) hypothetical distance (i.e., the extent to which an object is perceived to be real or imaginary) (Liberman & Trope, 2008). An event is perceived as psychologically distant if it happens in the far future (vs. near future), occurs in physically remote places (vs. near places), is less (vs. more) related to the self, and is less (vs. more) likely to occur. Research has shown that the four dimensions of psychological distance are interrelated and can be integrated into a single psychological distance (Darke, Brady, Benedicktus, & Wilson, 2016; Kim, Zhang, & Li, 2008; Trope & Liberman, 2010). According to Trope and Liberman (2010), although each dimension of psychological distance is not directly related, people regard these dimensions as having a common meaning and able to access all dimensions automatically. As a result, one dimension of distance affects other dimensions of distance. For example, Darke et al. (2016) found that physical distance of a retailer influences the overall psychological distance of the retailer which subsequently affects trust and purchase intentions. 2.2. Psychological distance of luxury brands According to a widely accepted definition in consumer research, a luxury brand refers to a brand that is characterized by a set of unique factors including exclusivity, high price, quality, and symbolic attributes (Heine, 2012; Riley, Lomax, & Blunden, 2004). One of the core principles of luxury brand management is creating psychological distance between luxury brands and the mass-market (Kapferer, 1997; Kapferer & Bastien, 2012; Wiedmann, Hennigs, & Siebels, 2009). In the current research, psychological distance of a luxury brand is defined as consumers' subjective perception of the distance between a luxury brand and the mass-market consumers. Luxury brand consumption is driven by the desire to enhance one's social status and to own an exclusive product that only a small number of people can possess (Kapferer & Bastien, 2012; Wiedmann et al., 2009). Luxury brands evoke perceptions of rarity and exclusivity due to the difficulty of attaining them (Miyazaki, Grewal, & Goodstein, 2005), which enlarges the perception of psychological distance between the luxury brands and the average consumers. Luxury brands strategically limit attainability of the brand by tightly controlling many aspects of their business practices such as the price of products, distribution channels, aesthetic dimensions of products (Kapferer, 1997) to maintain the perception of exclusivity among consumers. In addition, luxury brand advertisements communicate superiority, exclusivity, and distance by invoking social segregation, and exclusion (Jiang, Gao, Huang, DeWall, & Zhou, 2014). Therefore, the following hypothesis is proposed. H1. Luxury brands, compared to casual brands, will be perceived as more psychologically distant. 2.3. Luxury brand-consumer engagement on social media Social media often becomes a place for socialization and building potential friendship though repeated conversations and exchanges among users (Tsimonis & Dimitriadis, 2014). Similarly, brands often build relationships with consumers on social media by engaging with them. In this study, brand-consumer engagement is defined as a brand's motivational state to connect and build social relationships with all consumers. For example, brands engage with consumers through responding to consumer comments, publishing user-generated contents on the brands' social media page (Peterson, 2015), and posting interactive content such as clickable icons or a quiz that consumers can take (Heavey, 2017). An engagement level between a brand and consumers on social media can range from high to low (Gilbert & Karahalios, 2009). An active, high level of brand-consumer engagement will focus more on building an intimate, close relationship between a brand and all consumers. For instance, luxury brands may actively engage with consumer by responding to all consumers' comments in a friendly way or encouraging them to engage in the brand's social media events. On the contrary, luxury brands with a low level of brand-consumer engagement may selectively respond to consumers' comments and just focus on communicating their brand messages on social media. This study argues that a level of brand-consumer engagement is an important determinant of psychological distance, specifically social distance of luxury brands. According to Akerlof's (1997) model of social distance, socially closer individuals are more likely to interact with each other while those who are distant have little interaction. Likewise, Bourdieu (1989) argues that social distance represents a symbolic space between status groups with different lifestyles, and people who are socially distant rarely interact. Therefore, an extensive engagement between a brand and consumers can lead to greater intimacy and closeness between the brand and the consumers (Hudson, Huang, Roth, & Madden, 2016) and give consumers feelings of friendship (Gummerus, Liljander, Weman, & Pihlström, 2012). In this sense, a high level of brand-consumer engagement on social media is likely to reduce psychological distance of luxury brands. Therefore, the following hypothesis is proposed. H2. A luxury brand with a high (vs. low) level of brand-consumer engagement on social media, will be perceived as less (vs. more) psychologically distant. 2.4. Psychological distance and value perceptions of luxury brands Researchers have proposed that there are multiple dimensions that constitute consumers' value perceptions of luxury brands (Hennigs, Wiedmann, Behrens, & Klarmann, 2013; Kapferer, 1997; Vigneron & Johnson, 2004). That is, consumers seek multifaceted values through consumption of luxury brands (Hennigs et al., 2013). The consensus is that there are three key dimensions that create value perceptions of M. Park et al. Journal of Business Research 117 (2020) 529–542 530

luxury brands: social, uniqueness, and quality value perceptions. These key value dimensions should be ensured to create a lasting luxury brand (Vigneron & Johnson, 2004). Social value perception refers to the perceived utility of a luxury brand for enhancing social status (Vigneron & Johnson, 2004; Wiedmann et al., 2009). This social value dimension is an outer- directed value which aims to create a favorable social image within consumers' social groups or to fit into groups consumer aspire to through the acquisition of conspicuous products (Park, Rabolt, & Jeon, 2008; Wiedmann et al., 2009). Consumers use products to integrate the symbolic meanings of the products into their identities and to communicate how they define themselves to others (Holt, 1995). Because luxury brands symbolize an affluent lifestyle (Dittmar, 1994), consumers use luxury brands as an important tool to signal wealth, high status, and a group membership to upper socio-economic class. Uniqueness perception is concerned with the perceived exclusivity and rarity of a luxury brand (Wiedmann et al., 2009). Scarcity is an important aspect of luxury brands as it helps consumers feel unique and special (Tsai, Yang, & Liu, 2013) and it increases the value and dream of the brands (Brock, 1968; Dubois & Paternault, 1995). Empirical evidence confirmed that consumers regarded a scarce luxury brand as being valuable and desirable because it could enhance their image by signifying that they are unique and different from the rest of the others (Verhallen, 1982; Verhallen & Robben, 1994; Vigneron & Johnson, 2004). Thus, when a luxury product becomes available to everyone, it would no longer be regarded as luxury because it loses exclusive value (Wiedmann et al., 2009). Lastly, quality value perception is defined as a consumer's subjective belief that products of a luxury brand are of superior quality and performance (Wiedmann et al., 2009). Luxury brands are made of the best materials and hand-finished to ensure high quality (Fionda & Moore, 2009). Consumers regard this superior quality as a fundamental aspect of a luxury brand (Quelch, 1987). In addition, because high price is often linked to high quality (Rao & Monroe, 1989), consumers expect expensive luxury brands to have a high perceived quality value (Shukla & Purani, 2012). This research proposes that one important determinant of value perceptions of luxury brands (i.e., social perception, uniqueness perception, quality perception) is psychological distance of the brands. As discussed earlier, luxury brands are built on the concept of distance, meaning not everyone can own or have access to the brands. Researchers demonstrate that the core perceptions of luxury brands can be diluted when the brands become close to undesirable groups of consumers (e.g., mass market consumers) and when overdiffused into the mass market (Bellezza & Keinan, 2014; Kapferer & Bastien, 2012). In other words, increasing the accessibility of luxury brands for less affluent mass market consumers reduces psychological distance of the brands, which subsequently damages the high-status, symbolic character of the brands. Recent studies provide supports for the negative effects of reduced psychological distance on luxury brand perceptions (e.g., De Barnier, Falcy, & Valette-Florence, 2012; Fuchs et al., 2013). For example, Fuchs et al. (2013) found that user-designed luxury products, compared to company-designed products, decreased consumer demands for the products. When products were designed by users who were average consumers not by the luxury brands' elite experts, the brand was perceived to be close to mass market consumers which undermined perceived social value. Similarly, De Barnier et al. (2012) found that accessible luxury brands, which are psychologically closer to mass-market consumers than other luxury brands, were associated with lower perceived social value. In line with this finding, this research argues that reduced psychological distance prompted by a high level of brand-consumer engagement may make an impression that the luxury brand is for every day consumers rather than a selective group of people. In turn, it will undermine the core value perceptions of luxury brands. Therefore, the following hypotheses are formulated. H3. A luxury brand with a high level of brand-consumer engagement on social media, compared to one with a low level of engagement, will lead to lower brand perceptions (i.e., social perception, uniqueness perception, quality perception). H4. The effect of brand-consumer engagement on perceptions of luxury brands is mediated by reduced psychological distance. The Fig. 1 illustrates the research model. 3. Study 1 The objective of Study 1 was to provide a preliminary test of the prediction that luxury brands, compared to mainstream brands, will be perceived as more psychologically distant (H1). 3.1. Study design The study used a 2 (brand category: luxury vs. mainstream) × 2 (brand replicates) mixed-model design in which the brand category was a between-subject factor and the brand replicates were a within-subject factor. Following previous research on categorizing apparel brands based on brand associations (Dew & Kwon, 2010; Fuchs et al., 2013), Versace and Prada were used as the luxury brand replicates and American Eagle and Old Navy were used as the mainstream brand replicates. Fig. 1. The research model. M. Park et al. Journal of Business Research 117 (2020) 529–542 531

