207 ase fourm of international Busnes Sues (2007 38 5) to Brat Al gn eaaed 0012586 $300 A social institutional approach to identifying generation cohorts in China with a comparison with American consumers Kineta H Hung’, Flora Fang Gu* and Chi Kin (Bennett) Yim" "school of Business, University of Hong Kong, ‘ong Kong; 7Hong Kong Poytechnic University, ‘Hong Kong Correspondence: Kineta H Hung, School of Business, University of Hong Kong, Pokfulam Road, Hong Kong. Tel: + 852 2857 8248; Fax: + 852 2858 5614; E-mail: kineta@business.hku.hie Received: 6 September 2005 Revised: 27 April 2006 ‘Accepted: 3 October 2006 Online publication date: 24 May 2007 Abstract Identifying distinctive target segments is a fundamental challenge faced by international marketers. This paper describes an approach to understanding ‘consumer market structures in an important international market, China, and to segmenting Chinese consumers by integrating insights from generational ‘cohort and social institutional theories. We conduct two empirical studies to verify how China’s recent momentous ideological events could give rise to and affect the life experiences of different generation cohorts in the country. The results of Study 1 support the conceptual framework and establish three distinct cohorts in China: Red Guards, Moder Realists, and Global Materialist. Study 2 builds on the findings of Study 1: it links cohort differences to differences in consumer values (materialism) and choice behaviors (foreign vs local brands), and then compares them with parallel consumers in the United States. This paper outlines and tests an approach to segmentation that can help, international fiers identify distinct segments in and design effective marketing strategies for China, Journal of International Business Studies (2007) 38, 836-853. {010.105 7ipalgravejivs8400288 Keywords: socal institutional theory; generation cohort; segmentation; China Introduction A conceptual understanding of how consumers behave in different cultures represents the first step in designing effective international marketing strategies (Johansson and Thorelli, 1985). Such under- standing is crucial, because it enables managers to interpret what they observe in foreign markets and postulate how consumers might respond to firm strategies (Zou and Cavusgil, 2002; Walters and Samiee, 2003). Because today’s market is increasingly char- acterized by hypercompetition and well-defined brand positioning, the identification of distinctive target segments offers a fund: mental challenge to international marketing (Zou and Cavusgil, 2002; Batra and Tse, 2003). International segmentation studies have identified multiple categories that might serve as bases for segmentation, The availability of published secondary data (e.g., Eutomonitor, United Nations, World Bank) makes country-level variables such as ‘economic indicators and political characteristics the most popular segmentation bases (Steenkamp and Hofstede, 2002). These basesSoci iatutlonal porch to Meatihea gentratlon cohorts tnt nna to % enable MNCs to identify clusters of target countries, although they generally do not account for intra- cultural variations (Au, 1999). Other studies use disaggregated information such as demographics (e.g, age, gender, income) and psychographics (eg., VALS) to provide managers with more detailed understanding of market structures. Among disag- gregated bases, a hybrid approach - cohort segmen- tation - takes advantage both of the stability that age segmentation offers (Steenkamp and Hofstede, 2002) and of the insights into consumer motiva- tions that value segmentation offers. Although cohort segmentation has attracted research interest in the US market, it has seldom been applied in an international marketing context, ‘A generational cohort refers to a consumer segment that uses a consumer's coming-of-age year as a proxy to postulate his or her value priorities developed through life experiences during his or her formative years, which may persist throughout that person's lifetime. For example, the baby boomers in the United States, who came of age during an era of general prosperity but lost confidence in government and other institutions, became the ‘me’ generation (Meredith and Schewe, 2002). The increased interests in their well-being prompted baby boomers to redefine the notion of middle age by investing heavily in anti-aging products, cosmetic plastic surgery, fitness programs and sports cars: this behavior contrasts with that of middle-aged citizens in previous generations, who accepted the prevalent notion of middle age rather than actively resisting the aging process (Weiss, 2002). If middle age can mean something quite different from one generation of Americans to another, who grew up in environments that are relatively stable, the need to identity distinct cohorts in transitional economies, in which the environ- ment is by definition volatile, must be acute. In a country such as China, which has undergone considerable changes during the past 50 years, people are likely to pursue values and behaviors that reflect the value priorities of their coming- of-age era. If so, how effective is cohort as a basis for segmenting consumers across generations in Chi- na? Can this approach help identify segments unique to China? What socioeconomic processes give rise to and characterize these segments? And, finally, what insights will the segmentation ratio- nale and accompanying processes offer to intera- tional marketing managers regarding segments’ value orientations and consumption patterns? The current study attempts to address these questions by investigating the impact of changing institutions on cohorts of Chinese consumers. We draw on social institutional theory - well estab- lished for firm-level studies in China (North, 199 Walder, 1995; Boisot and Child, 1999) — and supplement it with historic generational cohort theory and a collective memory explanation to develop an integrated framework applicable to consumer studies. We then identify three cohorts of consumers in China, and verify in two surveys the differences in their values and consumption behaviors. We also improve on previous attempts to Investigate cohorts in China using approximations (eg, 1951-1960, 1961-1970) to define cohort membership, which may not tie in with soci historical events in the country (eg., Egri and Ralston, 2004), Understanding consumer behavior in China Among the pioneering attempts to understand how the 1.3 biition Chinese consumers behave, Schmitt (1997) highlights age group differences as a fundamental characteristic of the Chinese market. Whereas some age group differences can be attributed to intergenerational gaps, which constitute an integral part of social development, others point to drastic changes in China's social institutions. Belk and Zhou (2002) explore the onset of China’s revolutions and. political campaigns during the past 50 years, and postulate that these events help explain how Chinese consumers behave today. Fong (2004) notes that major changes in the country’s institutions have socialized parents and their children with conflicting ideologies, and have presented them with different opportunities and constraints. Egri and Ralston (2004) examine the values among managers and professionals. in China and the United States, and find that the Chinese/US cohorts who grew up during China's closed-door policy evince the least similar values. These observations suggest that China's eventful recent past, at the societal level, has ‘mapped its people's life experiences at the indi dual level, a relationship that deserves further investigation. Another reason why we chose China as our research context Is that China has undergone multiple distinct and momentous revolutions and political campaigns during the past SO years. Few countries have undertaken such rapid social exper mentations for successive age groups in the way [paral af iterations! Bonne Sue7 seuttnl penchant chart ie China has (Rosen, 2000; Belk and Zhou, 2002). Therefore consumer demand in China, a socialist market economy, may be more strongly affected by institutional decree than it would be in a full market economy. Moreover, China is