The devil wears Prada and Chanel and Calvin Klein.

The devil wears Prada and Chanel and Calvin Klein. Uncovering the patterns of luxury brand competition. Sarah Patrick 110079788 Supervisors Professor Jenni Romaniuk Professor Byron Sharp Dr Virginia Beal A thesis submitted for the degree of Masters by Research (Marketing) Thesis submission: February 2017 1

TABLE OF CONTENTS LIST OF FIGURES... 5 LIST OF TABLES... 6 ABSTRACT... 13 CHAPTER 1... 21 Introduction... 21 1.1 Luxury Brand Competition and Overview of Research... 21 1.2 Background: Luxury Brands... 22 1.3 The Competing Demands of Luxury Brand Management... 23 1.4 Research Approach... 25 1.5 Main Findings and Contributions... 31 CHAPTER 2... 33 Luxury Brand Owner Profiles... 33 2.1 Brand Competition... 33 2.2 Segmentation... 34 2.3 Brand User Profiles: Mass Markets... 35 2.4 Studies of Brand User Profiles... 38 2.5 Where Does Luxury Fit?... 41 2.6 Consumer Characteristics and Luxury Brands... 44 2.7 Hypotheses... 48 CHAPTER 3... 52 Method: Brand Owner Profiles... 52 3.1 Replication and Multiple Sets of Data... 52 3.2 Sample Profiles... 53 3.3 Motive Profiles... 60 CHAPTER 4... 64 Results Study One... 64 Do rival brands appeal to customers with distinct demographic profiles or motivations?... 64 4.1 Competition and Brand Owner Profiles... 64 4.2 Demographic Brand Owner Profile Analysis... 65 2

4.3 Consistency of Deviations... 82 4.4 Hypotheses: Demographic Profiles of Competing Brands... 83 4.5 Motivations for Buying Luxury... 85 4.6 Motive Profiles: Consistent Deviations... 102 4.7 Hypotheses: Motive profiles of competing brands owners... 104 4.8 Summary of Results... 105 CHAPTER 5... 107 Discussion: Brand Owner Profiles... 107 5.1 Overview and Main Findings... 107 CHAPTER 6... 112 Duplication of Purchase... 112 6.1 Brand Competition... 112 6.2 Duplication of Purchase Law: Background and Scope... 113 6.3 DoP Studies: Scope of Markets Tested and Relevance to Luxury... 117 6.4 Strategic Value of DoP Law... 117 6.5 Luxury Brand Competition... 122 6.5.1 DoP Law and Luxury... 123 6.6 Luxury Theory and Submarkets... 125 CHAPTER 7... 130 Method: Duplication of Purchase... 130 7.1 Brand Competition... 130 7.2 Duplication of Purchase Analysis... 135 CHAPTER 8... 144 Results Study Two... 144 How do luxury brands share owners?... 144 8.1 Brand Competition... 144 8.2 DoP: Fashion and Accessories Brands... 145 8.3 DoP Law: Goodness of Fit... 146 8.4 Consistency of DoP Fit... 150 8.5 Deviations: Fashion and Accessories Brands... 151 8.6 DoP Law: Goodness of Fit - Makeup and Skincare Brands... 152 8.7 Consistency of DoP Pattern... 160 8.8 Deviations from the DoP Pattern: Makeup and Skincare Brands... 160 8.9 Hypothesised Submarkets... 162 3

CHAPTER 9... 171 Discussion: Duplication of Purchase... 171 9.1 Overview... 171 9.2 The DoP Law... 171 9.3 Deviations in Customer Sharing... 172 CHAPTER 10... 175 Conclusions, Implications and Future Research... 175 10.1 Summary of Findings... 175 10.2 Implications: Marketing Theory... 177 10.3 Implications: Marketing Practice... 179 10.4 Limitations and Future Research... 181 Appendix 1... 184 Appendix 2... 194 Appendix 3... 309 Appendix 4... 317 References... 324 4

LIST OF FIGURES FIGURE 1: MOTIVES TESTED - ORGANISED BY THEME... 60 FIGURE 2: SUMMARISED MOTIVES FOR TESTING... 62 FIGURE 3: MOTIVES FOR TESTING BRANDS' OWNER PROFILES... 85 FIGURE 4: EXPECTED PATTERN OF SHARING IN LINE WITH DOP LAW... 136 5

LIST OF TABLES TABLE 1: SAMPLE PROFILES - 2014... 54 TABLE 2: SAMPLE PROFILES - 2015... 55 TABLE 3: SAMPLE OF BRANDS TESTED IN EACH CATEGORY... 57 TABLE 4: AGE PROFILES IN PERCENTAGES... 58 TABLE 5: CALCULATING THE DEVIATION FROM AVERAGE... 59 TABLE 6: TABLE OF DEVIATIONS FROM AVERAGE... 59 TABLE 7: DEVIATIONS OF +/- 10 PP FROM AVERAGE... 60 TABLE 8: DEVIATIONS OF +/- 10 PP FROM AVERAGE... 63 TABLE 9: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY BRAND OWNERS (HK 2014)... 66 TABLE 10: SUMMARY OF GENDER PROFILES - WATCH/JEWELLERY OWNERS... 68 TABLE 11: SUMMARY OF AGE PROFILES - WATCH/JEWELLERY OWNERS... 68 TABLE 12: SUMMARY OF INCOME PROFILES - WATCH/JEWELLERY OWNERS... 68 TABLE 13: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES BRAND OWNERS (IN 2014) 70 TABLE 14: SUMMARY OF GENDER PROFILES - FASHION/ACCESSORIES OWNERS... 73 TABLE 15: SUMMARY OF AGE PROFILES - FASHION/ACCESSORIES OWNERS... 73 TABLE 16: SUMMARY OF INCOME PROFILES - FASHION/ACCESSORIES OWNERS... 73 TABLE 17: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE BRAND OWNERS (CH 2014)... 76 TABLE 18: SUMMARY OF GENDER PROFILES - MAKEUP/SKINCARE OWNERS... 78 TABLE 19: SUMMARY OF AGE PROFILES - MAKEUP/SKINCARE OWNERS... 78 TABLE 20: SUMMARY OF INCOME PROFILES - MAKEUP/SKINCARE OWNERS... 78 TABLE 21: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (HK 2014)... 80 TABLE 22: SUMMARY OF GENDER PROFILES - SPIRITS OWNERS... 81 TABLE 23: SUMMARY OF AGE PROFILES - SPIRITS OWNERS... 81 TABLE 24: SUMMARY OF INCOME PROFILES - SPIRITS OWNERS... 81 TABLE 25: SUMMARY OF DEVIATIONS IN DEMOGRAPHIC PROFILES - ACROSS YEARS AND COUNTRIES... 82 TABLE 26: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (CH 2015)... 88 TABLE 27: SUMMARY OF MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS - 2014 AND 2015... 90 TABLE 28: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (HK 2014)... 92 TABLE 29: SUMMARY OF MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS - 2014 AND 2015... 94 TABLE 30: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (HK 2014)... 96 TABLE 31: SUMMARY OF MOTIVE PROFILES FOR MAKEUP/SKINCARE OWNERS - 2014 AND 2015... 98 6