3.2. Instruments The measurement items of psychological distance and brand awareness were adopted from previous research. The measures of psychological distance were adapted from Darke et al. (2016). They were a three-item semantic differential scale that measure various dimensions of psychological distance (i.e., when I think about brand X and its characteristics, I think it is… social close/distant, temporally close/distant, physically close/distant). In order to rule out the possible confounding effects of brand awareness on the results of the study, a three-item scale of brand awareness adopted from Aaker (1996) was included (e.g., I have heard of this brand; 1 = Strongly disagree to 7 = Strongly agree). 3.3. Data collection and experimental procedure Participants were recruited from Amazon MTurk. The study was advertised as a consumer brand perception study. Participants were told that the researcher was interested in their perception of two apparel brands. Next, participants were randomly assigned to one of the two brand category conditions (i.e., luxury vs. mainstream) in which two different brands for each category were presented in a random order. The participants were provided with the name of each brand and then asked to complete the measures of psychological distance and brand awareness. Lastly, they answered questions related to demographic information such as gender and income. 3.4. Results 3.4.1. Participant characteristics Fifty-nine MTurk workers (male = 65%, mean age = 28 years) participated in the study. The median household income of participants ranged from $30,000 to $49,999. 3.4.2. Hypothesis testing Prior to testing H1, the mean score of brand awareness for each of the four stimuli brands was compared using an Independent sample ttest. The results revealed that there were no significant mean differences among the four brands (p > .05). Therefore, the data across the brand replicates were collapsed. The results of Independent sample ttest showed that luxury brands, compared to mainstream brands, were perceived as more psychologically distant, as predicted in H1 (Mluxury = 4.66 vs. Mmainstream = 3.63; t(116) = 4.59, p = .00). Therefore, hypothesis 1 was supported. 3.5. Discussion The results of Study 1 provide supports for our initial prediction that luxury brands are inherently psychologically distant than mainstream brands (H1). This was a condition that needed to be met to continue with other hypotheses, as the research is built on the argument that luxury brands need to maintain psychological distance from the mass market consumers. Also, the results rule out the possible effect of brand awareness, as there were no significant mean differences among the four stimuli brands. 4. Study 2 The objectives of study 2 were to test the effect of brand-consumer engagement on psychological distance of luxury brands (H2) and the value perceptions of luxury brands (H3) and the mediating role of psychological distance (H4). 4.1. Study design and stimuli development A single factor (brand-consumer engagement level: High vs. Low) between-subjects design was used. To manipulate the level of brandconsumer engagement, two versions of a relatively unknown luxury watch brand (i.e., Vacheron Constantin's) mock Facebook pages, varying the degree of responsiveness to consumers' comments and consumer participations, were created. Compared to a very famous luxury watch brand such as Rolex, Vacheron Constantin has significantly less followers on social media (6 M vs. 510 K on Facebook), implying a relatively lower level of brand awareness. Using a relatively unknown luxury brand can minimize possible confounding effects resulting from previous perceptions about the brand. For the high brand-consumer engagement condition, the luxury brand responded to consumers' comments on the brand's Facebook posting in a friendly way with use of emojis. Also, the brand encouraged consumers to share their photos using a brand hashtag and displayed consumers' photos wearing the brand's products on its Facebook page. For the low brand-consumer engagement condition, the luxury brand did not respond to any consumers' comments on its Facebook posting. Also, it only displayed the images of their products and did not show any images of consumers wearing their products (see Appendix A for the stimuli). 4.2. Instruments In study 2, to better apply psychological distance to the context of relatively unknown luxury brand evaluation, perception of formality (i.e., I think this brand is… casual-formal) and unattainability (i.e., I think this brand is… attainable-unattainable) were measured for psychological distance. Previous research suggests that formality (Slepian, Ferber, Gold, & Rutchick, 2015) represents a form of psychological distance and attainability of a target object determines psychological distance of the object (Gjesme, 1981) (see Table 1). Social value perception was measured with six items that assess conspicuousness and status of the brand (e.g., To what extent can this brand indicate a person's social status? 1 = Not at all, 7 = Very much) adopted by Truong, Simmons, McColl, and Kitchen (2008). Quality value perception was measured by four items that assess perception of the brand's product quality (e.g., This brand's product has the best quality; 1 = Strongly disagree, 7 = Strongly agree) (Hennigs et al., 2013; Hung et al., 2011), and uniqueness value perception was measured by two items that measures rarity and exclusivity of the brand (e.g., This brand's product is exclusive; 1 = Strongly disagree, 7 = Strongly agree) (Hung et al., 2011; Lee, Chen, & Wang, 2015). The same measures of brand awareness (Aaker, 1996) from Study 1 were used for Study 2. The manipulation of brand-consumer engagement level was checked by one item, “The brand I just saw actively interacts with consumers on Facebook; 1 = Strongly disagree, 7 = Strongly agree”. 4.3. Data collection and experimental procedure Participants were recruited from Amazon MTurk. The study was advertised as a study about a luxury brand's social media pages. Participations were randomly assigned to one of the two brand-consumer engagement conditions. They first viewed the corresponding Facebook pages for their condition and then responded on questionnaire items measuring brand perceptions (i.e., social, uniqueness, quality value perceptions), psychological distance of the brand (i.e., formality and unattainability), brand-consumer engagement level (manipulation check item), brand awareness to control for previous knowledge and perception, and demographic information. 4.4. Results 4.4.1. Participant characteristics A total of 74 participants (male = 59.5%) were recruited from Amazon MTurk to a luxury brand evaluation study. The median annual M. Park et al. Journal of Business Research 117 (2020) 529–542 532

household income of participants was $30,000–49,999. 4.4.2. Manipulation checks The manipulation of the level of brand-consumer engagement was successful. Participants in the high-level brand-consumer engagement condition rated significantly higher on brand-consumer engagement than those who were in the low-level interaction condition (Mhigh = 5.12 vs. Mlow = 2.31; t(72) = 11.34, p < .00). 4.4.3. Hypothesis testing To test the effect of brand-consumer engagement on psychological distance of luxury brands (H2) and the value perceptions of luxury brands (H3), a one-way MANCOVA analysis was performed. Brand awareness was entered as a covariate to prevent possible confounding effects. As predicted, the results showed that the participants in the high brand-consumer engagement condition (vs. low) indicated lower psychological distance of the brand (formality: Mhigh = 4.62 vs. Mlow = 5.51; F = 3.96, p = .05, unattainability: Mhigh = 3.32 vs. Mlow = 4.30; F = 6.67, p < .05), thereby supporting H2. Also, as predicted in H3, the results revealed that the participants in the high brand-consumer engagement condition (vs. low brand-consumer engagement condition) showed lower value perceptions (i.e., social, uniqueness, and quality value perceptions) of the luxury brand (p ≤ .01) (see Fig. 2). Furthermore, a mediation analysis was conducted to examine the mediating role of psychological distance on the relationships between brand-consumer engagement on brand perceptions (H4). The procedure suggested by Zhao, Lynch, and Chen (2010) was followed using the Preacher and Hayes (2008) macro for mediation analysis. Brand awareness was entered as a covariate to control possible confounding effects. The results showed that formality (psychological distance) partially mediated the relationships between brand-consumer engagement and social and quality perceptions, but not for uniqueness perception. Specifically, when social perception was regressed on brandconsumer engagement, including formality decreased the beta weight for brand-consumer engagement from 0.92 (t(70) = 3.49, p < .01) to 0.70 (t(70) = 2.72, p < .01) (see Fig. 3). The bootstrapping technique also supported the proposed mediation relationship. When 1000 bootstrapped samples were used, 95% BCa (bias corrected and accelerated) bootstrap confidence interval did not include zero (indirect effect: β = 0.22, SE = 0.13, 95% CI = 0.02 to 0.51). Similarly, when quality perception was regressed on brand-consumer engagement, including formality decreased the beta weight for brand-consumer engagement from 0.87 (t(70) = 3.25, p < .05) to 0.64 (t(70) = 2.46, p < .05) (see Table 1 Measurement items of variables in Study 2. Variables Items Scale Psychological distance When I think about the brand and its characteristics, I think it is…. Casual – formal Attainable – unattainable 7-Point semantic differential scale Social value perception To what extent can this brand indicate a person's social status? To what extent is this brand a symbol of achievement? To what extent is this brand a symbol of wealth? To what extent is this brand a symbol of prestige? To what extent does this brand attract attention? To what extent can this brand impress other people? 1 = not at all, 7 = very much Uniqueness value perception This brand's product is rare This brand's product is exclusive. 1 = strongly disagree, 7 = strongly agree Quality value perception This brand's product has the best quality. This brand's product has rich workmanship. This brand's product lasts a long time. This brand's product is crafted. 1 = strongly disagree, 7 = strongly agree Brand awareness I know what this brand stands for. I have an opinion about this brand. I have heard of this brand. 1 = strongly disagree, 7 = strongly agree Manipulation check The brand I just saw actively interacts with consumers on Facebook. 1 = strongly disagree, 7 = strongly agree Fig. 2. The mean differences in the outcome variables (Study 2). Note. *p ≤ .05, **p ≤ .01, ***p ≤ .001. M. Park et al. Journal of Business Research 117 (2020) 529–542 533

Fig. 4). The bootstrapping technique also supported the proposed mediation relationship (indirect effect: β = 0.23, SE = 0.14, 95% CI = 0.02 to 0.57). On the other hand, another measure of psychological distance, unattainability, did not mediate any relationships. Therefore, H4 was partially supported. 4.5. Discussion Study 2 empirically demonstrated the impact of brand-consumer engagement on psychological distance and value perceptions of the luxury brand. As hypothesized, the study revealed that a high (vs. low) level of brand-consumer engagement shortened psychological distance between the luxury brand and the consumers. This result is consistent with previous research that found a positive relationship between the level of engagement and psychological distance (Hudson et al., 2016). Importantly, participants indicated lower value perceptions when the luxury brand's social media page displayed a high level of brandconsumer engagement than a low level of engagement. Moreover, formality, a measure of psychological distance, partially mediated the relationship between brand-consumer engagement and social and quality value perceptions. Specifically, a high level of brand-consumer engagement decreased formality of the luxury brand, which in turn decreased the value perceptions of the brand. 5. Study 3 The primary objective of Study 3 was to replicate, extend, and increase generalizability of the findings of Study 2. To do so, Study 3 manipulated the level of brand-consumer engagement in a different way, controlled the effect of brand awareness using a hypothetical luxury brand, and collected samples that are more representative of the U.S. consumers. Regarding the measures, Study 3 focused on measuring the social distance dimension of psychological distance to better reflect the research context of brand-consumer engagement on social media. Additionally, participants' attitudes toward the luxury brand's social media engagement strategy were measured in the study. This item was included to examine whether attitude is a good measure to capture the full impact of brand-consumer engagement of the luxury brand. 5.1. Study design and stimuli development Similar to Study 2, a single factor (brand-consumer engagement level: High vs. Low) between-subjects design was used. A fictitious luxury watch brand called “Suissse Majestät” was created to control prior knowledge and perceptions of brands (Shin, Eastman, & Mothersbaugh, 2017). The brand was described as a leading luxury brand made in Switzerland and the brand's watches are the symbols of excellence and performance (Shin et al., 2017). Study 2 decided to provide written scenarios to manipulate brandconsumer engagement to control possible confounding factors resulting from the attractiveness of stimuli photos and different levels of visual information. For the high level of brand-consumer engagement condition, it was described that Suisse Majestät has decided to increase engagement with all social media users as a social media strategy. Then, it was described that Suisse Majestät will follow back every social media user who follows or likes Suisse Majestät on social media. Also, Suisse Majestät will reach out social media users who tag the brand and leave Fig. 3. The mediation effect of formality on social perception Note: *p ≤ .05 **p ≤ .01 ***p ≤ .001. Fig. 4. The mediation effect of formality on quality perception Note: *p ≤ .05 **p ≤ .01 ***p ≤ .001. M. Park et al. Journal of Business Research 117 (2020) 529–542 534