the largest developing economy, and attracts many MNCs that could benefit from the findings of our research. In summary, this research outlines and_ tests a segmentation approach that may enable MNCs to use readily available information (e.g., census data about country demographics) to postulate information that is usually difficult to collect in transitional economies (e.g., motivations to con- sume), A comparison with US consumers in the latter part of this research highlights the effects of, changing institutions on consumers in China, and contributes to our understanding of international segmentation. Social institutional theory Social institutional theory pertains to the way Institutions emerge, function, and evolve to affect the behaviors of persons and firms in a society. Institutions include: formal organizations, such as governments; legitimizing bodies, such as public opinion, professions, religions, and ideologies (Scott, 1987); ‘invisible’ rules of operation, such as culture and norms (DiMaggio and Powell, 1983); and the process of habituation, according to which, actions that are repeated become cast into a pattern (DiMaggio and Powell, 1983). These institutions represent the structure and mechanisms that func- tion as the ‘rules of the game’ in a society (North, 1999). Institutions typically change incrementally. How- ever, cataclysmic events such as World War II bring forth breaking points and discontinuous institu- tional changes (Inglehart, 1977). More recentiy, when formerly planned economies adapted to the free market system, many institutions underwent rastic reform (Lau et al., 2002) to render firms more ‘marketized’ (Nee, 1992) and profitable (Walder, 1995). These disruptive social changes have given Nobel laureates in economics (North, 1990), leading sociologists (Walder, 1995), and pioneering organization theorists (Boisot_ and Child, 1999) a context from which to derive implications about firms in China. Historic generation cohort theory ‘The concept of generation and its influence on cohorts of people has been discussed widely in sociology, anthropology, and social psychology inte go af - —_ esearch, Inglehart (1977, 1990) and his colleagues (Abramson and Inglehart, 1995; Inglehart and Norris, 2003) have demonstrated, using longitudi- nal data from 40 countries, that once a society embarks on industrialization, fundamental changes in values occur, including a greater eriphasis on belonging, self-expression, and quality of life (or “postmodern values’) in lieu of economic and physical security (or ‘materialist values’), Central to Inglehart’s thesis is that value changes take place gradually ata younger generation, socialized within a particular socioeconomic system during its pre- adult years, replaces an older generation as the adult population in a society. Rogler (2002) expands Inglehart’s thesis into an integrated theory by postulating that the emer- gence of distinct generations and the formation of their collective identity can be attributed to similar life experiences. Rogler’s work adopts a historical perspective with three underlying theses. First, cataclysmic events (ie., wars, revolutions, natural or man-made disasters) that shake the foundation of the existing social order ‘give birth’ to a new generation. Second, because a person forms his or her value system during his or her pre-adult years, but experiences a crystallization of the process by the time he or she reaches adulthood (McCrae et al., 2002), cataclysmic events impose a stronger Imprint on a coming-of-age group than on other age groups living through the same historical period. Third, consolidated orientations, including a person's values and goals, as supported by his or her peers, persist throughout his or her life, In short, cataclysmic events enforce a stronger, longer- lasting value imprint on the coming-of-age group than on other age groups who live through the same historic period. Collective memory explanation An important theoretical tenet that underlies institutional theory and historical” generation cohort theory is collective memory. A person's memory encompasses shared memory with the collectivities of which the person is a part (Belk, 1988, 1991; Lipsitz, 1990). Because memories are often shared and jointly enacted, as well as forced by rituals and artifacts (Halbwachs, 1950}, societies remember collectively, and the rituals are maintained among peers (Connerton, 1989; Holbrook and Schindler, 1989; Belk, 1991). ‘Thus intergenerational differences in values and behaviors may be driven by different events and life experiences. ipo oer Bana SuierSocial inattutlonal approach to Id on cohorts Social institutional changes and generation cohorts in China We identify the Cultural Revolution (including the Sent-Down Movement) (1966-1979), the Economic Reform (1980-1991), and the globalization of China (1992- ) as major events that have given birth to recent generation cohorts in China. The Cultural Revolution was the most disruptive of these events, and the destruction it caused has been widely publicized. The Economic Reform and globalization of China, though less disruptive at a personal level, introduced new social orders, ideol- ‘gies, and institutional drivers that, according to Rogler (2002), cultivate coming-of-age groups’ values. We categorize persons into cohorts on the basis of the specific event they experienced during their coming-of-age years. Cohort 1: Red Guards ‘The Red Guards came of age during the Cultural Revolution (1966-1979), when China was closed to the outside world and commercial activities came to a halt to give way to social/political class struggles. During the early years of the movement, many students were inspired to become Red Guards. They sacrificed their youth, material com- forts, and family life for the welfare of the country and the Communist Party. During the climax of the movement, nearly 17 million young intellectuals, who were led to believe that they were In need of re-education by peasants, relocated to rural areas, where the living conditions were harsh. Although the Cultural Revolution appeared to be a move- ment engineered to indoctrinate young people with heroism and idealism, and spur them to improve China’s conditions through revolutionary efforts (Rosen, 2000), it was believed to be at least in part political and aimed at defusing potential social unrest due to massive unemployment (Yao and Lu, 1993), ‘The Red Guards have often been considered the ‘Jost’ generation of China (Chen, 1999), given the harsh, bitter, and violent struggles they endured during their formative years, which did not prepare them for the changing world. In turn, this cohort may be less open to change (Egri and Ralston, 2004). Thus the social/political ideologies the Red Guards acquired during their coming-of-age years probably continued to condition their attitudes and consumption behavior, even after China entered the open economy era. We propose that the Red Guards will be pessimistic in their outlook, and remain frugal and conservative in their con- sumption attitudes and behavior. Cohort 2: Modern Realists ‘The Modern Realists came of age during the Economic Reform (1980-1991), and grew up in a society whose economy was transforming from a centrally planned system that stressed egalitarian- ism to a market system that stressed competition, entrepreneurship, and individual 2ccomplishment. Unlike the Red Guards, who were subjected to social/political class struggles during their coming- of-age years, the Modern Realists were encouraged by the government to pursue a relatively comfor- table life - or Xiaokang (Lu, 2000). The political will behind the Economic Reform was supported by Institutional changes, such as establishing special ‘economic zones, encouraging overseas Chinese to invest in the motherland, deregulating the market and state-owned firms, and maintaining robust economic growth (China Statistical Yearbook, 2002). Many members of the cohort took advan: tage of these institutional changes by moving to special economic zones, accumulating personal wealth, and achieved Xiaokang. A survey conducted between 1989 and 1992 showed that attitudes toward monetary wealth among youths