TABLE 32: MOTIVE PROFILES OF SPIRITS OWNERS (HK 2014)... 100 TABLE 33: SUMMARY OF MOTIVE PROFILES OF SPIRITS OWNERS (2014 AND 2015)... 101 TABLE 34: DEVIATIONS IN BRAND MOTIVE PROFILES PERSISTING ACROSS BOTH YEARS AND MORE THAN ONE COUNTRY... 103 TABLE 35: STUDIES WHERE DOP LAW IS OBSERVED... 115 TABLE 36: DOP STUDIES: SHARING BETWEEN PRICE AND LUXURY TIER... 119 TABLE 37: FASHION/ACCESSORIES BRANDS TESTED... 133 TABLE 38: FASHION/ACCESSORIES BRANDS REPORTED IN MAIN THESIS FOR DOP ANALYSIS... 134 TABLE 39: MAKEUP/SKINCARE BRANDS TESTED IN EACH COUNTRY... 135 TABLE 40: BRANDS SORTED BY BRAND SIZE (NUMBER OF OWNERS)... 137 TABLE 41: CUSTOMER SHARING BETWEEN BRANDS - CONVERTED TO PERCENTAGES... 138 TABLE 42: DEVIATIONS FROM AVERAGE SHARING... 139 TABLE 43: HIGH-END AND ACCESSIBLE BRANDS IN TWO CATEGORIES... 140 TABLE 44: FUNCTIONALLY DIFFERENT BRANDS IN TWO CATEGORIES... 141 TABLE 45: FASHION/ACCESSORIES BRANDS OWNED BY THE SAME PARENT COMPANY... 142 TABLE 46: MAKEUP/SKINCARE BRANDS OWNED BY THE SAME PARENT COMPANY... 143 TABLE 47: SHARING AND BRAND SIZE - FASHION/ACCESSORIES BRANDS... 145 TABLE 48: DOP SHARING MATRIX - FASHION/ACCESSORIES BRANDS (US 2014, N=302)... 147 TABLE 49: DOP SHARING MATRIX: FASHION/ACCESSORIES BRANDS (CHINA 2014, N= 303). 148 TABLE 50: DOP SHARING MATRIX: FASHION/ACCESSORIES BRANDS (HONG KONG 2014, N= 306)... 149 TABLE 51: DOP SHARING MATRIX: FASHION/ACCESSORIES BRANDS (SINGAPORE 2014, N= 300)... 150 TABLE 52: PERCENTAGE OF BRANDS SHARING IN LINE WITH DOP PATTERN... 150 TABLE 53: PERCENTAGE OF DEVIATIONS (BIDIRECTIONAL) FROM EXPECTED SHARING... 151 TABLE 54: BIDIRECTIONAL DEVIATIONS IN SHARING CONSISTENT ACROSS COUNTRIES... 152 TABLE 55: SHARING ACROSS MAKEUP/SKINCARE BRANDS... 153 TABLE 56: DOP SHARING MATRIX: MAKEUP/SKINCARE BRANDS (US 2014, N= 302)... 154 TABLE 57: DOP SHARING MATRIX: MAKEUP/SKINCARE BRANDS (CHINA 2014, N= 303)... 155 TABLE 58: DOP SHARING MATRIX: MAKEUP/SKINCARE BRANDS (HONG KONG 2014, N= 306)... 157 TABLE 59: DOP SHARING MATRIX: MAKEUP/SKINCARE BRANDS (SINGAPORE 2014, N= 300)... 159 TABLE 60: PERCENTAGE OF PAIRS SHARING IN LINE WITH SIZE... 160 TABLE 61: PERCENTAGE OF DEVIATIONS FROM EXPECTED SHARING... 161 TABLE 62: BRAND-LEVEL OVERSHARING CONSISTENT ACROSS COUNTRIES... 161 TABLE 63: BRAND-LEVEL UNDERSHARING CONSISTENT ACROSS COUNTRIES (FASHION/ACCESSORIES)... 162 7