a friendly comment on the users' posts or walls. For the low level of consumer engagement condition, Suisse Majestät was described that it has decided to maintain the current level of engagement with social media users. Specifically, Suisse Majestät was described that they will only follow social media users who are celebrities or brand ambassadors. Also, it was mentioned that Suisse Majestät will selectively respond to few social media users' comments on the brand's social media page (see Appendix B for the stimuli). 5.2. Instruments In terms of measurement, social, uniqueness, and quality value perception were measured using the same scale from Study 1. To better reflect the social distance dimension of psychological distance, two semantic differential measures, unapproachability (i.e., I think Suisse Majestät is very approachable (1) - unapproachable (7) to average consumers) and inaccessibility (i.e., I think Suisse Majestät is very accessible (1) - inaccessible (7) to average consumers) were used. Additionally, one item asking participants' attitude toward the brand's social media engagement strategy was measured on a semantic differential scale ranging from 1 (very negative) to 7 (very positive). As mentioned earlier, this item was included to examine whether the attitude is a good measure to capture the full impact of brand-consumer engagement. 5.3. Data collection and experimental procedure Using TurkPrime, MTurk participants were collected and those who completed Study 1 were excluded from Study 2 to prevent multiple submissions from the same participant. To increase the representativeness of the sample, participants with a wide range of household income were recruited until the approximate median income if the sample reached to $59,000, which is US median household income in 2016 (FRED, 2017). Participants were first introduced to the brand, Suissse Majestät. In order to increase involvement in the scenario, all participants were asked to imagine that they are financially well-off and they are an owner of a top-end Suissse Majestät watch. This technique was used in previous studies using a fictitious brand (Mandel, Petrova, & Cialdini, 2006; Shin et al., 2017). Then, the participants were told that they would read about the brand's social media strategy and answer some questions about it. On the next page, one of the two scenarios was randomly shown to the participants. After reading one of the two scenarios, participants completed a questionnaire. 5.4. Results 5.4.1. Participant characteristics A total of 248 participants completed the experiment. Participants' ages ranged from 19 to 74 years, with the average age of 38 years. Hundred thirteen (46%) of participants were men. The median household income ranged from $50,000 to $59,000. 5.4.2. Manipulation check The analysis of the manipulation check confirmed that the high level of brand-consumer engagement condition was perceived to have a higher consumer engagement level than the low level condition (Mhigh = 5.53 vs. Mlow = 2.42; t(246) = 16.13, p = .00). 5.4.3. Hypothesis testing Hypothesis 2 predicts that a luxury brand with a high level of brandconsumer engagement, compared to a low level of brand-consumer engagement, will be perceived as more socially close. The results the Independent t-tests confirmed that the high level of consumer engagement condition was perceived to be more socially close than the low level of consumer engagement condition (inaccessibility: Mhigh = 4.43 vs. Mlow = 5.58; t(246) = 5.26, p = .00, unapproachability: Mhigh = 3.98 vs. Mlow = 5.58; t(246) = 7.23, p = .00). Therefore, H2 was supported. To test H3, which predicts the effect of brand-consumer engagement on social, uniqueness, and quality perceptions, a series of t-tests were conducted. A significant effect of psychological distance was found on social value perception (Mhigh = 5.63 vs Mlow = 5.90, t(246) = −2.16, p < .05), uniqueness value perception (Mhigh = 5.61 vs Mlow = 5.92, t (246) = −1.91, p = .06), but not on quality value perception (Mhigh = 5.50 vs Mlow = 5.47, p > .05). Additional analysis showed that participants' attitude toward the brand's social media strategy was more positive when it had the high level of consumer engagement than the low one (Mhigh = 5.46 vs Mlow = 4.24, t(246) = 5.72, p = .00) (see Fig. 5). Following the same mediation procedure from Study 2, the mediating role of psychological distance on the relationship between consumer engagement and the outcome variables (H4) was tested. The results showed that inaccessibility and unapproachability, the measures of psychological distance, fully mediated the relationship between consumer engagement and social and uniqueness perceptions, but not Fig. 5. The mean differences in the outcome variables (Study 3). Note. *p ≤ .06, **p ≤ .01, ***p ≤ .001. M. Park et al. Journal of Business Research 117 (2020) 529–542 535

for quality perception. Specifically, when consumer engagement and inaccessibility were entered together as predictor variables of social and uniqueness perceptions, the beta weight for brand-consumer engagement became insignificant (social perception: c = 0.27 (t(246) = 2.16, p < .05) to c′ = 0.16 (t(245) = 1.21, p > .05); uniqueness perception: c = 0.31 (t(246) = 1.91, p = .05) to c′ = 0.11 (t(245) = 0.67, p > .05)) (see Fig. 6). Also, the bootstrapping supported the proposed mediation relationships (social perception: indirect effect: β = 0.11, SE = 0.04, 95% CI = 0.04 to 0.20; uniqueness perception: indirect effect: β = 0.20, SE = 0.06, 95% CI = 0.10 to 0.34). Similarly, when consumer engagement and unapproachability were entered together as predictor variables of social and uniqueness perceptions, the beta weight for brand-consumer engagement became insignificant (social perception: c = 0.27 (t(246) = 2.16, p < .05) to c′ = 0.16 (t(245) = 1.21, p > .05); uniqueness perception: c = 0.31 (t (246) = 1.91, p = .05) to c′ = 0.11 (t(245) = 0.67, p > .05)) (see Fig. 7). In addition, the bootstrapping supported the proposed mediation relationships (social perception: indirect effect: β = 0.12, SE = 0.06, 95% CI = 0.01 to 0.24; uniqueness perception: indirect effect: β = 0.22, SE = 0.07, 95% CI = 0.08 to 0.38). Therefore, H4 was partially supported. 5.5. Discussion Study 3 provides additional support for the findings of study 2 while ruling out the potential effect of brand awareness and using more representative sample of U.S. consumers. Specifically, a high level of brand-consumer engagement resulted lower social distance and social and uniqueness value perceptions. Moreover, Study 3 demonstrated that social distance of luxury brands (i.e., approachability, accessibility) fully mediated the relationship between brand-consumer engagement and social and uniqueness value perceptions of luxury brands. This suggests that social distance of luxury brands on social media is a critical variable that determines value perceptions of luxury brands. The insignificant main effect of brand-consumer engagement on quality value perception might be understandable because of the nature of perceived quality value. As discussed earlier, quality value perception may be more strongly influenced by factors such as price and craftmanship (Vigneron & Johnson, 2004) than social factors used in Study 3. Interestingly, participants' attitude toward the brand's social media engagement strategy was more positive in the high brand-consumer engagement condition while brand value perceptions were higher in the low engagement condition. This implies that although consumers have positive attitude toward a luxury brand's high level of consumer engagement on social media, it does not necessarily lead to higher perceived values for the brand. Therefore, the results suggest that the attitude measure is not adequate in capturing the impact of a luxury brand's consumer engagement strategy on brand values. 6. General discussion Across the experimental studies, this research highlights the impact of psychological distance of luxury brands triggered by a level of consumer-brand engagement on value perceptions of luxury brands. This research yields both theoretical and practical implications in the following ways. 6.1. Theoretical implications From a theoretical point of view, this study contributes to a body of literature concerning brand dilution and social media marketing of Fig. 6. The mediation effect of inaccessibility on social perception Note: *p ≤ .05 **p ≤ .01 ***p ≤ .001. Fig. 7. The mediation effect of unapproachability on uniqueness perception Note: *p ≤ .05 **p ≤ .01 ***p ≤ .001. M. Park et al. Journal of Business Research 117 (2020) 529–542 536

luxury brands, and to the theory of psychological distance. To date, the brand dilution literature has been dominantly examined from the aspects of brand extension strategies (e.g., Shin et al., 2017). On the other hand, this research demonstrates that brand-consumer engagement of luxury brands on social media can also dilute brand value perceptions. Therefore, this research's view on social media marketing as a source of brand dilution provides important insights and adds a new theoretical perspective to the literature on brand dilution. Also, while most of literature in social media marketing of luxury brands has documented positive aspects of social media on luxury brands (e.g., Kim & Ko, 2012), the current research contributes to the literature by uncovering negative aspects of social media marketing on luxury brands. Specifically, this study provides empirical evidence that a high level of brand-consumer engagement on social media can damage perceptions of luxury brands by revealing the underlying mechanism. Through a mediation analysis, it was found that this effect is partly due to the decrease of psychological distance of luxury brands. In regard to the theory, psychological distance has been examined and proved as a meaningful construct that is linked to strong consumer behavior outcomes such as product evaluations (Kim et al., 2008) and self-control (Kivetz & Simonson, 2002). This study also provides further support that psychological distance is an important construct that influences consumers' value perceptions of luxury brands on social media. While previous research has found that reducing psychological distance of online retailers is important for building trust (Darke et al., 2016; Edwards, Lee, & Ferle, 2009), the current study shows that, for luxury brands, it is vital to maintain the social distance dimension of psychological distance to protect value perceptions. Thus, this study offers insights concerning the applicability of the construal level theory of psychological distance to luxury brand perceptions on social media. 6.2. Practical implications With regards to managing luxury brands, one of the most important goals is to sustain the myth and dream of luxury (Kapferer & Bastien, 2012). This research provides a strong warning for luxury brands who may stray from the goal due to their social media strategies. The findings suggest that luxury brands should maintain sacred psychological distance on social media; otherwise it will undermine important value perceptions of the luxury brands such as exclusivity, status signaling, and quality, which eventually damages the luxury dream. Specifically, overly active and friendly brand-consumer engagement on social media may backfire luxury brands because consumers may perceive the brands to be too accessible and approachable to everyday consumers. Therefore, it may be more beneficial for luxury brands to selectively engage with consumers and only follow a certain group of consumers (e.g., high-profile celebrities or artists) on social media to demonstrate that it is maintaining psychological distance to mass market consumers. Moreover, as evidenced in Study 3, consumers' positive attitude toward the high level of brand-consumer engagement on social media does not necessarily translate into higher value perceptions of luxury brands. That is, highly active brand-consumer engagement which may appear as positive can actually lower the core perceptions of luxury brands. As managers of luxury brands attempt to increase brand-consumer engagement on social media extensively, they must be mindful of the potential negative consequences on how the brand is perceived. However, it is possible that a high level of brand-consumer engagement may offer other potential positive outcomes (e.g., WOM, higher brand awareness) for luxury brands. Therefore, luxury brands must weigh the benefits of actively engaging with consumers against the cost of reducing core value perceptions of the brands. 6.3. Suggestions for future research Among various dimensions of psychological distance, this study particularly focused on social distance of luxury brands in the context of social media marketing. Specifically, this study demonstrated that active brand-consumer engagement is an important antecedent of social distance of luxury brands that can damage important perceptions of the brands. Along with a social distance dimension, future studies should investigate other factors that may influence other dimensions of psychological distance of luxury brands, such as temporal and spatial distance. For example, some luxury brands on social media frequently provide direct links to their online stores along with product photos to increase both sales and consumer shopping convenience. However, this social media marketing tactic may be another factor that negatively influences perceptions of luxury brands because it may reduce perceived spatial distance of the brands. In other words, because an online store symbolizes increased accessibility of products and brands due to its ubiquity (Okonkwo, 2009), consumers may perceive the luxury brands as being close, accessible, and within reach, which subsequently undermines exclusivity perceptions of the brands. Moreover, future studies could investigate variables that may moderate the relationship between psychological distance and evaluations of luxury brands to show boundary conditions. Consumer-related variables such as power, social goals (i.e., competition, assimilation), and need for status may moderate such relationship because of the fit between the symbolism of luxury brands and consumers' needs. For example, consumers with high need for status may evaluate psychologically distant luxury brands more favorably than consumers with low need for status because such brands are perceived to be more conspicuous. Lastly, the current research measured psychological distance of luxury brands using measures that are more focused on social distance aspects. Although previous research has confirmed that social distance is an indicator of psychological distance, future research could further develop and test other measures of psychological distance. For example, developing measures that assess temporal distance and spatial distance relevant to the luxury brands context will create more comprehensive measures of psychological distance of luxury brands. M. Park et al. Journal of Business Research 117 (2020) 529–542 537

Appendix A. Manipulations in Study 2 High brand-consumer engagement manipulation M. Park et al. Journal of Business Research 117 (2020) 529–542 538

M. Park et al. Journal of Business Research 117 (2020) 529–542 539

Low brand-consumer engagement manipulation M. Park et al. Journal of Business Research 117 (2020) 529–542 540