in China were highly favorable (Tang and Parish, 2000). Another survey showed that respondents born between 1971 and 1975 were more entrepreneurial than were the generations that preceded them (Egti and Ralston, 2004). We suggest that these institu- tional changes encouraged Modern Realists to become novelty seekers. The Economic Reform also ‘reinvented! the advertising industry, which grew in leaps and bounds (China Advertising Association, 2002). Thus Modern Realists have been cultivated by advertising and marketing institutions to become materialistic. Cohort 3: Global Materialists Global Materialists were born after the Cultural Revolution, and came of age in 1992 or thereafter, when China was integrating into the global com- munity. During this period, the GDP continued to grow by approximately 10% annually, and China became Asia's largest advertising market (Xu, 2000). ‘Thus Global Materialists were exposed to increased levels of affluence in society and consumerist messages in the media. The Internet also helped this cohort connect with the outside world (Hung and Li, 2007), so its members developed tastes in fashion and products similar to those of youths Tourra of eration Baines Studies8) i around the world (Amett, 2002). Meanwhile, over- seas Chinese, especially those in Hong Kong, represented the main source of FDI in China throughout the 1980s. By 1992, FDI soared from USS4.37 billion a year earlier to USS11.01 billion as MNCS responded positively to the government's policy to globalize China, as Deng Xiaopeng indicated on his 1992 tour to southern China. FDI then increased fivefold over the next decade. This wave of investors included firms from Europe, Asia, and the US Zhang and Huang, 2003), and ensured the presence of global brands in China. ‘The younger segment of this cohort was also born after the implementation of the one-child policy 2000). These ‘little emperors’ grew up in an environment in which many were cared for by six doting family members each (parents and grand- parents), who tended to indulge them, We hypothesize that these institutional changes have cultivated in the Global Materialists even higher levels of novelty seeking and materialistic values, as well as a stronger orientation toward conspicuous consumption, including a preference for foreign brands Overview We postulate that institutional changes that arise from historic, cataclysmic events may explain the significant differences across cohorts of people in China. Thus the Red Guards may be negative in their outlook and frugal in their consumption behavior, the Modern Realists may be interested in monetary wealth and consumption of novel things, and the Global Materialists may be even more open than the Modern Realists and desire foreign goods. In the following, we report two empirical studies. Study 1 establishes China’s three distinct cohorts by examining differences in values (novelty seeking and pessimism) and consumption of products and services (shopping as leisure ot burden; use of traditional or novelty products). Novelty seeking is similar to consumer innovative- ness (e.g., Steenkamp et al., 1999), but it encom- passes a broader conceptualization, which goes beyond diffusion and product adoption. Pessimism triggers coping mechanisms that affect a person's consumption patterns (Zullow, 1991). Study 2 builds on the findings of Study 1, and links cohort differences to materialism and brand choice beha- vior (ie,, foreign ¥s local brands), Ia Table 1 we provide ‘a summary of the cohorts, including institutional changes, life experiences, alleged values, and consumption implications. Feral of amar Sana Sieg Institutional approach to Hlentfylng generation cohorts kine Hinge! oe Study 1: Cohort identification and Investigation Differences in personal value orientations Novelty seeking ‘The approach temperament indicated by novelty seeking captures both the entrepreneurial spirit to explore new ventures and the consumption need to try alternative and exciting things. We postulate that the novelty-seeking value will be embraced differentially by the cohorts. Modern Realists came of age when the predominant ideology was to lead a relatively comfortable life. Global Materialists came of age when society was becoming even more open and affluent. Therefore Modern Realists and Global Materialists should develop stronger novelty-seeking values thart Red Guards, as a result of their fife experiences. Pessimism The value of pessimism is characterized by the negativity and withdrawal that often accompanies economic and other forms of deprivation (Wang, 1994). A pessimist generally attends to the negative aspects of a given situation, and uses denial and disengagement when under stress (Gibson and Sanbonmatsu, 2004), Whereas Red Guards endured a harsh and deprived upbringing during the Cultural Revolution, only to be disappointed by their lack of relevant skills in the open economy, Modern Realists and Global Materialists were well provided for during their coming-of-age years, and enjoyed more opportunities as the economy improved, We therefore posit that Red Guards will be more pessimistic than the other two cohorts. Hypothesis 1: Modern Realists and Global Mate- rialists are (a) more novelty seeking and (b) less pessimistic than are Red Guards. Differences in shopping attitudes and product usage People tend to hold attitudes and behaviors con- sistent with their values and higher-order goals (Huffman et al., 2000). Prior research shows that a person with high novelty-seeking values is inter- ested in new products, and responsive to advertis- ing and promotions (Goldberg et al., 2003; Hung et al., 2005). We extend this line of research to examine cohort differences in shopping attitudes and product usage.Soci ltitlonal approch to entihig generton cohorts Kt ual 3K Table 1 Year of bith, institutional changes, fe experiences, values, and consumption: implications across cohorts Global materialist: ‘Coming of age 1992 and ater Year of birth 1974-1984 ‘Age at Study 1 (2002) 18-28 ‘Age at Study 2 (2000) 18-26 Social institutional changes Globalization '* Ideology/Social norms Global integration ‘@ Ad spend (million CNY) 67-4794 © FDI Gin billion USD) 11.914 >53 © GDP (annual growth) 9.79% Lite experiences ‘© Increased exposure to mass ‘media and the Internet ‘© Affected by global ‘consumer culture 1@ Increased exposure to MNCS and foreign brands Modem realists Red guards 1980-1991 1966-1979 1965-1973 1951-1964 29-37 38-51 27-35 36-49 Economic Reform Cultural Revolution Xigokang Socialist, Maoist ideology 1935 0.1 (1979) 0.16-+4.37 a 10.30% ra Venture to coastal cities and @ special economic zones 10 @ take advantage of the market economy Became Red Guards Relocated to rural areas in support ofthe ‘sent-down? movement Became entrepreneurial, @ Gave up education accumulated personal Strong bonding among wealth pees Novelty seeking? © Pessimistict Materiatistic® © Non-materialistic® ‘Shopping as leisure* © Shopping as burden Use novelty products" © Use traditional products* Use foreign brands? (© Use local brands? Aleged values ‘© Novelty seeking (highen* © Mateiaistic higher Implications for consumption © Shopping as lesure (bigher # Use novelty products (highen* © Use foreign brands _ (highen” Sua study 2. Consumers hold different attitudes toward shop- ping: some consider it a mundane task and dislike the activity, whereas others consider it a conspic- uous, enjoyable activity (Gardner and Sheppard, 1989). For these latter leisure shoppers, the shop- ping environment allows them to stay up to date with new products, enjoy social encounters, and experience sensory stimulation (Lunt and Living- stone, 1992), We postulate that Modern Realists and Global Materialists, whom we purport to be novelty seekers, will consider shopping a leisure activity, because they value the stimuli that shop- ping provides, whereas Red Guards, whom we Purport to be pessimists, will be likely to consider shopping a burden. Hypothesis 2: Modern Realists and Global Mate- rialists are (a) more likely to view shopping as leisure and (b) less likely to view shopping as a burden than are Red Guards. In terms of product usage, consumers have differ- ential preferences. We