TABLE 64: BRAND-LEVEL UNDERSHARING CONSISTENT ACROSS COUNTRIES (MAKEUP AND SKINCARE)... 164 TABLE 65: BRANDS UNDERSHARING BASED ON FUNCTIONAL DIFFERENCES... 166 TABLE 66: FASHION AND ACCESSORIES BRANDS OWNED BY THE SAME PARENT COMPANY 168 TABLE 67: MAKEUP AND SKINCARE BRANDS OWNED BY THE SAME PARENT COMPANY... 169 TABLE 70: MAKEUP/SKINCARE BRANDS TESTED (2014) - IN ALPHABETICAL ORDER... 187 TABLE 71: SPIRITS BRANDS TESTED (2014) - IN ALPHABETICAL ORDER... 188 TABLE 72: WATCH/JEWELLERY BRANDS TESTED (2015) - IN ALPHABETICAL ORDER... 189 TABLE 73: FASHION/ACCESSORIES BRANDS TESTED (2015) - IN ALPHABETICAL ORDER... 190 TABLE 74: MAKEUP/SKINCARE BRANDS TESTED (2015) - IN ALPHABETICAL ORDER... 192 TABLE 75: SPIRITS BRANDS TESTED (2015) - IN ALPHABETICAL ORDER... 193 TABLE 76: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (US 2014) - ORDERED BY BRAND SIZE... 194 TABLE 77: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (US 2014) - ORDERED BY BRAND SIZE... 195 TABLE 78: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (US 2014) - ORDERED BY BRAND SIZE... 196 TABLE 79: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (US 2014) - ORDERED BY BRAND SIZE... 197 TABLE 80: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (US 2014) - ORDERED BY BRAND SIZE... 199 TABLE 81: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (US 2014) - ORDERED BY BRAND SIZE... 200 TABLE 82: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (US 2015) - ORDERED BY BRAND SIZE... 201 TABLE 83: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (US 2015) - ORDERED BY BRAND SIZE... 202 TABLE 84: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES BRAND OWNERS (US 2015) ORDERED BY BRAND SIZE... 203 TABLE 85: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES BRAND OWNERS (US 2015) ORDERED BY BRAND SIZE... 205 TABLE 86: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE BRAND OWNERS (US 2015) ORDERED BY BRAND SIZE... 207 TABLE 87: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILE OF MAKEUP/SKINCARE OWNERS (US 2015) - ORDERED BY BRAND SIZE... 208 TABLE 88: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS BRAND OWNERS (US 2015) - ORDERED BY BRAND SIZE... 209 TABLE 89: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS BRAND OWNERS (US 2015) - ORDERED BY BRAND SIZE... 210 8

TABLE 90: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 211 TABLE 91: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 213 TABLE 92: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 215 TABLE 93: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 217 TABLE 94: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 219 TABLE 95: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 221 TABLE 96: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 223 TABLE 97: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNERS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 224 TABLE 98: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY (HONG KONG 2015) - ORDERED BY BRAND SIZE... 225 TABLE 99: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (HONG KONG 2015) - ORDERED BY BRAND SIZE... 227 TABLE 100: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (HONG KONG 2015) - ORDERED BY BRAND SIZE... 229 TABLE 101: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (HONG KONG 2015) ORDERED BY BRAND SIZE... 231 TABLE 102: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (HONG KONG 2015) - ORDERED BY BRAND SIZE... 234 TABLE 103: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (HONG KONG 2015) - ORDERED BY BRAND SIZE... 236 TABLE 104: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (HONG KONG 2015): ORDERED BY BRAND SIZE... 238 TABLE 105: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNERS (HONG KONG 2015) - ORDERED BY BRAND SIZE... 239 TABLE 106: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 240 TABLE 107: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 242 TABLE 108: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 244 TABLE 109: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES BRANDS (CHINA 2014) - ORDERED BY BRAND SIZE... 246 9

TABLE 110: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (CHINA 2014) ORDERED BY BRAND SIZE... 248 TABLE 111: BRAND-LEVEL DEVIATIONS IN MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 250 TABLE 112: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 252 TABLE 113: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNERS (CHINA 2014) - ORDERED BY BRAND SIZE... 253 TABLE 114: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 254 TABLE 115: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 255 TABLE 116: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 257 TABLE 117: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES BRANDS (CHINA 2015) - ORDERED BY BRAND SIZE... 259 TABLE 118: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 262 TABLE 119: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 264 TABLE 120: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 266 TABLE 121: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNERS (CHINA 2015) - ORDERED BY BRAND SIZE... 267 TABLE 122: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 268 TABLE 123: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 270 TABLE 124: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 272 TABLE 125: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 274 TABLE 126: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (SINGAPORE 2014) - ORDERED BY SIZE... 276 TABLE 127: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 278 TABLE 128: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (SINGAPORE 2015) - ORDERED BY BRAND SIZE... 280 TABLE 129: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (SINGAPORE 2015) - ORDERED BY SIZE... 281 10