Appendix B. Scenarios and manipulations in Study 3 Base scenario Suisse Majestät is a leading luxury watch brand founded by watch artisan Marco Müller in 1921 in Switzerland. Suisse Majestät produces one of the world's most finely crafted timepieces and its watches are the symbols of excellence and performance. Please imagine that you are financially well-off and you own the top-end Suisse Majestät watch. High brand-consumer engagement manipulation As a social media strategy, Suisse Majestät has decided to increase engagement with all social media users. Specifically, Suisse Majestät will follow back every social media user who follows or likes Suisse Majestät on social media. Also, Suisse Majestät will reach out social media users who tag the brand and leave a friendly comment on the users' posts or walls. Low brand-consumer engagement manipulation As a social media strategy, Suisse Majestät has decided to maintain the current level of engagement with social media users. Specifically, Suisse Majestät will only follow social media users who are celebrities or brand ambassadors. Also, Suisse Majestät will selectively respond to few social media users' comments on the brand's social media page. References Aaker, D. A. (1996). Measuring brand equity across products and markets. California Management Review, 38(3), 102–120. Akerlof, G. A. (1997). Social distance and social decisions. Econometrica: Journal of the Econometric Society, 65(5), 1005–1027. Bellezza, S., & Keinan, A. (2014). Brand tourists: how non–core users enhance the brand image by eliciting pride. Journal of Consumer Research, 41(2), 397–417. Blasco-Arcas, L., Holmqvist, J., & Vignolles, A. (2016). Brand contamination in social media: Consumers' negative influence on luxury brand perceptions—A structured abstract. (In L). Bourdieu, P. (1989). Social space and symbolic power. Sociological Theory, 7(1), 14–25. Brock, T. C. (1968). Implications of commodity theory for value change. In A. G. Greenwald, T. C. Brock, & T. M. Ostrom (Eds.). Psychological foundations of attitudes (pp. 243–275). New York: Academic. Chu, S. C., Kamal, S., & Kim, Y. (2013). Understanding consumers' responses toward social media advertising and purchase intention toward luxury products. Journal of Global Fashion Marketing, 4(3), 158–174. Darke, P. R., Brady, M. K., Benedicktus, R. L., & Wilson, A. E. (2016). Feeling close from afar: The role of psychological distance in offsetting distrust in unfamiliar online retailers. Journal of Retailing, 92(3), 287–299. De Barnier, V., Falcy, S., & Valette-Florence, P. (2012). Do consumers perceive three levels of luxury? A comparison of accessible, intermediate and inaccessible luxury brands. Journal of Brand Management, 19(7), 623–636. Desai, F. (2016). How digital challengers are redefining luxury. Retrieved from http:// www.forbes.com/sites/falgunidesai/2016/07/16/how-digital-challengersareredefining-luxury/2/#66a2a3d96bff. Dew, L., & Kwon, W. S. (2010). Exploration of apparel brand knowledge: Brand awareness, brand association, and brand category structure. Clothing and Textiles Research Journal, 28(1), 3–18. Dittmar, H. (1994). Material possessions as stereotypes: Material images of different M. Park et al. Journal of Business Research 117 (2020) 529–542 541

socioeconomic groups. Journal of Economic Psychology, 15(4), 561–585. Dubois, B., & Paternault, C. (1995). Observations: Understanding the world of international luxury brands: The “dream formula”. Journal of Advertising Research, 35(4), 69–76. Edwards, S. M., Lee, J. K., & Ferle, C. L. (2009). Does place matter when shopping online? Perceptions of similarity and familiarity as indicators of psychological distance. Journal of Interactive Advertising, 10(1), 35–50. Fionda, A. M., & Moore, C. M. (2009). The anatomy of the luxury fashion brand. Journal of Brand Management, 16(5-6), 347–363. FRED (2017). Real median household income in the United States. Retrieved from https://fred.stlouisfed.org/series/MEHOINUSA672N. Fuchs, C., Prandelli, E., Schreier, M., & Dahl, D. W. (2013). All that is users might not be gold: How labeling products as user designed backfires in the context of luxury fashion brands. Journal of Marketing, 77(5), 75–91. Gilbert, E., & Karahalios, K. (2009, April). Predicting tie strength with social media. Proceedings of the SIGCHI conference on human factors in computing systems (pp. 211– 220). ACM. Gjesme, T. (1981). Is there any future in achievement motivation? Motivation and Emotion, 5(2), 115–138. Gummerus, J., Liljander, V., Weman, E., & Pihlström, M. (2012). Customer engagement in a Facebook brand community. Management Research Review, 35(9), 857–877. Heavey, D. (2017). Examples of engaging social media posts. Retrieved from https:// thrivehive.com/examples-engaging-social-media-posts/. Heine, K. (2012). The concept of luxury brands. Retrieved from http://www. conceptofluxurybrands.com/content/Heine_TheConceptofLuxuryBrands.pdf. Hennigs, N., Wiedmann, K. P., Behrens, S., & Klarmann, C. (2013). Unleashing the power of luxury: Antecedents of luxury brand perception and effects on luxury brand strength. Journal of Brand Management, 20(8), 705–715. Holt, D. B. (1995). How consumers consume: A typology of consumption practices. Journal of Consumer Research, 22(1), 1–16. Hudson, S., Huang, L., Roth, M. S., & Madden, T. J. (2016). The influence of social media interactions on consumer–brand relationships: A three-country study of brand perceptions and marketing behaviors. International Journal of Research in Marketing, 33(1), 27–41. Hung, K. P., Huiling Chen, A., Peng, N., Hackley, C., Amy Tiwsakul, R., & Chou, C. L. (2011). Antecedents of luxury brand purchase intention. The Journal of Product and Brand Management, 20(6), 457–467. Jiang, M., Gao, D. G., Huang, R., Dewall, C. N., & Zhou, X. (2014). The devil wears Prada: Advertisements of luxury brands evoke feelings of social exclusion. Asian Journal of Social Psychology, 17(4), 245–254. Kapferer, J. N. (1997). Managing luxury brands. Journal of Brand Management, 4(4), 251–259. Kapferer, J. N., & Bastien, V. (2012). The luxury strategy: Break the rules of marketing to build luxury brands. Kogan Page Publishers. Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of social media. Business Horizons, 53(1), 59–68. Kim, A. J., & Ko, E. (2012). Do social media marketing activities enhance customer equity? An empirical study of luxury fashion brand. Journal of Business Research, 65(10), 1480–1486. Kim, K., Zhang, M., & Li, X. (2008). Effects of temporal and social distance on consumer evaluations. Journal of Consumer Research, 35(4), 706–713. Kivetz, R., & Simonson, I. (2002). Earning the right to indulge: Effort as a determinant of customer preferences toward frequency program rewards. Journal of Marketing Research, 39(2), 155–170. Labrecque, L. I. (2014). Fostering consumer–brand relationships in social media environments: The role of parasocial interaction. Journal of Interactive Marketing, 28(2), 134–148. Lee, H. C., Chen, W. W., & Wang, C. W. (2015). The role of visual art in enhancing perceived prestige of luxury brands. Marketing Letters, 26(4), 593–606. Liberman, N., & Trope, Y. (2008). The psychology of transcending the here and now. Science, 322(5905), 1201–1205. Mandel, N., Petrova, P. K., & Cialdini, R. B. (2006). Images of success and the preference for luxury brands. Journal of Consumer Psychology, 16(1), 57–69. Miyazaki, A. D., Grewal, D., & Goodstein, R. C. (2005). The effect of multiple extrinsic cues on quality perceptions: A matter of consistency. Journal of Consumer Research, 32(1), 146–153. Okonkwo, U. (2009). Sustaining the luxury brand on the internet. Journal of Brand Management, 16(5–6), 302–310. Park, H. J., Rabolt, N. J., & Jeon, K. S. (2008). Purchasing global luxury brands among young Korean consumers. Journal of Fashion Marketing and Management, 12(2), 244–259. Peterson, R. (2015). 10 best brand examples of social media engagement. Retrieved from http://barnraisersllc.com/2015/03/10-best-brand-examples-of-social-mediaengagement/. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. Quelch, J. A. (1987). Marketing the premium product. Business Horizons, 30(3), 38–45. Rao, A. R., & Monroe, K. B. (1989). The effect of price, brand name, and store name on buyers' perceptions of product quality: An integrative review. Journal of Marketing Research, 26(3), 351–357. Reed, D. (2015). Why does the digital world challenge luxury brands? Retrieved from http://www.wearejourney.co.uk/why-does-the-digital-world-challenge-luxurybrands/. Riley, F. D. O., Lomax, W., & Blunden, A. (2004). Dove vs. Dior: Extending the brand extension decision-making process from mass to luxury. Australasian Marketing Journal (AMJ), 12(3), 40–55. Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on consumer perceptions of brands. Journal of Marketing Communications, 22(2), 189–214. Schneider, M. (2015). 24 statistics that show social media is the future of customer service. Retrieved from http://www.socialmediatoday.com/social-business/24- statistics-showsocial-media-future-customer-service. Shin, H., Eastman, J. K., & Mothersbaugh, D. (2017). The effect of a limited-edition offer following brand dilution on consumer attitudes toward a luxury brand. Journal of Retailing and Consumer Services, 38, 59–70. Shukla, P., & Purani, K. (2012). Comparing the importance of luxury value perceptions in crossnational contexts. Journal of Business Research, 65(10), 1417–1424. Slepian, M. L., Ferber, S. N., Gold, J. M., & Rutchick, A. M. (2015). The cognitive consequences of formal clothing. Social Psychological and Personality Science, 6(6), 661–668. Trope, Y., & Liberman, N. (2010). Construal-level theory of psychological distance. Psychological Review, 117(2), 440–463. Truong, Y., Simmons, G., McColl, R., & Kitchen, P. J. (2008). Status and conspicuousness–are they related? Strategic marketing implications for luxury brands. Journal of Strategic Marketing, 16(3), 189–203. Tsai, W. S., Yang, Q., & Liu, Y. (2013). Young Chinese consumers' snob and bandwagon luxury consumption preferences. Journal of International Consumer Marketing, 25(5), 290–304. Tsimonis, G., & Dimitriadis, S. (2014). Brand strategies in social media. Marketing Intelligence & Planning, 32(3), 328–344. Tungate, M. (2009). Luxury world: The past, present and future of luxury brands. London, UK: Kogan Page Publishers. Verhallen, T. M. (1982). Scarcity and consumer choice behavior. Journal of Economic Psychology, 2(4), 299–322. Verhallen, T. M., & Robben, H. S. (1994). Scarcity and preference: An experiment on unavailability and product evaluation. Journal of Economic Psychology, 15(2), 315–331. Vigneron, F., & Johnson, L. W. (2004). Measuring perceptions of brand luxury. Journal of Brand Management, 11(6), 484–506. Wiedmann, K. P., Hennigs, N., & Siebels, A. (2009). Value-based segmentation of luxury consumption behavior. Psychology and Marketing, 26(7), 625–651. Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206. M. Park et al. Journal of Business Research 117 (2020) 529–542 542