differentiate specifically between traditional products that have a long history in China (e.g,, tea) and novelty products that have become popular in China only recently (c.g,, coffee). Few products marketed by MNCs in China are traditional products; rather, novelty products often are introduced to Chinese consu- mers by MNCs, We postulate that the novelty- seeking Modern Realists and Global Materialists will have a higher propensity to use novelty products, whereas the pes: Red Guards will have a higher propensity to use traditional pro- ducts. Hypothesis 3: Modern Realists and Global Mate- rialists are (a) more likely to use novelty products: and (b) less likely to use traditional products than, are Red Guards Socioeconomic control variables Previous research on segmentation suggests that socioeconomic variables are useful for differentiat- ing segments of consumers. We include the follow- Ing covariates in our analyses of variance: city tier, a Tourra of reratona Bases Saassect maton sepreech to enihingsenerton shorts tes tg a! _ proxy for regional economic development; family income and education, proxies for a_ person's socioeconomic status; and gender, which indicates men’s and women’s different category and shop- ping needs. Regional differences represent an important socioeconomic factor in China (Walters and Samiee, 2003). Specifically, coastal cities have attracted a higher proportion of FDI and have benefited more from economic and globalization reforms than have inland cities, We divide the cities we survey into two tiers on the basis of per capita GDP to account for regional differences. In cross- tabulation tests, we run the analyses per city tier to provide insights into cohort effects. Methods and results A major research firm collected the data for Study 1 A total of 2970 respondents between the ages of 18 and 51 years participated. During face-to-face inter- views, respondents rated items pertaining to their values (novelty seeking, pessimism) and shopping attitudes (leisure, burden) on five-point scales (1=strongly disagree, S=strongly agree), and indi- cated, using yes/no responses, whether they used or consumed on a regular basis specific traditional products (tea, rice wine) or novelty products (cellular phones, coffee). The measures of aovelty seeking (Steenkamp and Baumgartner, 1992) and pessimism (Lunt and Livingstone, 1991) indicate the cohorts’ dispositions. The measures of shopping attitudes were adapted from Lunt and Livingstone (1992), We selected tea and rice wine to represent traditional products. Cellular phones and coffee represent products introduced to China more recently. We tested a replication of two products in each category to enhance the reliability of our findings, The survey was conducted in 2002. On the basis of the respondents’ age at the time of the survey, we classify them into cohorts in accordance with the definitions provided in Table 1: 30.3% of the respondents are Global Materialists, 22.2% are Modern Realists, and 47.5% are Ret Guards. The data cover 10 cities, which we categorize into tier 1 cities (Beijing, Shanghai, Guangzhou, mean per capita GDP=CNY 36,987) and tier 2 cities (Cheng- du, Daffan, Nanjing, Shenyang, Wuhan, Xi'an, and Zhengzhou, mean per capita GDP=CNY 19,396) (China Statistical Yearbook, 2002). We also include the covariates of gender, family income (four levels: CNY 800 oF less, CNY 801-1500, CNY 1501-3000, CNY 3001 or more), and education (four levels’ ‘amal 9 rteotinal Busy Satie junior high or below, high school, college, uni- versity and postgraduate study). We performed a principal component factor analysis (EFA) with varimax rotation. Using the eigenvalue (> 1) and Scree test as criteria, we obtain four theorized latent constructs with no cross- loadings greater than 0.30. The four factors explain 52.28% of total variance. We delete one item from pessimism that has 2 low factor loading (=0.44). We then conduct a CFA that yields a satisfactory model fit (77=485.348, d.f.=84, P<0.001, CFI=0.943, GFI-0.978, AGF1=0.969, RMSEA=0.040). All con- struct reliabilities (0.75 for novelty seeking, 0.60 for pessimism, 0.71 for shopping as leisure, 0.67 for shopping as burden) satisfy the standard desirable value of 0.60 for exploratory research (Bagozzi and Yi, 1988). We then assess the discriminant validity of our measures by testing the confidence intervals of the phi estimates of our constructs: none of them Includes unity. We also conduct a series of CFAS, as recommended by Bagozzi et al. (1991). The results suggest that, for every pair of factors in the measurement model, a two-factor model fits the data significantly better than a one-factor model, demonstrating satisfactory discriminant validity (Anderson, 1987; Bollen, 1989). We provide details of the measures in Table 2. Cohorts, values (H1), and shopping attitudes (H2) We hypothesize that the Global Materialists and Modem Realists are novelty seekers (Ha), whereas the Red Guards are pessimists (H1b). The results of an ANCOVA (with city tier, gender, income, and education as covariates) show significant overall cohort effects for Global Materialists, Modern Realists, and Red Guards (novelty seeking: F(6,2824)=46.59, P<0.001, M's=2i.74, 20.72, 19.86, respectively; pessimism: F(6,2824)=3.96, P<0.05, M's=4.86, 5.06, 5.12, respectively). The Global Materialists (P<0.001) and Modern Realists (P<0.001) score significantly higher mean values ‘on novelty seeking than do the Red Guards, whereas the Red Guards score significantly higher mean values on pessimism than do the Global Materialists (P<0.01), though not the Modem Realists (P>0.05). These findings provide full support for Hla and partial support for HIb (Table 3). Meanwhile, the Global Materialists score significantly higher mean values on novelty seek- ing (P<0.001) compared with the Modern Realists, but there are no significant differences between these two cohorts on pessimism (P>0.05). The covariates exert differential effects on the reported% Social institutional 10 Mentlfying generation cohorts kines Ming el wa Table 2 Measures in Study 1 Factor loading Foctor 1: Novelty seking (15.486 of variance explained, construct reliabilty=0.75) 1 ike to try challenging things that | have never dane before 0.730 IF1 am interested in something, | will do it in spite of risks 0.706 Like to try things new and different 0.692 | am very interested in new technology reported in the media 0.633 | am interested in knowledge and information in various fields 0.599 Factor 2: Pessimism (9.196 of variance explained, construct reliobilty=0.60) { don't like many aspects of myself 0.828 In general, | believe that life is full of distress and sadness 0.798 Factor 3: Shopping as leisure (14.396 of variance explained, construct reliabilty~0.71) ‘My friends always accompany me to shop 0.748 enjoy sharing shopping experiences with others 0.748 {ike to go shopping with friends 0.742 Purchasing gives me great enjoyment 0.654 Factor 4: Shopping as burden (13.496 of variance explained, construct relablity=0.67) 1am reluctant to accompany others to shop 0.729 ‘walk away when other people stat talking about shopping 0.704 1 don’t ie people who like to shop 0.695 '1go shopping only when others ask me help buy something 0.657 Model fit indices 2.485.348 (B4 A), Pe<0.001, CF1=0.943, GF1=0.978, AGF [Note No erossioading coetcent is greater than 0.30, values: education, income, and gender exert sig- nificant effects on both novelty seeking (P<0.001) and pessimism (P-<0.05), whereas city tier exerts, significant effects only on pessimism (P<0.01). We also hypothesize that Global Materialists and Modern Realists view shopping as a leisure activity (H2a), whereas the Red Guards view it as a burden. (Hb). The results of an ANCOVA (with city tier, gender, income, education as covariates) again show significant differences (leisure: F(6,2824)= 70,33, P<0.001, M’s=7.33, 6.72, 6.32, respectively; burden: F(6,2824)=17.20, P<0.001, M’s=5.63, 5.78, 6.10, respectively). Compared with Red Guards, Global Materialists (P-<0.001) and Modern Realists (P<0.001) are significantly more likely to view shopping as a ieisure activity, whereas the Red Guards are significantly more likely to view