TABLE 130: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (SINGAPORE 2015) ORDERED BY BRAND SIZE... 283 TABLE 131: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (SINGAPORE 2015) - ORDERED BY BRAND SIZE... 285 TABLE 132: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (SINGAPORE 2015) - ORDERED BY BRAND SIZE... 288 TABLE 133: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (SINGAPORE 2015) - ORDERED BY SIZE... 289 TABLE 134 BRAND SIZE: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (SINGAPORE 2015) - ORDERED BY BRAND SIZE... 291 TABLE 135: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNERS (SINGAPORE 2015) - ORDERED BY BRAND SIZE... 292 TABLE 136: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF WATCH/JEWELLERY OWNERS (INDONESIA 2014) - ORDERED BY BRAND SIZE... 293 TABLE 137: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF WATCH/JEWELLERY OWNERS (INDONESIA 2014) - ORDERED BY BRAND SIZE... 295 TABLE 138: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (INDONESIA 2014) - BRANDS ORDERED BY SIZE... 297 TABLE 139: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (INDONESIA 2014) BRANDS ORDERED BY SIZE... 299 TABLE 140: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF MAKEUP/SKINCARE OWNERS (INDONESIA 2014) - BRANDS ORDERED BY SIZE... 301 TABLE 141: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF MAKEUP/SKINCARE OWNERS (INDONESIA 2014) - ORDERED BY BRAND SIZE... 303 TABLE 142: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF SPIRITS OWNERS (INDONESIA 2014) BRANDS ORDERED BY SIZE... 305 TABLE 143: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF SPIRITS OWNES (INDONESIA 2014) - ORDERED BY BRAND SIZE... 306 TABLE 144: BRAND-LEVEL DEVIATIONS: DEMOGRAPHIC PROFILES OF FASHION/ACCESSORIES OWNERS (TAIWAN 2014) - ORDERED BY BRAND SIZE... 307 TABLE 145: BRAND-LEVEL DEVIATIONS: MOTIVE PROFILES OF FASHION/ACCESSORIES OWNERS (TAIWAN 2014) - ORDERED BY BRAND SIZE... 308 TABLE 146: PERCENTAGE POINT DEVIATIONS IN SHARING FOR FASHION/ACCESSORIES BRANDS (US 2014) - ORDERED BY BRAND SIZE... 309 TABLE 147: PERCENTAGE POINT DEVIATIONS IN SHARING FOR FASHION/ACCESSORIES BRANDS (SINGAPORE 2014) - ORDERED BY BRAND SIZE... 311 TABLE 148: PERCENTAGE POINT DEVIATIONS FOR FASHION/ACCESSORIES BRANDS (HONG KONG 2014) - ORDERED BY BRAND SIZE... 313 TABLE 149: PERCENTAGE POINT DEVIATIONS IN SHARING FOR FASHION/ACCESSORIES BRANDS (CHINA 2014) - ORDERED BY BRAND SIZE... 315 11

TABLE 150: BIDIRECTIONAL OVERSHARING BETWEEN FASHION BRANDS (US)... 317 TABLE 151: BIDIRECTIONAL OVERSHARING BETWEEN FASHION BRANDS (CHINA)... 317 TABLE 152: BIDIRECTIONAL UNDERSHARING BETWEEN FASHION BRANDS (CHINA)... 318 TABLE 153: BIDIRECTIONAL OVERSHARING BETWEEN FASHION BRANDS (HONG KONG)... 319 TABLE 154: BIDIRECTIONAL UNDERSHARING BETWEEN FASHION BRANDS (HONG KONG).. 320 TABLE 155: BIDIRECTIONAL OVERSHARING BETWEEN FASHION BRANDS (SINGAPORE)... 320 TABLE 156: BIDIRECTIONAL UNDERSHARING BETWEEN FASHION BRANDS (SINGAPORE)... 320 TABLE 157: BIDIRECTIONAL OVERSHARING BETWEEN MAKEUP BRANDS (US)... 322 TABLE 158: BIDIRECTIONAL UNDERSHARING BETWEEN MAKEUP BRANDS (US)... 322 TABLE 159: BIDIRECTIONAL UNDERSHARING BETWEEN MAKEUP BRANDS (CHINA)... 322 TABLE 160: BIDIRECTIONAL OVERSHARING BETWEEN MAKEUP BRANDS (HONG KONG)... 322 TABLE 161: BIDIRECTIONAL UNDERSHARING BETWEEN MAKEUP BRANDS (HONG KONG)... 323 TABLE 162: BIDIRECTIONAL OVERSHARING BETWEEN MAKEUP BRANDS (SINGAPORE)... 323 TABLE 163: BIDIRECTIONAL UNDERSHARING BETWEEN MAKEUP BRANDS (SINGAPORE)... 323 12

ABSTRACT Luxury Brand Competition A brand s survival ultimately rests on its ability to sell products. With other brands vying for a share of consumer spending, brands must be strategic about how they compete. Empirical research can help by uncovering regularities in buying behaviour and providing norms to guide decision-making. Empirical studies in mass markets have revealed a lot about underlying patterns of competition. However competition in luxury markets has received comparatively less empirical attention. Various luxury theorists have shown interest in this area but there are differing opinions among experts about how brands compete. For example, some theorists suggest brands compete based on their level of luxury or luxury tier (e.g. De Barnier, Falcy, & Valette-Florence, 2012; Silverstein & Fiske, 2003). Others argue luxury buyers may be classified based on the qualities they seek and therefore different brands are thought to appeal to these customer types (e.g. De Barnier et al., 2012; Kapferer, 1998). There are also theoretical reasons for suspecting the law-like patterns of competition in mass markets may not apply to luxury goods (e.g. Keller, 2009; Sicard, 2013). The two studies conducted in this thesis empirically test some widely held beliefs about luxury brand competition using a framework of established empirical generalisations. Study One examines whether luxury brands compete by appealing to distinct segments of consumers. Study Two empirically tests whether brand size influences competition and if particular luxury brands compete differently based on their luxury tier, functional differences or ownership by the same luxury conglomerate. 13

Objective of Thesis This research examines how luxury brands compete. Specifically, it explores whether luxury brands compete in separate silos where segments of customers are drawn to each brand s unique offering. This thesis tests whether directly competing brands attract customer bases with distinct demographic profiles and motives for buying the category. The research documents the degree to which brands customer bases overlap. High overlap would indicate brands do not compete by servicing specific segments of customers. Data This thesis examines competition between more than 250 different brands, spanning four luxury categories: 1) fashion and accessories, 2) makeup and skincare, 3) watches and jewellery, and 4) luxury spirits. This thesis uses data collected in 2014 and 2015 by Affluential, a panel database representative of the top 25 percent of income earners in the countries tested. The data looks exclusively at current luxury brand ownership - encompassing brands bought for oneself and received as gifts. The 2014 dataset is comprised of samples from six countries (US, China, Singapore, Hong Kong, Indonesia and Taiwan) with a total sample exceeding 1600 respondents. The 2015 dataset is comprised of over 1200 participants from four countries (US, China, Singapore and Hong Kong). Rather than analyse as a single dataset, this research uses these different conditions to explore the generalisability of patterns of competition. Main Findings Study One: Brand Owner Profiles Across four luxury categories in six countries, two consecutive years and over 200 brands, demographic owner profiles of directly competing brands are generally similar. Depending on the category, 81-95 percent of demographic profiles are within 10 percentage points (PP) of the average profile and overall a mean absolute deviation (MAD) of 5 PP from the average profile is found. Most differences in demographic profiles are explained by gender 14