Contents lists available at ScienceDirect Journal of Business Research journal homepage: www.elsevier.com/locate/jbusres Behind the runway: Extending sustainability in luxury fashion supply chains Hakan Karaosmana,⁎ , Patsy Perryb , Alessandro Bruna , Gustavo Morales-Alonsoc a Politecnico di Milano, Italy b The University of Manchester, UK c Universidad Politecnica de Madrid, Spain ARTICLE INFO Keywords: Sustainability Supply chain management Luxury fashion Qualitative Case study ABSTRACT From a resource-based view, this paper investigates sustainability integration across multiple tiers in two Italian luxury supply chains producing fashion and leather footwear, a complex and fragmented industry sector that is highly dependent upon raw materials and a human skills base. Qualitative in-depth interview data were collected from senior industry informants within 10 businesses, spanning multiple supply chain tiers. Industry practices are systematically decomposed into product, process and supply chain levels to analyse supply chain sustainability. Findings reveal that product-level practices focused on raw materials more than design initiatives, with operational benefits of cost reduction and market benefits of consumer value-add. Process-level practices in water and energy reduction were motivated by cost reduction benefits more than environmental concerns. At supply chain level, traceability projects and supplier audits were limited by a lack of end-to-end supply chain visibility, despite the criticality of raw materials and evidence of close and long-term trading relationships. Supply chain transparency and supplier engagement are critical areas for development. Both technical and relational resources must be developed across supply networks. Current practices are geared towards reducing negative impacts associated with current operations, falling short of the radical strategies needed to address root causes and embrace sustainability at large. 1. Introduction Globalization and growing reliance on network relationships have led to a renewed focus on supply chain management (SCM) for enhanced competitiveness (Liu, Zhu, & Seuring, 2017). Nevertheless, exploitation of resources across globally dispersed supply chains results in many environmental and social issues (Freise & Seuring, 2015). One of the biggest challenges of global corporations with globally dispersed supply chains is the integration of social and environmental protection in business (Gualandris, Klassen, Vachon, & Kalchschmidt, 2015). Sustainability refers to the nexus of social, environmental and economic considerations in business. It exceeds organizational boundaries (Marshall, McCarthy, Heavey, & McGrath, 2015), as the most critical environmental and social issues are generated by suppliers located in the second tier or further upstream (Lee & Klassen, 2008; Tachizawa & Wong, 2014). While smaller suppliers may not be subject to global critique for irresponsible behaviours (Carrigan, McEachern, Moraes, & Bosangit, 2017), their unethical actions could damage a buying firm's sustainability performance (Huq, Chowdhury, & Klassen, 2016; Sancha, Gimenez, & Sierra, 2016) and thus companies cannot be considered more sustainable than their suppliers (Sancha et al., 2016). Companies must therefore seek to improve the performance of not only their own operations but also their suppliers' (Fritz, Schöggl, & Baumgartner, 2017). However, suppliers' ability to adhere to social and environmental sustainability requirements is often compromised by lead retail buyers' unrealistic expectations of cost and speed (Perry, Wood, & Fernie, 2015). The luxury industry traditionally traded upon premium quality, craftsmanship and exclusivity. Although it has not been associated with concerns about environmental and social responsibility to the same extent as the mid-market and fast fashion sector, a number of recent trends such as changing social norms, consumer and investor expectations, have led to an increased focus on luxury sustainability (Winston, 2016). Furthermore, the recent democratization of luxury, whereby luxury brands have extended themselves to affordable offerings for the masses (Cristini, Kauppinen-Raisanen, Barthod-Prothade, & Woodside, 2016), has resultant implications for cost, volume, lead times and sustainability. Management of sustainability in luxury supply chains becomes more critical as luxury companies are increasingly faced with the pressures of the mid-market fashion sector, including downward price pressure, international sourcing, high product variety and low predictability (Perry et al., 2015). Although luxury's higher price points https://doi.org/10.1016/j.jbusres.2018.09.017 Received 13 November 2017; Received in revised form 18 September 2018; Accepted 19 September 2018 ⁎ Corresponding author. E-mail address: [email protected] (H. Karaosman). Journal of Business Research 117 (2020) 652–663 Available online 29 September 2018 0148-2963/ © 2018 Elsevier Inc. All rights reserved. T

have traditionally protected it from negative media attention and consumer activism around social and environmental responsibility, this is no longer the case (O'Flaherty, 2017). In 2017 for example, it was reported that Louis Vuitton shoes were made in the lower labour cost country Romania before being shipped to Italy for the final production process, including soles being stamped with ‘Made in Italy’ (Lembke, 2017). Product responsibility failures have been reported, such as children's wear products from luxury brands including Dior, Hermès and Louis Vuitton testing positive for one or more of hazardous chemicals (Greenpeace, 2014). Dior scored zero in Fashion Revolution's Transparency Index, in which companies are ranked according to how much they disclose about their social and environmental policies, practices and impact (Rivera, 2017). However, increasing implementation of social and environmental initiatives is apparent (Acabou & Dekhili, 2013; Guercini & Ranfagni, 2013; Winston, 2016), partly as a response to increasing stakeholder pressure from regulators, NGOs, consumers and the media, including high profile campaigns such as Greenpeace's ‘Detox’, Fashion Revolution's ‘Who Made My Clothes?’ and Zero Discharge of Hazardous Chemicals' (ZDHC) programme to eliminate hazardous chemicals from global supply chains. Despite its increasing importance, the principles of sustainability can be perceived as incompatible with those of luxury, given luxury's focus on unique, valuable and rare materials that offer the dream factor to the consumer, rather than environmental or ethical concerns (Yang, Han, & Lee, 2017). Although raw material production is critical in luxury, this is usually beyond the scope and expertise of the lead firm, and some materials are scarce in terms of the limited geographical ranges from which they derive (Crowley et al., 2015), leading to complexity in supply chains. Italy is a key global sourcing centre for luxury fashion and leather products. The Italian tanning industry is the global leader, accounting for 11.6% of the global tannery sector and 13% by value of all leather goods items exported worldwide (Conseil National du Cuir, 2017). Italy is also known for its high service levels, product excellence and country of origin credentials (Brun et al., 2008) and for distinctive strategies in production, sourcing and distribution processes (Caniato, Caridi, Castelli, & Golini, 2011), all of which are critical in luxury fashion. In addition, Italian companies, as key actors in the luxury supply chain, have had to respond to increasing levels of sustainability regulation and calls for greater transparency in social and environmental policies and practices from a variety of stakeholders. For example, Caniato, Caridi, Crippa, and Moretto (2012) showed how internal, market and legislation-based drivers push luxury companies to implement environmental sustainability practices at product, process and supply chain levels. This paper presents a multiple case study of two Italian luxury supply chains to investigate how companies within a sector where production is dependent upon raw materials and specialised technical competencies cope with environmental and social issues, and to what extent businesses in different tiers create and implement strategies to control their impact and enhance sustainability across the supply chain. The paper is structured as follows: Section 2 reviews the literature and presents the theoretical background to sustainable supply chain management (SSCM), Section 3 explains the research methodology and Section 4 presents the findings from the empirical research. Section 5 provides a discussion of the key findings, and Section 6 delivers conclusions and identifies implications for theory and practice, as well as suggestions for future research. 2. Sustainable supply chain management (SSCM) Literature on SSCM encapsulates a number of key themes, including motivations for environmental and social sustainability implementation, resources needed, the importance of upstream supplier integration and the influence of contextual variables. SSCM involves the management of material, information and capital flows and the cooperation between actors throughout supply chains, whilst considering all three dimensions of sustainable development (Seuring & Muller, 2008). It aims to meet stakeholder requirements and improve profit and competitiveness, alongside increasing environmental efficiency and social responsibility (Gualandris, Golini, & Kalchschmidt, 2014). Some earlier research presents sustainability as a single concept combining environmental and social sustainability (Pagell & Wu, 2009; Parmigiani, Klassen, & Russo, 2011). However, environmental and social sustainability are separate concepts with their own antecedents, processes and results (Pullman, Maloni, & Carter, 2009); and there could be a tradeoff between these (Pfeffer, 2010) as companies with strong sustainability practices in one dimension could have weak practices in another (Huq et al., 2016). The vagueness of what sustainability and sustainable development actually refer to could even allow businesses to claim that they are committed to sustainable development whilst actually perpetrating unsustainability (Missimer, Robèrt, & Broman, 2017b). Motivation can largely be explained by self-interest: companies are more inclined to implement social sustainability practices to mitigate the negative impact of their operations rather than build truly sustainable supply chains (Rodriguez, Gimenez-Thomson, Arenas, & Pagell, 2016). There is ample evidence of environmental sustainability in SCM literature (Caniato et al., 2012; Laari, Töyli, & Ojala, 2017; Lee & Klassen, 2008; Vachon & Klassen, 2006; Wu & Pagell, 2011), much of which suggests that it ‘pays to be green’ (Hart & Dowell, 2011), but limited attention has been paid to social sustainability, which is often difficult to quantify in objective measures (Fritz et al., 2017; Huq et al., 2016; Missimer et al., 2017b; Missimer, Robèrt, & Broman, 2017a; Quarshie, Salmi, & Leuschner, 2016; Sancha et al., 2016; Winter & Lasch, 2016; Yawar & Seuring, 2017), making it prone to subjectivity (Beske-Janssen, Johnson, & Schaltegger, 2015). Notwithstanding recent noticeable studies (e.g. Carrigan et al., 2017; Eizenberg & Jabareen, 2017; Köksal, Strähle, Müller, & Freise, 2017), there is a lack of empirical and theoretical investigations pertaining to social sustainability (Ansari & Kant, 2017; Eizenberg & Jabareen, 2017). Social sustainability is oversimplified in extant theoretical constructs (Missimer et al., 2017b) and there are few studies in which the interrelationships between social and environmental practices are explored (Marshall et al., 2015; Wang & Dai, 2018), even though practices must be taken into account simultaneously and strategically since supply chain management encompasses both natural and human resources. Much empirical SSCM research has been conducted in manufacturing industries, but with limited investigation in fashion and textiles (Ansari & Kant, 2017) or luxury supply chains (Towers, Perry, & Chen, 2013), despite the detrimental environmental and social issues that characterise these sectors. 2.1. Resources needed for SSCM According to the resource-based view (RBV), collaboration can enable partnering companies to develop valuable, rare and difficult to imitate resources that could lead to competitive advantage (Sancha et al., 2016). The natural resource-based view (NRBV) extended the RBV by taking into consideration the natural environment in terms of three key strategic capabilities of pollution prevention, product stewardship and sustainable development (Hart, 1995; Hart & Dowell, 2011). NRBV assumes a positive relationship between capabilities and organizational performance (Aragon-Correa & Sharma, 2003; Wagner, 2015), for example pollution prevention technologies could enable the creation of complex capabilities which are difficult to duplicate (Schoenherr, 2012). According to Vachon and Klassen (2008), upstream collaboration impacts process-based environmental performance while downstream collaboration is more linked to product-based performance. Proposing a dynamic view of resources, NRBV considers resource transferability an imperative and proposes that establishing an environmental management strategy for resources can lead to sustained competitive advantage (Vachon & Klassen, 2008). As noted by Marshall H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 653