shop- ping as a burden than either Global Materialists (P<0.001) or Modern Realists (P<0:001). Mean- while, Global Materialists are more likely than Modern Realists to view shopping as a leisure activity (P<0.001), though there is no significant difference between the two cohorts with regard to shopping as a burden (P>0.05). City tier exerts 969, RMSEA=0.035 significant effects on shopping as a_ burden (P<0.001). Gender exerts significant effects on both shopping as leisure (P<0.001) and shopping as burden (P-<0.001). Neither income nor educa- tion exerts significant effects (P>0.08). Thus the findings provide support for H2a and H2b (Table 3). Cohorts and product usage (H3) H3 suggests that Global Materialists and Modern Realists are more likely to be novelty product users (H13a), and Red Guards are more likely to be traditional product users. We conduct a series of cross-tabulations (per city tier) to assess the cohorts! product usage. We find that, in tier 1 cities, novelty products are used by a significantly higher propor- tion of Global Materialists and Modern Realists (cell phone: 77=34.16, P<0.001, M's=67, 59.9, 45.4%; 7.19, P<0.001, M's=23.5, 15.8, 11.8%), whereas traditional products are used by a signifi- cantly higher proportion of Red Guards (teé 2.51, P<0.01, M’s=44.8, 49.3, $7.9%; rice wine: 77=22.20, P<0.001, M's=5.7, 7.9, 16.4%). ‘We report the findings for tier 2 cities in Table 3: Toro Tara] BO SO% Table 3 Values and consumption attitudes and behaviors (Study 1) Social inttutlonal approach to Mentlfying generation cohorts nts Hung al ue Global materiatists (18-28) Modem realist (29-37) Red guard (38-51) P value ANCOVA tests (covariates=city tir, family income, education, gender) Values Novelty seeking 22.23 20.93 19.44 0.001 Pessimism 470 499 5.25 0,001 ‘Consumption attitudes Shopping as leisure 7.28 676 632 0.001 Shopping as burden 5.68 5.80 6.09 0.001 Chisquare tests (96 within cohort) Novelty (nontraditional) product usage Cellular phone City tert 67.0 59.9 45.4 0,001 City tier2 46.2 470 286 0.001 Coffee ity tert 235 158 ns 0.001 City tier? 141 101 67 0.001 Traditional product usage Tea City tert 448 493 379 001 City tier2 39.0 482 52.9 0,001 Rice wine City tier! 87 79 16.4 0.001 City tier2 126 18.8 2.0 0.001 their cohort effects are very similar, in support of both H3a and H3b. Study 2: Materialism, foreign brand consumption, and cross-national comparison Study 1 provides empirical support for the sugges- tion that consumers who have been exposed to the same historic cataclysmic events and altering social Institutional conditions in China exhibit similar values, shopping attitudes, and product usage. Study 2 builds on these findings and examines cohort differences in materialism and foreign brand usage that should be of interest to international marketers. Study 2 also provides a comparison of findings on consumption patterns between Chi- nese and US consumers to provide additional evidence that the effects we identify are due to cohort, rather than life cycle, differences. Materialism Materialism refers to @ consumption-based orienta- tion toward happiness (Inglehart, 1981). At the zhest level, possessions assume a central place in ‘a materialist’ life, and provide him or her with the greatest sources Of satisfaction and dissatisfaction (Belk, 1984). Materialism also may act as a sub- Fearne oT ronal Ba SI stitute for a lost sense of community (Easterlin and Crimmins, 1991; Achenreiner, 1997), and enable an sitive person to participate in an imagined perfect life (Campbell, 1987). Thus a person high in materialistic values is possessive, desires a higher income, and places less emphasis on interpersonal relationships (Richins and Dawson, 1992). For our study, whereas Red Guards acquired socialist/Maoist Ideology and endured the Cultural Revolution when they were coming of age, Modern Realists and Global Materialists came of age when advertising spending, FDI, and GDP were growing rapidly and the political climate was more open. Although Study 1 shows that Global Materialists and Modern Realists are more prone to novelty seeking, a closer look suggests that the cohorts differ in the way they realize theit novelty-seeking values: Mod- ern Realists are more intent on entrepreneurship (Egri and Ralston, 2004), whereas Global Materialists| focus on consumption as a result of their greater exposure to media advertising, the Internet, and global consumer culture, Because of the cohorts’ Gifferential experiences, we suggest that Global Materialists will be more materialistic than either Modern Realists or Red Guards. In contrast, Red Guards enjoy strong bonds with thelr peers, and willSocial institutlonal approach to identifying generation cohorts ‘have little need for an imagined substitute commu- nity, In short, we suggest that Global Materialists ‘will be the most materialistic and Red Guards the least materialistic cohorts. Hypothesis 4: Global Materialists are (a) more materialistic than Modern Realists, (b) who in turn are more materialistic than Red Guards. Foreign brand usage across cohorts Foreign brands usually promote global/Western images when marketed in transitional economies (Alden et al., 1999), a strategy that enhances the desirability of foreign brands among local consu- mers, who associate foreignness with quality and prestige because of the brands’ higher prices and relative scarcity (Batra et al., 2000; Steenkamp ¢¢ al, 2003). If Global Materialists are the most materi- alistic cohort, they should value the quality and prestige associated with foreign brands more than do Modern Realists or Red Guards. In addition, Global Materialists perceive themselves as members of a global community, as a result of their greater exposure to MNC products and global consumer culture, Prior research shows that congruity between a product and self-image facilitates pro- duct purchases (Sirgy et al., 2000), so we argue that Global Materialists, who relate more strongly to foreign brands, will have a higher likelihood of using them than the other cohorts, Hypothesis 5: Global Materialists are more likely to be foreign brand users than either (a) Modern Realists or (b) Red Guards, Foreign brands, product image, and self-congr In addition to categorizing a product into a local or foreign brand, prior research has classified products as utilitarian, symbolic, or routine on the basis of the primary purpose it serves and the consumer's differential involvement (Vaughn, 1986). Thus we can differentiate foreign brands into purpose- related subcategories. A Louis Vuitton fashion acces- sory, which is a foreign brand (for Chinese consumers) and a value-expressive product, would be likely to offer twofold symbolic values. In contrast, Colgate toothpaste also represents a for- eign brand, but serves a low-involvement purpose. The toothpaste probably does not contain high- involvement symbolic values similar to those attributed to the fashion accessory, even though both are of foreign make. We have suggested that the congruity between, product origin and a cohort member's self-image facilitates a Global Materialist’s purchase of foreign brands. Research on attitude formation and pur- chase intentions also shows that a value-expressive product (e.g., fashionable clothing) has greater relevance for consumers who are concerned about their self-image (Shavitt, 1990; Johar and Sirgy, 1991), We posit that the product make/consumer self-image congruity we consider with HS will be more pronounced if the focal product is also value expressive Hypothesis 6: Global Materialists are more likely to be users of foreign-made, value-expressive products than either (a) Modern Realists or (b) Red Guards. Cross-national comparisons We purport to identify in this research three generation cohorts in China, each of which has differential value orientations and consumption behaviors. Because cohort membership is deter- mined by the coming-of-age year proxy, and people are likely to pass through different life cycle stages as they age, we compare Chinese and American consumers’ consumption patterns, as well as th life cycle stages, to ascertain whether the effects we identify are indeed due to the cohorts’ differential life