o On average 28 percent of deviations in demographic profiles (across both years and categories) are explained by gender compared to an average of 11 percent each for age and income) o Overall MADs for gender range from 5-9 PP depending on the category (compared to 4-6 PP for age and 4-5 PP for income) Demographic differences between brand owner profiles observed in this study are slightly higher than deviations reported in past research. For example, a study of user profiles in 42 product categories (including chocolate bars, credit cards, private health insurance and package holidays in the UK), reported competing brands demographic user profiles differ by an average of 3 PP (Kennedy, Ehrenberg, & Long, 2000). Competing brands customer bases have similar motives for buying luxury brands (based on nine different motivations for buying luxury brands). Depending on the category, 94 97 percent of motive profiles are within 10 PP of the average profile. Brands rarely attract customer bases with unique motivations for buying luxury goods (overall MADs range from 2-7PP depending on the category) While differences between brands customer motive profiles are small, they are more pronounced than findings from similar studies in mass markets. For example, Kennedy and Ehrenberg (2001a) report 1 percent of cases deviate 10 PP from the average profile (this thesis reports levels of 3-6 percent). Study Two: Duplication of Purchase Across more than 50 brands in two luxury categories, each in four countries, predictable customer sharing in line with brand size is the norm; with larger brands drawing more from the customer bases of smaller brands (Agostini, 1961; Goodhardt & Ehrenberg, 1969). 92 percent of competing fashion/accessories brands share customers in line with brand size (based on over 1380 cases) 91 percent of competing makeup/skincare brands share customers in accordance with brand size (based on over 330 cases) 15

These findings are consistent with the Duplication of Purchase law (DoP law) of customer sharing between competing brands documented in mass markets (e.g. Bass, 1974; Ehrenberg, Uncles, & Goodhardt, 2004; Uncles & Ellis, 1989). The findings also support empirical evidence of brands sharing customers in line with size in luxury fashion and Champagne competition (Romaniuk & Sharp, 2016) and prestige skincare brand competition (Romaniuk & Winchester, 2007). Consistent with past DoP empirical research, the present findings show some cases of functionally different brands competing less intensely with one another (e.g. sweetened vs. unsweetened breakfast cereal brands as noted by Hammond, Ehrenberg, and Goodhardt (1996). Studies One and Two come together to show (in the categories tested), brands mainly compete in line with size in a fairly unsegmented luxury market. Some brands appeal more than average to males or females, and certain brands compete less directly based on differences in their product ranges etc. However luxury brands rarely compete in separate silos by attracting distinct customer bases. Implications and Contributions This thesis extends the scope of brand user profile and duplication of purchase research to include luxury brand competition in multiple new product categories. The findings challenge popular themes in luxury theory about brand segmentation and competition. This research contributes to existing scientific knowledge by demonstrating the luxury category is not a boundary condition to the patterns of brand competition observed in mass markets. The research has two major implications for luxury brand marketing. First, directly competing brands generally have similar customer bases. This means in the categories tested, brands rarely attract customer bases with unique motives for buying luxury or distinct demographic profiles (though certain brands do appeal more to males or females). Second, luxury brands compete primarily on the basis of size. For brand managers, this research reveals norms for luxury 16

brand competition. This information can improve strategic decision-making such as deciding whether to enter a new product category, and help brands set realistic expectations in novel settings (e.g. entering new markets, in response to acquisition by a luxury brand conglomerate etc.). Limitations and Future Research One of the main challenges associated with luxury brand research is securing representative samples of brand owners, particularly for cross-brand ownership. To maintain the representativeness of the samples and reliability of the results, some smaller (often more expensive/exclusive) luxury brands are omitted from the analysis. Another data-related limitation is the use of combined product categories (e.g. the makeup/skincare category includes makeup, fragrance, skincare and nail polish brand ownership). More defined categories would enable clearer comparisons between brands. There are opportunities to further test the generalisability of this research with more product categories and different countries or regions (e.g. Europe). Future inquiry could build on the present research to test whether the Double Jeopardy pattern (McPhee, 1963) applies to luxury brand ownership. This research would test another established empirical generalisation in the luxury market and concurrently examine loyalty to luxury brands - another area of interest among luxury theorists (e.g. Kapferer & Bastien, 2009a; Nueno & Quelch, 1998). Future research could also examine competition between luxury and non-luxury brands, building on evidence of luxury/prestige submarkets within the general market (e.g. Colombo, Ehrenberg, & Sabavala, 2000; Romaniuk & Winchester, 2007). Such research could examine whether luxury and non-luxury brands compete directly, and if so, whether this is more prevalent in certain product categories and price tiers. 17

DECLARATION I declare that: this thesis presents work carried out by myself and does not incorporate without acknowledgment any material previously submitted for a degree or diploma in any university; to the best of my knowledge it does not contain any materials previously published or written by another person except where due reference is made in the text; and all substantive contributions by others to the work presented, including jointly authored publications, is clearly acknowledged. Sarah Patrick...15 February 2017 18

ACKNOWLEDGEMENTS Thank you Jenni for passing on your knowledge, for providing direction and opportunities for me to learn and gain experience. Thank you for constantly challenging me. Thank you Byron for the motivation, your big picture thinking and always encouraging me to think about the relevance of my research. Thank you Gin for the ongoing support, for checking in and always being there to answer my questions. Alicia, thank you for all your assistance and attention to detail. I would have run out of time if it were not for you! Thank you to my family and friends, especially Mum, Dad, Alex, Alex A., Rupert, Chlöe, Kathleen, Kate, Nick, Morgs and Simon. A special thank you to Ella, Kirsten, Pippi, Adam, James Kelly and Nicole. This research received support from an Australian Government Research Training Program Scholarship, and from the various sponsors of the Ehrenberg- Bass Institute s research. 19