et al. (2015), the complexity, path dependence and learning from sustainable supply chain practices can result in the development a unique source of competitive advantage (Reuter, Foerstl, Hartmann, & Blome, 2010), especially where such practices help to solve problems, sense opportunities and threats, make market-oriented decisions and change a company's resource base (Barreto, 2010). Organizational capabilities and managerial cognition are necessary to achieve financial benefits from sustainability practices (Hart & Dowell, 2011). By working with suppliers, companies may develop resources that lead to better environmental performance (Sancha et al., 2016). Knowledge exchange requires relational resources and Ashby (2018) identified three types of strategic resources including physical (e.g. raw materials), tacit (e.g. skills, reputation, firm culture) and socially complex resources (e.g. friendship, shared vision, trust and commitment) needed to progress towards a closed loop supply chain. The nature of buyer-supplier relationships and the existence of trust and commitment between trading partners influence sustainability performance (Perry et al., 2015). Wilhelm, Blome, Wieck, and Xiao (2016) identified a number of contingencies across multi-tier supply chains which impede the extension of sustainability to upstream suppliers, including supply chain complexity and transparency, power asymmetries and supplier technical expertise. Alongside technical expertise, collaboration and commitment emerge as key antecedents of SSCM, but there remains a need for empirical understanding of the origin of key resources and the link between such resources and sustainability practices. 2.2. Extending SSCM across multiple tiers Despite increasing academic attention to SSCM, there is some disconnect between real-life sustainability problems and what has been covered in research. Sustainability measurement must get closer to where the real impact is generated within the chain (Beske-Janssen et al., 2015). However, much existing debate is restricted to the first tier supplier level rather than considering the entire chain across multiple tiers of suppliers (Yawar & Seuring, 2017) and it remains unclear how environmental change could be stimulated at further levels upstream (Lee, Klassen, Furlan, & Vinelli, 2014). Increased cooperation is needed along the chain for companies to achieve sustainability goals (Govindan, 2018) but there is a need for greater understanding of how sustainability collaboration could be developed at supply chain level (Blome, Paulraj, & Schuetz, 2014). To achieve broader stakeholder engagement, knowledge sharing and dialogue is needed and should be guided by principles of materiality, reliability, accuracy, completeness and responsiveness (Gualandris et al., 2015). Extant literature calls for more real-world examples in terms of practitioners' understanding to further interpret the concept of SSCM (Busse, Schleper, Niu, & Wagner, 2016), including an understanding of activities needed to engage suppliers beyond tier one (Quarshie et al., 2016) and examination of how sustainability practices could translate to second movers or to an entire industry (Pagell & Wu, 2009). The adoption of inter-organizational practices could help improve the sustainability performance of the buying firm (Gualandris et al., 2014), but it remains unknown how supply chain actors could make decisions for a variety of products with asymmetric information settings (Govindan, Seuring, Zhu, & Azevedo, 2016). Furthermore, buying firms need to develop certain capacities to implement sustainability strategies across the chain; however they may possess insufficient knowledge to do so alone, especially if their business has wide coverage and high diversity (Liu, Zhang, Hendry, Bu, & Wang, 2018). Thus, the role of contingencies such as industry and/or time-based factors should be considered (Busse et al., 2016; Pagell & Wu, 2009). Further in-depth evaluations within specific industries are thus necessary (Stindt, 2017), within a variety of company and industry specific investigations (Laari et al., 2017), as well as a multidisciplinary approach to explore the connections between product design, production processes and supply chain stages (Curwen, Park, & Sarkar, 2013). 2.3. The influence of firm size on SSCM Contextual factors including industry sector, supply chain position, company size, and company location may affect sustainability implementation. Company size may particularly impact sustainability activities and related outcomes (Pagell & Wu, 2009), as it influences the development of capabilities and the adoption of green strategies (Liu et al., 2017). Small and medium-sized enterprises (SMEs) often encounter challenges in responding to sustainability; for example, meeting social responsibility requirements might have high cost implications in terms of time and knowledge (Zhang, Pawar, & Bhardwaj, 2017). Caniato et al. (2012) found that small companies operating in footwear and outwear industries had greener processes, but did not measure environmental performance in a structured quantitative way due to their small size, which made it difficult to set up a planned performance measurement system. Most studies addressing supplier development strategies and their role in managing social sustainability in SCM do not address firm size, even though it is a determinant of social and economic performance (Yawar & Seuring, 2017). Already limited comprehension of responsible business practices in SMEs is further intensified by a failure to consider differences in SMEs and the contextual impacts in traditional sectors (Carrigan et al., 2017). SMEs are pivotal in SCM, representing more than 90% of all businesses and 60% of the employment rate of all countries (Fritz et al., 2017), but are not well researched, especially in terms of social sustainability (Nakamba, Chan, & Sharmina, 2017). In Italy, textile manufacturing is mostly characterised by SMEs (Froud, Hayes, Wei, & Williams, 2017) and they are an important part of Italian luxury fashion production. While the environmental and social consequences of actions taken by individual SMEs may be small, their joint impact may be significant. Hence, consideration of organizational and contextual variables that could influence sustainability implementation is necessary. 2.4. Summary and research questions The strategic importance of sustainability is clear, however more research is needed to understand the practices required to implement it, particularly for social sustainability (Huq et al., 2016) and the influence of context-specific challenges on the development of sustainable supply chains (Roy, Charan, Schoenherr, & Sahay, 2018). Despite pressure on suppliers to implement social standards in the management of social sustainability across multi-tier supply chains (Nakamba et al., 2017), little research has been conducted on what facilitates suppliers to respond to SSCM requirements (Roy et al., 2018) and few studies consider social practices at the micro level, in the context of supplier businesses or SMEs (Nakamba et al., 2017) or the willingness and ability of smaller suppliers to engage in partnerships within the SSCM context (Beske, Land, & Seuring, 2014). Hence, the efforts and challenges in implementing supply chain sustainability beyond the first-tier buyersupplier dyad seem underexplored. The following research questions were formulated to address these knowledge gaps: RQ1: How is sustainability extended across multiple supply chain tiers in the Italian luxury fashion industry? RQ2: How do contextual factors of the luxury fashion sector influence sustainability in supply chain management? 3. Research methodology Empirical research fails to capture some of the most essential ethical dimensions and the majority of the literature has adopted rather rationalist approaches, with relatively few contributions utilizing qualitative approaches (Dubey, Gunasekaran, Childe, Papadopoulos, & Wamba, 2017). By moving beyond the first tier buyer-supplier H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 654

relationship and examining multiple tiers, this study provides empirical insight on SSCM at the level of sub-suppliers, previously lacking in the literature (Wilhelm et al., 2016). A qualitative multiple case study was conducted, adopting purposive sampling to select a total number of ten companies within two Italian luxury supply chains for garments and leather shoes, encompassing brand owners, garment producers and raw material processors, all located in Italy. Both supply chains are comparable in size and maturity of network, with most brand owners and suppliers originally founded as family businesses in the 1940s–1950s and evidence of long-term trading relationships spanning more than one generation. The leather supply chain is mostly concentrated in Italy, while the textile supply chain is more geographically fragmented. Focusing on a single country and industrial sector allowed the researchers to control for variances in processes and materials that would not be feasible in cross-industry settings (Winter & Lasch, 2016). The case study process followed Yin's (2014) guidelines and previous qualitative research in sustainable supply chain management literature (Caniato et al., 2012; Carrigan et al., 2017; Pagell & Wu, 2009; Wilhelm et al., 2016). Data were collected from face-to-face semistructured interviews with relevant senior company representatives within each supply chain during 2015–16. Semi-structured interviews allowed the researcher freer rein to probe interesting areas that arose, while the respondent had greater opportunity to tell their own story and become more active in shaping the interview process (Smith & Eatough, 2007). Questions were used as a guide to generate discussion and gather factual information, starting with initial questions about general company information, including history, products and customers. Respondents were assured of personal and company confidentiality in order to build rapport and facilitate an honest dialogue (Perry et al., 2015). The second set of questions elicited detailed information on the theoretical constructs. Table 1 shows the sample company characteristics and interview details. In smaller companies, the president or CEO plus a sustainability officer participated in the interviews, while in large companies, the managers in charge of operations and SCM as well as sustainability team member(s) joined the meetings. Companies were structured differently and job descriptions significantly vary, especially when it comes to sustainability. In most companies, the responsibility for sustainability was divided and integrated into the jobs of multiple managers, meaning that there was no single individual assigned to sustainability. Additionally, the managers interviewed were often in charge of one or more of the functions of interest, which helped to reduce the number of interviews needed, but increased interview length. Facility tours and internal company documents were part of the data collection process at organizations with on-site production facilities. Finally, data were gathered from publicly available resources. Most interviews lasted over 1 h and all were recorded with permission of interviewees. Subsequent to each company visit, interview tapes were transcribed and interview notes were edited, which generated on average 10 pages of interview transcripts per company, in addition to relevant parts of publicly available reports and notes from site visits. For data analysis, the theoretical perspectives and constructs were used to provide an initial coding scheme, shown in Appendix A, which builds on Karaosman, Morales-Alonso, and Brun's (2016) taxonomy to evaluate real-life SSCM practices in multiple supply chain tiers. Withincase analysis was a process of data reduction and data management to structure, define, reduce and re-assemble the data. First, the business models of sample companies were studied. Then, programs, policies, and practices were categorised to understand how sustainability was implemented in terms of product design, production processes and supply chain execution. Next, sustainability practices were cross-referenced with the research taxonomy, shown in Appendix B. The final step was to synthesise all aforementioned elements. Following Yin's (2014) case study guidelines, data triangulation was achieved through the use of a case study protocol, multiple sources of evidence (interviews, documents, observation during site visits) and discussion of results with multiple investigators within the research team. Table 1 Sample company breakdown. Company Product category Supply chain activities No. of employees Interviewees Interview duration Leather 1 (Brand owner) Leather goods Design: In-house Production: Entirely outsourced Retail: Package finished products and transport to stores ~240 - Quality manager (leather goods) - Quality manager (readyto-wear) 1 h 20 min Leather 2 (Brand owner) Leather goods Design: In-house Production: From tanning processes to final product Retail: Package finished products and transport to stores ~1650 - Planning and distribution manager - Sustainability officer 1 h 40 min Leather 3 (Leather processor) Leather shoes Production: Re-tanning, dyeing and finishing processes ~200 - Production manager 1 h Leather 4 (Brand owner and OEM) Leather shoes Design: In-house for own brand, by client for OEM Production: Purchase finished leather and assemble final product Retail: Package finished product and transport to stores ~25 - CEO 1 h 10 min Textiles 1 (Brand owner) Cotton, silk, wool Design: In-house Production: Purchase yarn (and fabric) then produce fabric and manufacture garments Retail: Package finish products and transport to stores ~2000 - Head of global operations - Supply chain operations project intern 2 h Textiles 2 (Accessory manufacturer) Silk Purchase fabric then implement finishing treatments, manufacture products, package and transport to stores ~50 - CEO 1 h 35 min Textiles 3 (Fabric producer and garment manufacturer) Cotton, wool Purchase yarn, knit and finish fabric, conduct quality control of finishing treatments, manufacture garments, package and transport to stores ~50 - Founder 1 h 10 min Textiles 4 (Fabric producer and garment manufacturer) Wool, silk, cotton Purchase yarn (and in some cases fabric) and weave into fabric, manufacture garments, package and transport to stores ~50 - Founder 1 h Textiles 5 (Fabric producer) Silk Purchase yarn then produce fabric, send finished fabrics to clients ~50 - Production manager 1 h Textiles 6 (Textile processer) Silk, wool Purchase fabric and carry out dyeing, printing and finishing treatments ~100 - Director 1 h H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 655