experiences. As Egri and Ralston (2004) show, the consumers who grew up during China’s closed- door policy hold the least similar values. Thus we propose that not only will Chinese and Americans Of similar ages (and life cycle stages) hold different values as a result of cross-national differences, but the extent of value difference will vary in accor- dance with the respective dominant ideology in these countries during the cohorts’ coming of age. People in different life cycle stages have different consumption needs, affected in part by their family responsibilities (Schaninger and Danko, 1993). If the effects we identify are due to life cycle differences, consecutive Chinese cohorts should display consumption patterns similar to consecu- tive American cohorts, because Chinese and Amer- ican consumers pass through similar life cycle changes as they age. Alternatively, if the effects we identify are due to cohort differences as a result of people's life experiences, consecutive cohorts of ‘Chinese consumers will display different consump- tion patterns from those of their American counter- parts, even if they are in similar life cycle stages, because of the differences in their fe experiences.HE sot ttn certin cabot ue ‘These cross-national differences should be greatest among people who grew up during China's closed door policy (i.e., Red Guards in China, born 1951- 1964, and baby boomers in the United States, born 1946-1964),' when the dominant ideology in each country was least similar The baby boomers grew up during a time of general prosperity in the US, and display a con- sumption pattern that reflects their relatively affluent upbringing (Meredith and Schewe, 2002). Unlike previous generations of 50-year-old Amer- cans who increased their savings when they reached this age bracket, the boomers have reversed the trend by continuing to spend and taking out oans (Russell, 1999; Francese, 2001). This habit is widespread because the boomers, who grew up during the free-spirited 1960s, have never learned to save (Meredith and Schewe, 2002). This and similar observations led Weiss (2002: 31) to con- ‘clude that ‘the past income and spending trends we have observed among 55- to 64-year-olds will likely be of little help in predicting what Boomers will do when they move into that age category.’ In short, baby boomers are easy spenders, whereas Red Guards are careful spenders. Home and car ownership across cohorts in the United States and China Spending on homes and cars are indicative of the consumption profile of Americans at different life cycle stages (Schaninger and Danko, 1993). As well, the residential pattern among Chinese consumers changes over life cycle stages (Chen, 2005). We outline American cohorts’ spending on homes and cars, and then develop hypotheses to compare this spending with levels of home and car ownership actoss cohorts in China. We review publications that report consolidated American data from the Bureau of Census and Bureau of Labor Statistics (Russell, 1999; Russell and Mitchell, 1999), These publications show that baby boomers are more likely than younger Americans to be home owners (Russell and Mitchell, 1999). Using an index of 100 to represent average spending on homes, the indexed spending is 18 for Americans under 25 years of age, 78 for those aged 25-34 years, 106 for age group 35-44, and 120 for those 45-54 aged years (Russell and Mitchell, 1999). Thus home-ownership spending may be described as an upward-sloping curve across cohorts in the US. Similarly, car spending indicates an upward-sloping curve across American cohorts (indexed spending: under 25, 53; age 25-34, 82; age 35-44, 112; age 45- Journal at rteratinal Bases Sadie eta ua stb 54, 147) (Russell and Mitchell, 1999). The higher spending by baby boomers on these purchases falls, in line with expectations regarding the life cycle stage that baby boomers occupy: they are more likely to be married (=60%) and have larger families (family size=3.05) than younger Americans (Rus- sell, 1999). In view of Chinese and American cohorts’ differential life experiences, we anticipate cross-national differences in home and car owner- ship across cohorts. Whereas the baby boomers’ spending continues to rise compared with younger cohorts, Red Guards’ ownership of homes and cars, is likely to be comparatively less than that of Modern Realists and Global Materialists. Hypothesis 7: Unlike baby boomers in the United States, Red Guards in China are less likely to be home owners than (a) Modern Realists or (b) Global Materialists. Hypothesis 8: Unlike baby boomers in the United States, Red Guards in China are less likely to be car owners than (a) Modern Realists or (b) Global Materialist. Methods Research design The data from 15 cities were collected by China Central Television (CCTV) between 1999 and 2001. We categorize the cities surveyed into tier 1 cities, which include the urban centers of Beljing, Shang- hai, Guangzhou, and Shenzhen (mean per capita GDP=CNY 39,248), and tier 2 cities, which include 11 economically less developed cities (Chengdu, Chongging, Dalian, Fuzhou, Hangzhou, Jinan, Nanjing, Shenyang, Xi'an, Tianjin, and Wuhan; ‘mean per capita GDP=CNY 19,400) (China Statistical Yearbook, 2002). Using the method of probability proportional to sizes, CCTV randomly selected respondents from each city. The interviewers com- pleted half the surveys during face-to-face inter- ‘views; respondents completed the remaining half on a self-administered basis, and then received a small token gift. Of the 48,000 respondents, 32,670 fell within the age range of the cohorts, whom we categorized on the basis of thelr age in the year 2000: 18.3% were Global Materialists, 38% were Modern, Realists, and 43.79 were Red Guards. Materialism This survey assesses respondents’ materialistic values using an eight-item measure adapted from.Socal nsthatlonal approach to ening genealn aborts set sang tt - 3 “ Table 4 Factor analysis of materialistic values (Study 2) Items Foctor loading Factor 1: Moteralism (eversecoded) (46.5% of variance explained) | would not replace products that are stil usable 0.775 Although | like 2 product, ll buy it only after considering its usefulness ong Furetionalty is much more important than appearance in daily necessities 0.626 The products | buy should wear well 0.592 Taylor Nelson Sofres (five-point scales ranging from. strongly disagree to strongly agree). We conducted a principal component factor analysis with varimax rotation and, using the eigenvalue (>1) and Scree test as criteria, obtain two factors. However, the composite reliability shows that only factor 7 (=0.62) meets the 0.60 criterion (Bagozzi and Yi, 1988), so we retain only this factor as indicated in Table 4 for subsequent analysis. A CFA demon- strates the factor’s acceptable goodness-of-fit statis- tics (7=86.649, df=2, P<0.001, CFI=0.978, GFI=0.995, AGFI=0.975, RMSEA=0.068). All stan- dardized loadings are significant at the P<0.001 level, Brand consumption Using yes/no responses, respondents indicated the brands they had used during the preceding 12 months; multiple responses were allowed. We examined two low-involvement products, tooth. paste and shampoo, and two value-expressive products, casual shoes and beer. The brands listed in the survey included major foreign and local brands in China: 36 brands of toothpaste, 59 brands of shampoo, 35 brands of casual shoes, and 62 brands of beer. We arranged the brands by product category and listed them alphabetically using Chinese pinyin. The only additional ‘brand’ that the respondents indicated they had consumed was ‘generics’ (no brand). One of the authors, a native of China, categorized the list into foreign and local brands and included generics within the local category. ‘We categorize users of the respective products into foreign and local users and define these groups such that foreign users have used at least one foreign brand in that category during the year. Therefore the foreign users group includes those who use foreign brands exclusively and those who use a mix of foreign and local brands, whereas local users use local brands exclusively. We classify the mixed mode as foreign users because very few