20

CHAPTER 1 Introduction Chapter One introduces the topic of luxury brand competition the focus of this thesis. This leads into a discussion of luxury goods and the conflicting demands of luxury brand management, which sets the scene for the studies in this thesis. Next the research objectives and method are outlined. The chapter closes with an overview of the findings and research contributions. 1.1 Luxury Brand Competition and Overview of Research Studies of brand competition reveal law-like patterns of buying behaviour underpin brand competition in mass markets. Our understanding of the scope of these regularities has developed over time through testing their robustness under different conditions (e.g. across different countries, product categories, over time etc.). While knowledge of the conditions under which these patterns persist in mass markets is vast, brand competition in the luxury market has received less empirical attention. Various luxury theorists have contributed to this topic, providing a range of perspectives about how luxury brands may compete. For example, some theorists envision distinct types of luxury customers who are drawn to different groups of brands (e.g. Kapferer, 1998), others suggest brands compete based on their level of luxury/luxury tier (e.g. De Barnier et al., 2012), and others question whether luxury brands compete at all (e.g. Roux, 1991). However, most of these ideas rest on the assumption that the luxury category is different to non-luxury (e.g. Kapferer & Bastien, 2012; Keller, 2009; Sicard, 2013). 21

Studies One and Two bridge this gap by applying two established empirical generalisations to the luxury category. Study One tests the pattern of similar brand user profiles between competing brands. Study Two tests the fit of the Duplication of Purchase law (DoP law) to luxury brand competition. Together, the results show rival luxury brands attract quite similar customer bases, with most brands competing in line with brand size within a broad luxury submarket. For brand managers, this research provides norms for luxury brand competition to guide brand strategy. 1.2 Background: Luxury Brands For centuries, luxury goods have been used to distinguish the social elites from the masses (Veblen, 1899). The demand for luxury goods continues today, with a global market value in excess of 850 billion (D'Arpizio, Levato, Zito, & De Montgolfier, 2014). Despite turbulence in global markets following the events of September 11, the financial crisis of 2009 and the Chinese stock market crisis in 2015 (Allen, 2015), between 1995 and 2015, the compound annual growth rate of the personal luxury goods market increased by 6 percent (Bain & Company, 2016). Emerging economies have also seen strong growth in the sales of luxury goods, and this demand is expected to continue well into the next decade (Euromonitor International, 2015; Kravets & Sandikci, 2014). Despite growth in the global luxury industry (Bain & Company, 2016) much remains unknown about how luxury brands should be optimally marketed. The luxury category has attracted a lot of speculation about what constitutes effective marketing, and the extent to which this should depart from the marketing of non-luxury products (e.g. Atwal & Williams, 2009; Tynan, McKechnie, & Chhuon, 2010). The present research addresses a central part of this question by examining luxury brand competition. The concept of luxury itself has received considerable attention from marketing theorists. A large part of this interest has been around how to best conceptualise luxury for marketing purposes (Berthon, Pitt, Parent, & Berthon, 2009; Vickers & 22

Renand, 2003). For instance, in defining luxury some theorists draw distinctions between terms such as luxury, deluxe and prestige, while others use them interchangeably (Cailleux, Mignot, & Kapferer, 2009; Kapferer & Bastien, 2009b). A consensus is yet to be reached about a comprehensive definition of luxury, however the present research will not contribute to this interesting debate. For the purposes of this research, the Atwal and Williams (2009) conceptualisation of luxury goods as items associated with exclusivity, status and quality (p. 338) will be used. 1.3 The Competing Demands of Luxury Brand Management Perhaps one reason why the conversation around defining luxury ensues is that luxury goods are considered different from non-luxury products. Luxury brands face challenges that are thought to require marketing approaches that are distinct from other products and services (Bastien & Kapferer, 2013). These approaches are based on logical assumptions from theorists and researchers about how luxury brands compete, and how luxury customers buy brands. For example, luxury theorists Professor Jean-Noël Kapferer and Professor Vincent Bastien, both from the prestigious HEC Paris, claim: in order to enter the luxury market, to build a successful luxury brand and to make it remain a luxury brand, one has to forget the classical marketing rules (2009b, p. 312). Kapferer and Bastien are not alone in this belief, and similar opinions are often expressed by luxury and marketing writers (e.g. Keller, 2009; Sicard, 2013). For example, while a luxury brand is expected to maintain high brand awareness and desirability among the masses (Phau & Prendergast, 2000), it is widely accepted that restricting ownership through high prices, limited distribution and waiting lists helps to maintain the appeal of luxury brands (Bastien & Kapferer, 2013; Nueno & Quelch, 1998; Štrach & Everett, 2006). 23

This perspective prevails in industry as well. Chairman of the LVMH Group (which oversees 70 luxury labels) (LVMH, 2015), Bernard Arnault highlights the contradictory demands of luxury marketing, stating the ideal luxury brand is: timeless, modern, fast growing, and highly profitable...there are fewer than ten star brands in the luxury world. It is very hard to balance all four characteristics at once If you have a star brand, then basically you can be sure you have mastered a paradox (Wetlaufer, 2001, p. 121). Arnault s comments highlight the complexity of successfully managing a luxury brand. As academics have noted, balancing these demands is even more difficult for brands aiming to grow sales, enter a new product category or establish a presence in new markets (e.g. Beverland, 2004). This research uses the framework of well-established patterns of mass market brand competition and buying behaviour to test whether there is empirical support for these popular luxury theories and marketing approaches. 1.3.1 Conglomerates Influence Competition The luxury market is dominated by luxury brand conglomerates (e.g. Richemont, Kering and LVMH), which continue to acquire once independent and familyowned luxury labels (Ijaouane & Kapferer, 2012). For example in Hong Kong, eight of the top ten luxury apparel and footwear brands for 2016 (ranked by retail value) were owned by luxury conglomerates, rather than independently owned (Euromonitor International, 2017). Formerly privately owned brands undergoing this transition are subject to intense pressure from shareholders and executives to maintain or enhance sales without surrendering their cachet (Ijaouane & Kapferer, 2012; Nueno & Quelch, 1998; Roberts, 2011; Štrach & Everett, 2006). Being part of a conglomerate provides brands with additional resources and marketing opportunities. With the backing of conglomerates, many luxury brands face decisions about how to allocate their marketing budgets. The 24