4. Findings To investigate individual environmental and social practices pertaining to different sustainability categories at various operational levels, practices were decomposed into product, process and supply chain-oriented practices. The following cross-case analysis shows how companies across multiple tiers implemented sustainability and which components were included in their operations, against the backdrop of business environment challenges which impacted upon each supply chain. 4.1. Sustainability integration at product level In the textile supply chain, use of certified organic materials emerged as a growing trend that, despite greater comparable material costs, resulted in long-term cost savings as a result of environmental footprint reduction and better product quality, supporting the argument that it ‘pays to be green’. Since quality is a direct market success factor for luxury industries, customers' purchase decisions are positively influenced and value can be added. Organic materials were perceived as being beneficial in terms of better quality that could be reflected in price, as well as cost reduction benefits due to the reduction of water and chemical use in processing. As a result of increasing stakeholder pressure in the sector from powerful NGOs such as Greenpeace's ‘Detox’ campaign and multi-stakeholder groups such as the ZDHC Programme, chemical reduction was also a key concern across the textile companies and some had set targets to reduce certain chemicals from their production lines. ‘Textiles 1’ was an important example in this respect, having committed to the Zero Discharge of Hazardous Chemical by 2020 project, set up a ‘Product Safety Control System’ to guarantee a continuous and methodical control process, and developed a materials substance list, which was shared with direct suppliers. This materials substance list emerged as a strategic priority in the textile supply chain towards responsible chemicals management. Client tests, where the lead luxury brand would conduct tests in its own laboratories, emerged as a pivotal driver for suppliers to improve product quality and product related performance. SMEs in particular paid close attention to chemical use and use of better quality materials, knowing that their clients would double-check the product in their own laboratories. Leather companies were increasingly engaging with material assessment and chemical restrictions to ensure product safety, but a number of conflicts emerged between market requirements for material finish and feel, and sustainability issues. Chemical elimination is challenging, as certain chemicals are necessary to give the finished leather product a fine feeling which meets luxury market requirements. Materials research and determination of replacements is time-consuming, and the Sustainability Officer of ‘Leather 2’ explained that relational aspects were critical in working outside of ‘business-as-usual’ towards hazardous chemical elimination: “you need to find supportive partners in your supply chain that are willing to develop new kind of methods in order to fulfil your requirements”, and therefore substitution and elimination of chemicals from production processes was challenging. Eliminating chemicals, albeit important, could compromise the end product quality. Furthermore, lead brand owners were increasingly concerned about consumer perception, as sustainability and the use of ‘eco’ and/or ‘recycled’ terminology are still far from being attached to the luxury concept, and the end consumer may thus perceive sustainability negatively. Nevertheless, ‘organic’ was perceived to be a positive differentiator. Sustainability does not currently lead the design concept of luxury but rather takes a supporting role. Table 2 below summarizes best practices implemented to integrate sustainability at product level across the textile and leather supply chains. 4.2. Sustainability integration at process level Sustainability integration in operations management mainly emerged in terms of water management and energy efficiency, in response to growing concerns about pollution prevention. Companies tracked their energy use, water consumption and GHG emissions. ‘Textiles 2’ evidenced committed to environmental impact reduction through a number of energy-saving projects, including installation of a photovoltaic plant, a cogeneration plant and an internal depuration plant. The CEO of ‘Textiles 2’ confirmed that cost reduction was the main driver for energy efficiency: “the cogeneration plant has been implemented to reduce production costs”. ‘Textiles 5’ also achieved cost savings after installing photovoltaic panels in 2013. Water emerged as a key concern for leather companies in particular, due to its critical role in processing and its expensive nature and direct association with operational costs. According to the Production Manager of ‘Leather 3’, “the main problem of a tannery is the water consumption; that is an important voice in the balance sheet.” Sustainability integration was thus driven by cost reduction motivations, as explained by the Quality Manager in ‘Leather 1’: “The tanneries work with a lot of water to make the leather, so one of the main targets of the tannery is to decrease the water waste… Not because they think of sustainability, because they think of money (due to the fact that) water is quite expensive”. ‘Leather 2’ collected and stored rainwater for gardening reuse in their atelier. As water remediation plays a vital role in purify wastewater, this company also implemented a MRSL (manufacturing restricted substances list) project to eliminate chemicals from production processes and thus avoid discharge of such during the process. Chemical reduction combined with a proper waste management system is essential to reach water stewardship, which could eventually provide financial returns, as noted by the Sustainability Officer in ‘Leather 2’: “Eliminating chemicals or heavy metals in tanning processes can lead to water savings”. Sustainability management needs strong organizational commitments and there must be an alignment between financial and non-financial goals, as exemplified by ‘Leather 2’, which initiated an environmental profit and loss to move towards linking financial statements with non-financial goals. Leather companies revealed that eliminating certain chemicals from tanning processes resulted in reduced water consumption and improved wastewater treatment performance. However, evidence of struggles emerged in terms of how to integrate pollution prevention techniques in daily business activities. It is not feasible to control energy consumption or carbon footprint in independent networks in which the buying firm acts as a coordinator by outsourcing its business functions to third party suppliers. Vertical integration is becoming weaker in luxury fashion supply chains, resulting in difficulties to monitor suppliers' emission and energy performance. Therefore, textile companies in particular pursued a reactive arms' length approach to comply with regulations. Supply chain complexity impedes companies to take greater steps towards operational sustainability excellence because they mostly outsource their products and raw material processing is hidden. Italy's silk production, for example, has Table 2 Best practices in sustainability integration at product level. Best practice category Textile supply chain Leather supply chain Sustainable product design Chemical reduction Chemical reduction A matrix to categorize all chemical components A matrix to categorize all chemical components Use of certified raw materials Use of certified raw materials Client tests Client tests Life cycle assessment Life cycle assessment Use of eco-friendly (recycled, organic) materials H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 656

mostly been outsourced to Asia, hence production processes are not directly controllable, as explained by the Production Manager of ‘Textiles 5’: “Sixty years ago we had entire production of silk internally… we started from silkworm cultivation to the final product… but years passed and globalization created complexities, it was too costly to produce everything internally; it was less costly to outsource it”. Downstream, the consumer experience of luxury was observed being a potential hindrance for sustainability. Shoe boxes, for example, were made much bigger than what a pair of shoes would require; however, as part of the consumer experience, luxury brands provided oversized items to carry the product. In order to protect boxes from damage during transportation and to maximise in-store availability, vehicles are not fully loaded while transferring products from distribution centres to the retail stores, with negative consequences for carbon footprint. Notwithstanding some companies' use of hybrid vehicles in their fleets, packaging size and logistics optimization seemed to be underestimated issues. Store-to-store transfer further added to the issue, but could be reduced by distributing the right merchandise to the right store every day. Optimising car fleet use including hybrid vehicles could reduce carbon emissions and also increase operational efficiency, as seen in the case of ‘Leather 2’'s use of electrical vehicles to move items from store to store in Milan, as these were permitted to access restricted city centre zones to save time as well as environmental impact. Table 3 below displays the best practices at process level in terms of sustainable operations management. 4.3. Sustainability integration at supply chain level Raw materials are acquired from nature-based systems, which can lead to a number of environmental issues such as droughts, biodiversity degradation and water scarcity; while social issues such as job losses and health and safety issues challenge sourcing practices. Supply chain level includes sustainable sourcing management and organizational commitment to sustainability. However, supply chain mapping was clearer in the leather supply chain, where tanneries were the direct suppliers of the raw material, in comparison to the textile supply chain, where raw material production was located at the second tier or further upstream. Sustainability was integrated into sourcing management through traceability projects and supplier audits. Auditing was the main practice used to monitor sustainability, yet a number of issues challenge supplier audits to become an established practice and extended further upstream to raw material suppliers. Raw material processing takes place in a wide range of countries, which means brand owners may only conduct traceability projects using an arm's length approach. Traceability of raw material processing and production was challenging as large product varieties and supply fragmentation made it difficult for companies to trace their suppliers and subcontractors, in order to make processes controllable and completely accountable. The Production Manager of ‘Textiles 5’ acknowledged the geographic fragmentation of raw material supply which led to challenges in traceability: “Raw materials come from all over the world: cotton from Egypt or Turkey, polyester from Bulgaria or Romania, silk from India or China… the raw silk comes from China to Como then receives dyeing and other elaboration…. we do not exactly know all the components within (the fabrics)”. Similarly, the Director of ‘Textiles 6’ admitted that “Before the textile arrives to us, we don't know which kind of tests the suppliers have done”. Hence, visibility over raw material production and processing emerged as one of the most challenging facets across the textile supply chain. Despite the criticality of raw materials and evidence of close and longterm trading relationships in both supply chains, an inclusive picture of end-to-end supply chain sustainability was missing so actual risks remained unknown. ‘Leather 1’, ‘Leather 2’ and ‘Leather 3’ applied traceability practices and were able to get closer to their raw material processing suppliers, as raw materials originating from abroad were mostly processed in their own region. Certification of origin was a common practice. ‘Leather 3’ and ‘Leather 4’ asked their suppliers to provide them with certificates of material origin. However, brand owners did not pressure suppliers to provide them with such certificates. All SMEs (namely ‘Textiles 3’, ‘Textiles 4’, ‘Textiles 5’, ‘Textiles 6’, ‘Leather 3’ and ‘Leather 4’), revealed their key clients did not ask for detailed information about their subcontractors, nor about raw material sourcing suppliers. The CEO of ‘Leather 4’ stated: “When we ask our suppliers about their records for traceability, our suppliers, including ‘Leather 3’, are efficient to trace the leather but generally our clients don't ask us to trace or provide a specification, so we don't ask our suppliers (about their sustainability performance) … our clients, regarding social sustainability, never asked and never checked”. Similarly, ‘Leather 4’ perceived that commercial pressures had taken over the industry and the pressure of meeting performance-based operational requirements jeopardized detailed assessment in terms of material traceability. Furthermore, lack of trust was a significant barrier. SMEs tended to support each other well but did not always trust the lead brands. Many SMEs were aware that providing details of their own suppliers and subcontractors to support traceability might also enable their large clients to cut them out and directly approach other suppliers further upstream. The CEO of Leather 4 revealed: “I know it happened to some other companies that some jobs were lost because they were bypassed by their clients (after these companies gave their upstream suppliers information to their clients)”. Trust and knowledge transfer thus appeared to be rather weak in certain places, despite being of paramount importance in linking noneconomic goals with financial objectives. Further, multiple demands for sustainability monitoring and nonstandardized assessments impeded some suppliers in advancing their operational performance due to the time-consuming nature and multiplicity of different codes and requirements for each brand. The Production Manager of ‘Textiles 5’ explained: “Lots of different requests are coming from different clients, different types of certifications.” Consequently, SMEs in the textile supply chain struggled to create coherent measurement systems to measure, report and improve their overall performance. In terms of organizational commitment, there was evidence of positive perceptions towards moving away from ‘business as usual’ and embracing a sustainable future. The CEO of ‘Leather 2’'s holding company considered sustainability as an opportunity for business improvement, and this was evidenced by its range of innovative sustainability initiatives. Despite fairly limited evidence of sustainable practices, textile companies seemed to be aware that practices Table 3 Best practices in sustainability integration at process level. Best practice category Textile supply chain Leather supply chain Sustainable operations management Water stewardship Water stewardship Wastewater treatment Wastewater treatment Process innovation Process innovation Reduce/reuse of textile waste Reduce/reuse of textile waste Renewable energy generation Emission reduction Carbon offsetting Logistics planning optimization H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 657