respondents used foreign brands exclusively. To =86.649 (2 dif), P<0.001, CFi=0.978, GFl=0.995, AGFI=0.975, RMSEA=0.068. ensure the reliability of our classification scheme, we conducted sensitivity tests with altemative mixes of local and foreign brands: the results of these tests remain consistent, and reflect the brand usage pattern we reveal with our initial classifica- tion, in which Global Materialists use more foreign brands than do Modern Realists and Red Guards. Materialism across cohorts (H4) H4 suggests that (a) Global Materialists are more materialistic than Modern Realists, who in turn (b) are more materialistic than Red Guards. According to our ANCOVA run (with city tier, gender, income, and education as covariates), the results demon- strate a significant cohort effect for Global Materi- alists, Modern Realists, and Red Guards (F(2,7412): 104.55, P<0.001; M’5=10.13, 9.90, 8.96, respec- tively). Pairwise comparisons show significant mean differences, with the Global Materialists scoring higher mean values than Modern Realists (P<0.001), who in turn score higher mean values than the Red Guards on materialism (P'<0.001), in support of both H4a and Hiab. The findings show that two covariates - education (F(1,7421)=136.74, P-<0,001) and family income (F(1,7421)=93.03, P<0.001) ~ also exert significant effects Foreign brand usage across cohorts (HS) HS suggests significant differences across cohorts in foreign brand usage, such that Global Materialist ‘have the highest likelihood of using foreign brands. ‘We conduct a series of logistic regressions, one for each product, in which we employ respondents’ brand usage data as the dependent variable (for- eign=1, local=0, non-users excluded). We also include independent variables in the logistic regres sions, namely city tier, education, and family income; as well as gender, marital status (single, married, other), and family size (=1, 2, 3, 4, 5, 6, 7, >7) to serve as proxies for the respondent's life cycle stage; and housing option (private, rental, firm-owned) and car ownership (yes, no) as proxies for major purchases that could reduce respondents’ Journal ot ateratonal oarer Soies3 set natal aprsch to Wentng geen chon ae level of disposable income. In China, private- owned residences and rental residences require payment at market value, whereas firm-owned residences refer to subsidized housing that gener- ally requires below-market payments. We code cohort, a categorical variable, as an indicator contrast, for which we use Global Materialists as the reference: that is, cohort (1) contrasts Modern Realists and Global Materialists, and cohort (2) contrasts Red Guards and Global Materialists. The remaining categorical data (e.g., family income, housing) also provide indicator contrasts, so that the base level of each variable (e.g., CNY 2000 or below, private ownership) serves as the reference. ‘The regression models are robust, yielding high percentages of correct classifications (60.1-90.2%, Table 5, bottom row). The results (Table 5) indicate that six of the eight cohort coefficients (2 cohort contrasts x 4 products) register significant effects (P<0.05). Global Materi- alists are more likely than Modern Realists to use foreign brands of shoes (°°, P<0.05) and beer (e°#, P<0.01), but not toothpaste (9%, ota tg at - P>0.08) or shampoo (e "5, P>0.05). They also are more likely than Red Guards to use foreign brands of all four products: toothpaste (e °7", P-<0.05), shampoo (¢ °2%4, P<0.05), shoes (e-°47°, P<0.05), and beer (e~°*"?, P<0.001). The logit coefficients for the categorical cohort variables indicate that a unit increase in cohort is associated with a b, change in the log odds of foreign brand consumption. For example, compared with a Global Materialist, it is only e~°25° (=0.70) times as likely for a Modern Realist or only €-°47° (0.63) times as likely for a Red Guard to have used foreign casual shoes duting the preceding 12. months. ‘These findings provide partial support for HSa and full support for Hb. Usage of foreign brands of value-expressive products across cohorts (H6) 16 suggests that Global Materialists are more likely to use the foreign brands of value-expressive products than are elther Modern Realists or Red Guards, We again refer to Table 5, which shows that all four cohort coefficients related to value- Table 5 Foreign brand usage as a binary logic function of cohort and socioeconomic factors Product categon? Toothpaste Shompoe Casual shoes seer Coton” Cohor (1) -0.093 0195 =0.356" -oasere Cohort 2) “0.216 =0.394" 0.470" “0612 Scioeconomic variables City ter ~0233ee ~o40re ~0.4see -oariee Education 0.0956 0.080% =0.029 0.089% Family income 0.396" 0329" 0.342 0.326" nnership of mojr items Home ownership Rental vs pvate-owned 0.030 -0216 ons? 0278 Firm 5 petvate-awned =o.se 0.080 oo7t =01is ar ownership 0.000 0.162 0.390 0036 Lie cyl stoges ond gender ‘Gender® 0.324 0.266 -0.400re 049 Maral status? 0070 0250 =0087 ~0.296" Family size ois 0.098 “ona =0010 Nagelerke 0.06 0.03 007 0.09 200d!) 300.88 96.76" 109.695 27239" eof correct classeation 60.1 902 6 633 “Dependent variable=1 ifthe respondent consumed atleast one foreign brand in the preceding 12 months in the category, Of otherwise. "Cohort is measured as a categorcal variable ving the frst cohort (Global Mates) asthe reference for contrast with the rest of cohorts ‘Gt ter measures the contrast eflect of ier T cites Bling, Shanghai, Guangzhou, Shenzhen) against ethers Gender contrasts female against mat ‘mtatal status: Tmsingle, 2=maried, 3=dhvorced/widowed, ?-<0.05; *P-<0.01; **P-<0.001 ‘uma iterations Business SaisSocal inttutlonal approach to Kdentlfying generation cohorts expressive products (2 cohort contrasts x2 pro- ducts) register significant effects (P<0.05). Global Materialists have a higher likelihood of using foreign casual shoes and beer than either Modern Realists (shoes: e~°*, P<0.05; beer: e°**?, P-<0.01) ot Red Guards (shoes: €°4”, P<0.05; beer: ¢°°61, P<0.01). In contrast, only two of the four cohort coefficients related to low-involvement products register significant effects (P-<0.05). The significant cohort coefficients include those con- trasting Global Materialists with Red Guards (cohort (2), toothpaste; cohort (2), shampoo), but neither of the cohort coefficients contrasting Global Materialists with Modern Realists (cohort (J), toothpaste; cohort (1), shampoo) registers significant effects (P>0.05). Because the findings pertaining to low-involvement products differ from those pertaining to value-expressive products (cohort (1), casual shoes, P<0.05; cohort (1), beer, P<0,01) for the same cohort contrasts, we consider Hé6a and Héb supported In addition to these cohort effects, several inde- pendent variables register significant effects. For example, city tier (P<0.001, tier 1 cities more likely) and family income (<0,001, higher income more likely) register significant effects on all four pro- ducts. Gender registers significant effects on three of the four products: toothpaste (P<0.001, female more likely), shampoo (P< 0.01, female more likely), and casual shoes (P<0.001, male more likely) Cross-national comparisons (H7, H8) 17 and H8 suggest that unlike baby boomers in the United States, whose spending on home and cars is greater than that of younger Americans, Red Guards are less likely to be home owners (H7) or car owners (H18) than Global Materialists or Modern Realists. We conduct cross-tabulation tests to assess cohort effects on these variables. The results regarding the housing option show significant differences (;?