conflicting demands of luxury marketing makes this a challenging task. Therefore, empirical insights about how to most effectively market luxury goods are especially valuable. Added to new marketing opportunities, previously independent brands gain access to centralised resources such as production sites, new distribution channels and real estate (Doran, 2013; Shipilov & Godart, 2015). These resources can be leveraged by brands to enhance market share, which in turn affects the competitive structure of markets (Doran, 2013; Farris, Olver, & de Kluyver, 1989; Weiss, 1968). 1.4 Research Approach Discoveries in marketing science reveal law-like relationships, which are the recurrence of empirical findings under varied conditions (Bass, 1995; Ehrenberg, 1969). Law-like relationships provide benchmarks for testing under different experimental conditions (Uncles & Kwok, 2013) and contribute to greater understanding of the appropriateness of different marketing strategies (Dawes, 2008). Two important empirical patterns are the Duplication of Purchase Law (DoP) (Agostini, 1961; Goodhardt & Ehrenberg, 1969) and the similarity of Brand User Profiles between directly competing brands (Kennedy & Ehrenberg, 2000a; Uncles, Kennedy, Nenycz-Thiel, Singh, & Kwok, 2012). The patterns of buyer behaviour and brand competition observed in massmarkets shape the research approach in this thesis. The luxury sector is an important market to extend marketing science research, first, as it has received only initial empirical attention and second; there are theoretical reasons why the law-like patterns observed in other contexts may not hold for luxury. There are a number of reasons for this (covered in the chapters to follow). Uncovering whether the same regularities of competition apply to luxury brands provides an opportunity to test the scope of these law-like patterns. 25

This research investigates whether there are generalisable patterns of competition underpinning the luxury market and if these differ from competition in mass markets. The research is conducted in two parts. Study One tests whether luxury brands compete by appealing to segments of consumers seeking a brand s unique product offering. Another possibility is that brands compete head-on in a mainly unsegmented luxury submarket. Study Two empirically tests if brand size influences luxury brand competition and whether certain brands compete more intensely with one another. This is important because it allows luxury brands to identify which competitors they are most likely to lose customers to, or steal customers from. Together, Studies One and Two test popular theories of luxury brand competition whilst contributing to preliminary empirical research in luxury brand competition. Differentiated replication studies test the robustness of scientific findings and reveal the conditions where these patterns may not apply (Hubbard & Armstrong, 1994; Uncles, 2011). This research uses a multiple sets of data (MSoD) approach to further test the robustness of the findings under varied conditions (i.e. different countries, categories and over time). Testing across these different sets of data allows systematic patterns to emerge and the practical and academic implications to be assessed (Dawes, 2008). The next section describes the laws for testing and their application to luxury brands. 1.4.1 Luxury Brand Owner Profiles A brand s user profile is its composition of users or its customer base (Ehrenberg & Goodhardt, 1970). In mass markets, evidence shows there is little difference between competing brands demographic and psychographic customer profiles (Kennedy & Ehrenberg, 2001b; Uncles et al., 2012). Deviations are rare and predictable. For example, some deviations reflect functional differences between competing brands such as sweetened and unsweetened breakfast cereals, which appeal to children more than adults (e.g. Hammond et al., 1996; Kennedy & Ehrenberg, 2001b). 26

In contrast to mass market goods, competing luxury brands are generally agreed to appeal to different types of owners (e.g. Kapferer, 1998; Kuksov & Xie, 2012; Wiedmann, Hennigs, & Siebels, 2009). This perspective is based on the assumption that consumers are drawn to brands based on the different characteristics they seek in luxury goods (Kapferer, 1998; Wiedmann et al., 2009). For example, experts such as Kapferer (1998) and Kuksov and Xie (2012) explain the brands customers choose can reflect how much they value exclusivity or membership in a select group. The difference has also been considered from a demographic perspective, such that younger customers with lower incomes (relative to older and wealthier customers) are thought to be drawn to more affordable brands (Kapferer, 1998). Study One uses brand user profile analysis to test for the presence of distinct demographic and motive profiles between the customer bases of competing brands. To identify any systematic patterns across different conditions, Study One compares user profiles across two consecutive years, six countries (US, China, Singapore, Taiwan, Indonesia and Hong Kong) and four product categories 1) watches/jewellery, 2) fashion/accessories, 3) makeup/skincare and 4) spirits. Due to the conflicting evidence offered by brand user profile research and luxury theory and research, competing hypotheses are used to guide Study One. Presenting competing hypotheses is appropriate where there is more than one possible hypothesis for the research (Armstrong, Brodie, & Parsons, 2001). Informed by the findings from empirical studies of mass markets in addition to luxury theory and research, the first part of Study One hypothesises: H1a. Directly competing luxury fashion/accessories brands will have owner bases with similar demographic profiles. H1b. Directly competing luxury fashion/accessories brands will have owner bases with distinctive demographic profiles. 27