implemented individually could eventually make the chain stronger. SMEs located in the same industrial districts disseminated knowledge and technical information with their counterparts. Suppliers separated commercial imperatives from sustainability goals, as shown by some textile companies that embarked on industrial collaborations in production zones with their competitors, as explained by the CEO of ‘Textiles 2’: “We share the same dyeing technology (in the zone). If someone finds a new dyer that is more efficient and effective, he shares this information with everyone here in the district… the competences are shared…” Preserving Italian values and promoting Italian product were important internal drivers for both lead buying firms and SME suppliers. Developing a better region in which to live and operate became more than an organizational commitment, but a personal one too. SMEs' commitment to sustainability was linked to their sense of attachment to the region, from which their families originated and their children would inherit. This feeling of sentimentality and sense of pride in the region in which they lived and worked was a powerful driver for SMEs' commitment to sustainability. Both supply chains made philanthropic donations to local communities, provided their own employees with retirement benefits such as insurance, provided education and training to future employees in an attempt to preserve the labour force and craftsmanship skills at risk of dying out in Italy. This sense of pride motivated some of the smaller companies to move beyond regulatory compliance and to proactively introduce some initiatives, without receiving client demands. Lastly, evidence suggested that providing decent, agreeable and healthy working conditions is fundamental to increase employee welfare. One of the most profound social outcomes is that ‘sustainable’ companies invest in decent work and labour management. However, client demands for social sustainability also played a part, as acknowledged by the CEO of ‘Leather 4’: “clients have become more conscious (regarding labour management), and therefore the illegal payments and work have almost disappeared in this zone and everything is well regulated”. Table 4 below summarizes best practices implemented at supply chain level in terms of sourcing management and organizational commitment to sustainability. 5. Discussion This study delivers important insights by investigating the critical role of first and second tier SME suppliers in a supply chain, and by providing an understanding of how SMEs could respond to commercial pressures and sustainability imperatives, despite their limited resources. Applying NRBV theory, the findings show how the characteristics of the business environment, such as supply chain complexity, uncertainty and lead-time pressure, impede sustainability implementation. Resources must be valuable, rare and inimitable, but organizational management processes are often socially complex, encompassing multiple actors which bring cultural and communication barriers as well as power dynamics. However, the implementation of environmental and social practices could create an inimitable resource, align sustainability with luxury goals and create potential to command a premium. In line with Ashby's (2018) findings, supplier relationships contribute to the achievement of SSCM. Shared vision appears to be a key resource enabling firms to move from reactive to more proactive environmental approaches. However, a proactive sustainability strategy is dependent on specific processes (including technological change), and a proactive approach to the natural environment requires the path dependence and embeddedness of, for example, stakeholder integration, higher-order learning, and continuous improvement (AragonCorrea & Sharma, 2003). For example, developing strategies for product stewardship requires an understanding of interdependencies. However, luxury fashion design is strictly protected and always performed in-house, so findings did not reveal any situation in which product stewardship strategies were pursued by including external stakeholders in the product design process. In accordance with Miemczyk, Howard, and Johnsen (2016), collaborative development and planning, and dynamic management of the chain emerged as two required antecedents. However, it is more difficult to make such radical changes in a complex business environment. Results reveal that information asymmetry and lack of trust increased the complexity of sustainability implementation. Many SMEs were reluctant to provide specific details of their own suppliers, which would enable greater visibility, because large buying firms could cut them out by directly approaching lower tier suppliers. Strong bargaining power possessed by large buying firms might oblige small and Table 4 Best practices in sustainability integration at supply chain level. Best practice categories Textile supply chain Leather supply chain Sustainable sourcing management Process certification Process certification Environmental certification of suppliers Environmental certification of suppliers Certificate of origin Certificate of origin Procurement certification Procurement certification Monitoring factory compliance Monitoring factory compliance Certification for social sustainability (to check working hours) Certification for social sustainability (to check working hours) Training and education material for suppliers Training and education material for suppliers Supplier/partner selection based on environmental and social sustainability practices Two tier supply chain audit system Developing programs for labour practices and decent work conditions Inspection and controls at potential supplier facilities Protecting female workforce Traceability Female empowerment through new jobs External and accredited auditor Organizational commitment to sustainability Philanthropic donations to local communities Philanthropic donations to local communities Training on Code of Ethics and ethical purchasing Training on Code of Ethics and ethical purchasing Multilateral dialogue with relevant bodies, institutions, associations and communities Multilateral dialogue with relevant bodies, institutions, associations and communities Insurance and retirement benefits Insurance and retirement benefits Educational activities for future employees Educational activities for future employees Providing a career path and better skilled human resource management Environmental profit and loss Fire safety training Providing sustainable working conditions to artisans Coaching sessions and counselling meetings Staff awareness programs for traceability H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 658

medium-sized suppliers to provide some strategic information in terms of material producers, but suppliers in both chains provided only limited information in order to preserve their own market share. This finding provides further explanation for Huq et al. (2016) finding that codes of conduct encourage suppliers to undertake ‘just enough’ responsible practices to avoid non-compliance; but prevent companies from significantly increasing social sustainability over the long term. There is a need to understand how to resolve social and environmental problems more broadly, rather than reducing negative impacts associated with current operations (Hart & Dowell, 2011); however the lack of trust presents a challenge to understanding how an ideal profile of sustainability collaboration could be created at supply chain level. It also hampers the development of more radical strategies to improve products and processes that must go beyond an arm's length approach. Commercial and lead-time pressures impeded large buying firms from engaging with their dispersed supply chain actors. Hence, more collaborative approaches in which different supply chain stages are efficiently and effectively coordinated are needed to achieve SSCM. These relationships could ultimately contribute to resources that a company could spread in its supply chain, and could provide the sustainable competitive advantage supported by the NRBV, as suggested by Ashby (2018). Results also support Aragon-Correa and Sharma (2003) and reveal that when a company develops a proactive sustainability approach in a complex business environment, it can create a competitive advantage. For example, creating resources that provide unique characteristics could lead to first-mover advantage, as seen with ‘Textiles 1’ which became ‘fur-free’ before many of its main competitors and thus managed to avoid a competitive market. Coordination and cooperation of supply chain parties plays a vital role because establishing sustainable partners, in case of limited resources, is essential to achieve sustainable development (Govindan, 2018). Resource transferability is an imperative aspect to support SMEs' sustainability implementation, but requires a relational orientation. Nonetheless, today's luxury sector falls short in terms of supplier engagement and integration. Large buying firms often form indirect supplier development programs in which the company provides limited resources to suppliers, and some common practices take place, such as supplier assessment, audits and performance measurement. However, results indicate that SMEs need large buying firms to create a shared value concept. Therefore, better engagement and relationship management are needed along the entire chain. SMEs need to be competitive and adaptive, nevertheless the required support from their clients was sometimes lacking. Contrarily, knowledge sharing and joint learning emerged as SME behavioural patterns by which, in the same manufacturing districts, they developed and shared technology, innovation, know-how and technical capabilities. Within SMEs in particular, a sense of attachment and pride in their region facilitated commitment to sustainability integration. In some cases, buying firms offered limited support and the supplier development and integration activities were merged with normal business processes of supplier selection and evaluation. Building upon Liu et al. (2018), findings reveal that impactful and extensive supplier development practices for sustainability are dependent upon the engagement and the resources of both buying firms and suppliers. Suppliers are the most important external members affecting the performance of the chain. Given the supplier's direct and significant impact on the reputation of the buying firm (Wang & Dai, 2018), sustainability management must be extended throughout the chain. Suppliers must be better integrated and supported in the big initiatives of lead firms because SMEs, although arguably more committed, have insufficient financial and technological resources to make a material difference. This supports Zhang et al.'s (2017) findings that advancing supply chain social responsibility is a challenging task for SMEs because meeting CSR requirements might involve high time and knowledge costs. Senior management commitment and awareness are pivotal for change management towards sustainability in organizational settings (Diabat, Kannan, & Mathiyazhagan, 2014; Faisal, 2011; Lee & Klassen, 2008). Leadership and organizational commitment affect companies' abilities to integrate environmentally and socially responsible practices across supply networks. In companies where top management Fig. 1. Framework of sustainability implementation in luxury fashion supply chains. H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 659

commitment was visible, more proactive and innovative initiatives became apparent. ‘Leather 2’, ‘Leather 3’, and ‘Textiles 3’ showed that relational capabilities such as monitoring and supplier development could lead to information sharing and increased commitment. To reach an ideal scenario in luxury fashion, values must be shared with supply chain actors and supplier integration must be prioritised. Much work still needs to be done in order that companies could understand the foundations of sustainability and engage their suppliers in sustainability. Rather than dictating, knowledge sharing and collaboration must be encouraged. In agreement with Stindt (2017), we argue that SSCM remains a complex issue for which guidelines and frameworks on how to pursue and implement it within business practices are still needed. As shown in Fig. 1 below, stakeholder integration, innovativeness and commitment emerged as antecedents of the development of proactive SSCM strategies, and also helped to overcome the negative characteristics of the business environment that impedes sustainability implementation. 6. Conclusion This study examined the implementation of social and environmental sustainability across multiple tiers in Italian luxury fashion supply chains, where most added value is generated by SMEs. The findings highlight that the key motivation for luxury fashion companies to implement sustainability is to reduce negative impacts generated by their operations, but more radical approaches needed to extend sustainability across the entire network are currently missing. The lack of end-to-end supply chain visibility limits supply chain traceability, which is needed to transfer best practices across the entire network. The type of practice implemented depends on the tier in which the company is situated. Upstream suppliers focus on process innovation to become greener, while downstream suppliers focus on product innovation. The sense of belonging to a local community is evidenced in SMEs, which makes organizational commitment to sustainability a more natural act. Individual attempts are thus more likely to be transformed into collective improvement strategies throughout the chain. In contrast, large buying firms find it more challenging to integrate their suppliers, which makes extending sustainable practices and sharing resources further upstream difficult. Relational factors of trust, collaboration and integration facilitated horizontal transfer of innovations and resources throughout the chain. While this paper focuses on Italian luxury supply chains, many luxury supply chains now extend far beyond country of brand origin to incorporate lower labour cost countries. Possible avenues for future research could include a cross-country analysis, which could be an interesting angle to examine the extent to which sustainability is affected by choice of production country. Further research could also focus on different supply chains outside of Italy with different characteristics, such as luxury wine, spirits or jewellery. External conditions such as future uncertainty and market complexity could affect the development of proactive sustainability strategies. Therefore, further research is called for to investigate how companies could integrate, build and reconfigure internal and external capabilities to ensure supply chain sustainability in rapidly changing environments. Acknowledgements This research was conducted within the framework of the European Doctorate in Industrial Management (EDIM) funded by the European Commission’s Education, Audio-visual and Culture Executive Agency (EACEA) under Erasmus Mundus Action 1 programs. Consejo Social de la Universidad Politecnica de Madrid also financially supported the research. Appendix A. Theoretical perspectives utilized for data analysis NRBV Implications for SSCM Theoretical perspective Implications for SSCM Reference Environmental proactivity Going beyond legislative compliance Bowen, Cousins, Lamming, & Faruk, 2001; Managing environmental risks Gold, Seuring, & Beske, 2010 Product-based green supply Recycling initiatives that require collaboration with suppliers Bowen et al., 2001; Gold et al., 2010 Efforts with suppliers to reduce environmental impact Greening the supply process Strategic supplier selection Bowen et al., 2001; Gold et al., 2010 Supplier evaluation Monitoring and mentoring activities to improve environmental performance Environmental collaboration Direct involvement of company with suppliers in planning jointly for environmental management and solutions Vachon & Klassen, 2008 Path dependency and learning from pollution prevention Strategy complementarity through product stewardship, pollution prevention and sustainable development Miemczyk et al., 2016 Development, learning and sharing Antecedents of social sustainability Drivers and barriers of integrating social aspects into GSCM From various SSCM models & own development Enabling factors and resources supporting SSCM Organizational and operational structure From various SSCM models & own Knowledge and learning processes development Resources H. Karaosman et al. Journal of Business Research 117 (2020) 652–663 660