H2a. Directly competing luxury makeup/skincare brands will have owner bases with similar demographic profiles. H2b. Directly competing luxury makeup/skincare brands will have owner bases with distinctive demographic profiles. H3a. Directly competing luxury watch/jewellery brands will have owner bases with similar demographic profiles. H3b. Directly competing luxury watch/jewellery brands will have owner bases with distinctive demographic profiles. H4a. Directly competing luxury spirits brands will have owner bases with similar demographic profiles. H4b. Directly competing luxury spirits brands will have owner bases with distinctive demographic profiles. The second aspect of Study One tests for brand-level segmentation in the motive profiles of luxury brand owners. This leads to the following competing hypotheses: H5a. The owners of rival luxury fashion/accessories brands will be similar in their luxury values and motives. H5b. The owners of rival luxury fashion/accessories brands will differ in their luxury values and motives. 28

H6a. The owners of rival luxury makeup/skincare brands will be similar in their luxury values and motives. H6b. The owners of rival luxury makeup/skincare brands will differ similar in their luxury values and motives. H7a. The owners of rival luxury watch/jewellery brands will be similar in their luxury values and motives. H7b. The owners of rival luxury watch/jewellery brands will differ in their luxury values and motives. H8a. The owners of rival luxury spirits brands will be similar in their luxury values and motives. H8b. The owners of rival luxury spirits brands will differ in their luxury values and motives. 1.4.2 Duplication of Purchase Study Two empirically tests whether there is an underlying pattern to luxury brand competition and if so, whether certain luxury brands compete more intensely with one another. Studies show customers exhibit polygamous loyalty to competing brands (Ehrenberg, Goodhardt, & Barwise, 1990; Martin, 1973; Wright, Sharp, & Sharp, 1998). The distribution of these repeat purchases results in brands sharing customers in accordance with their market share, whereby larger brands share fewer of their customers with the smaller competing brands (Goodhardt, Ehrenberg, & Chatfield, 1984). The Duplication of Purchase pattern (DoP) is observed across multiple buying situations including consumer packaged goods, subscription markets (e.g. 29

banking) and durable categories (e.g. cars and phone handsets) (Colombo et al., 2000; Dawes, 2014; Ehrenberg, 2000; Faulkner, Truong, & Romaniuk, 2014). Some preliminary testing by Sharp and Romaniuk (2016a) suggests DoP law also applies to luxury fashion and Champagne brands, whereby smaller brands share more of their owners with the larger brands. Study Two extends this work by applying the DoP analysis to brand ownership in two highly competitive categories: fashion/accessories and makeup/skincare brands each in four countries (US, China, Singapore and Hong Kong). Study Two addresses the following question: RQ1. Does ownership across competing luxury brands follow a DoP pattern? In addition, Study Two tests for deviations from the DoP pattern of sharing. Deviations can highlight submarkets of heightened competition between certain brands and can reveal information about the competitive structure of markets (Ehrenberg, 1975). For example, brands with functional similarities may compete more intensely/overshare customers with each other. Based on DoP research in mass markets and popular themes of luxury brand competition, Study Two addresses the following question: RQ2: Is there evidence of brands oversharing/undersharing, and is this consistent across international markets? On the basis of existing theory and empirical findings the following hypotheses are also tested: H9: Brands occupying a different luxury tier will undershare customers. H10: Functionally different brands will undershare customers. H11: Brands owned by a common parent company will overshare customers. 30

It is important for brand managers to make marketing decisions on the basis of reliable and stable findings rather than exceptional circumstances. For deviations to be reliable and managerially relevant, they need to persist across different conditions (e.g. over time or countries etc.). Identifying patterns of undersharing and excess sharing across different international markets provides a means to assess their stability and generalisability. 1.5 Main Findings and Contributions Study One shows competing brands attract customer bases with quite similar demographic profiles. Overall, 83 percent of deviations are within 10 percentage points (PP) of the typical owner profile. This similarity is reinforced by the overall mean absolute deviation (MAD) of 5 PP for demographic profiles across two years, six countries and four categories. This finding is broadly consistent with studies of demographic brand user profiles in mass market categories (e.g. Hammond et al., 1996; Uncles et al., 2012). Study One also shows owners of competing luxury brands have similar motive profiles for buying luxury brands, with 95 percent of deviations within 10 PP of the average profile and an overall MAD of 5 PP across nine different motivations. Rather than measuring users general attitudes and attitudes towards specific brands (as in previous research e.g. Kennedy and Ehrenberg (2001b)), the present study extends brand user profile research to include motivations for buying from the luxury category. Study Two reveals luxury brands (in two categories each in four countries) generally share customers in accordance with brand size (overall, 92 percent of brand pairs for fashion/accessories brands and 91 percent for makeup/skincare brand pairs show normal sharing). While brands do not appear to compete exactly as luxury theorists envision, there is evidence certain brands compete less intensely (than their size would suggest) due to price-related and functional differences. This result aligns with studies that noted lowered competition between functionally different brands in other categories (e.g. sporty and 31

luxury car brands and diet vs. non-diet flavoured milk brands (Colombo et al., 1999; Sharp, Sharp, & Redford, 2003)). Together, the results from Study One and Two highlight competing luxury brands do not generally attract unique customer bases and brand size primarily determines customer sharing. The findings broadly support arguments that luxury marketing should seek to generate widespread appeal, rather than narrowly target groups of luxury consumers (e.g. Dubois & Paternault, 1995; Kapferer & Bastien, 2009b; Keller, 2009). Digressing from Kapferer and Bastien s (2009b) perspective, drawing on a representative sample of luxury brand owners, the present findings imply luxury brands should treat category buyers as potential buyers of their brand. This stream of research clarifies how luxury brands compete. The findings provide norms for luxury marketers, which are particularly useful in setting expectations about competition. For example, knowing brand size will largely drive competition allows a brand manager to anticipate which competitors they are likely to gain and lose most of their customers (and therefore which competitors to watch more carefully). The findings also provide benchmarks for competition in novel contexts such as entering new international markets. The following chapter introduces the first study of this thesis. 32