=490.41, P<0.001); more home owners among Modern Realists (45.2 vs 44.6% and 42.7% for Global Materialists and Red Guards), more renters among Global Materialists (9.8 vs 8.7% and 3.3% for Modern Realists and Red Guards), and more residents of firm-owned housing, among Red Guards ($3.9 vs 45.5% and 46.1% for Global Materials and Modern Realists). Home own- ership thus reflects an inverted-U pattern across cohorts that peaks with Modern Realists, in con- trast to the upward-sloping pattern in the US housing data, which peaks with baby boomers. ‘These findings provide support for H7a and H7b, in manned oo that Red Guards are less likely to own a home than are the other cohorts. Regarding car ownership, the results also are significant (’=20.90, P<0.001). That is, more Modern Realists (4.5%) own cars than do either Red Guards (3.9%) or Global Materialists (2.8%). ‘The average of 3.9% car ownership in our sample, though low by US standards, is consistent with the national average in urban China (China Statistical Yearbook, 2002). Thus car ownership also suggests an inverted-U pattern across cohorts that peaks with Modern Realists; again, in the United States, the pattern slopes upwards, because it peaks with the baby boomers. These findings also show that Red Guards are less likely than Modern Realists but more likely than Global Materialists to own cars, in support of H8a but not H8b. We assess the marital status and family size across cohorts to evaluate whether the Red Guards exist in a life cycle stage similar to that of the baby boomers. The results for both marital status (cross-tabulation, 714,934.28, P<0.001) and family size (ANOVA, F=19.57, P<0.001) are significant: 93.4% of Red Guards are married, and their family size (mean=3.14) is the largest among cohorts in China. These findings suggest that Red Guards appear in a similar life cycle stage to that of the baby boomers in the United States, who also are more likely to be married and have larger families than younger Americans; however, Red Guards are less likely to own a home and/or car than are Modern Realists (home, car) or Global Materialists (home). These findings support our postulate that differences in consumption among Chinese cohorts are probably due to cohort rather than life cycle differences (Table 6) Conclusion This study offers a framework that involves historic cataclysmic events, changing institutions, and cohort membership to identify meaningful market segments in China, By retracing and understanding the societal events that have taken place, we identify and validate, with two studies, several segments of consumers in China with different values, attitudes, and consumption behavior. Red Guards, who came of age during the Cultural Revolution (1966-1979), are pessimistic and frugal, consider shopping a burden, and are less likely to use novelty products or foreign brands. Modern Realists, who came of age during the Economic Reform (1980-1991), are materialistic novelty see- ers who consider shopping a leisure activity and Toural of erations! Bases Suisitutonsl approsch to Mdentfying generation cohorts __ nts ug of Table 6 Home, car ownership and lfe cycle stages across cohorts ed guords Chinese cohort Global materass ‘Modern realists Age at Study 2 (2000) 18-26 27-35 36-49 Year of birth 1974-1984 1965-1973, 1951-1964 Home/cor ownership Housing options (449041, P<0,001) Private-owned 44.6% 45.2% 42.7% Rental 91856 87% 3.3% Fiem-owned 45.5% 46.1% 53.956 Car ownership (7220.90, P<0.001) Yes 2.8% 45% 4.9% Lie ce sages Mortal satus (?=14934.28, P<0.001) Single 87.2% 20.1% 2.8% Married 125% 78.6% 93.495 Others 0.396 13% 3.8% Family size (F=19.57, P<0.001) Mean 296 Baa aus American cohorts - ~ Baby boomers Age in 1999 <25 25-34 344 45-54 Year of birth 1975- 1965-1974 1946-1964 Indexed spending* Home t8 78 106 120 car 3 82 na 17 le ce stages Totat households ‘aby boomer households Married couples 53% 60% Family size 2.62 3.05 ‘Based on results of study 2 Source: Russell and Mitchel (1999), ‘Source; Russel (1999). use novelty products regularly. Finally, Global Materialists, who came of age during the globalizs- tion of China (1992-_), are the most novelty seeking and materialistic and have the highest propensity to use novelty products, especially foreign value- expressive products. These findings show that social institutional changes that arose from cataclysmic events have created distinct cohorts in China, whose life experiences during their critical coming-of-age years continued to shape their values and behavior long after the events took place, Our study also compares Chinese and American consumers, Whereas Red Guards and baby boomers are of similar ages and life cycle stages, Red Guards are less likely to own homes or cars than are younger Chinese, because these older consumers, ‘who came of age under the pre-reform regime in China, tend to persist in their less materialistic Fournat o reratona Baines dies ways and stay in firm-owned, more basic housing. in contrast, the US baby boomers increase their spending on homes and cars relative to younger Americans. These findings demonstrate that at least some cross-national differences in consumption behavior ate due to cohort differences. Thus researchers should investigate not only cross- national but also intracultural variations in inter- national markets. By including cohort membership in their segmentation schemes, international marketing managers could exploit consumers’ age groups, which reflect the events they experienced as they came of age, to facilitate product and com- munication campaigns. Property managers in China are already drawing on insights from the Global Materialists’ materialistic _ orientations and targeting the under-30 segment. This group of young consumers, who have been nicknamed_ sec nttlonlnpprnch o Wetifng ganarain carts nts 2 ‘home slaves’ (Sun, 2006), are keen to own not just a home but a brand new rather than a second- hand home. Similarly, the coffee industry is associating its caffeinated beverage with feisure and fine living to appeal to Global Materialists, who constitute the heaviest coffee drinkers in China (see Study 1 results). In the US, in contrast, older adults remain the heaviest coffee drinkers, even after we take Into consideration the Starbucks effect (Wolfe, 2006). Managers also use insights derived from the Red Guards’ nostalgia for the hardships of the Cultural Revolution to design products and persuasive messages, This trend can be seen in the emergence of Cultural Revolution theme restaurants, complete with basic fare and simple decor. Even advertise- ments have begun to use music and icons from the Cultural Revolution to appeal to a cohort that looks back with mixed emotions to an era of camaraderie and shared sacrifices. Thus an understanding of generational cohorts in China provides marketing managers with insights into this important inter- national market. Future research directions and limitations We have examined cohorts’ values, attitudes, and behaviors, but we do not measure specifically their perceptions of the life-changing events through which they lived. Further research could include such measures, and examine the mechanisms and processes that link cataclysmic events to the cohorts’ personal and consumption values, atti- tudes, and behaviors. Meanwhile, the trend of targeting Red Guards shows that, though cohorts exist because their members feel a sense of shared experience, which they reinforce through one another, they also continue to evolve as their collective memories become more selective over time. Accordingly, marketing managers should remain aware of the evolving values of the cohorts to which they ate marketing. Our findings show that socioeconomic variables such as income and education exert strong influ- ences on Chinese consumers’ foreign brand ‘usage. Similarly, life cycle events such as marital status and family size exert important influences References Abramson, P.R. and Inglehart, R. (1995) Value Change in Global Perspective, University oF Michigan Press: Ann Arbor, Ml on both Chinese and American consumers’ major purchases (except Red Guards). That is, the predicted outcomes are influenced by multiple factors, and cannot be explained entirely by cohort effects alone. Further research should build on these findings, and develop a matrix that includes multiple segmentation bases and different products to identify the circumstances in which a specific variable forms a meaningful segmentation basis for a particular product. The matrix also could offer a more comprehensive understanding of segment characteristics among Chinese consumers, and their similarities and differences in terms of corresponding segments in other countries, We also acknowledge several limitations to our study. Although we examine potential age con- founds (e.g, life cycle stages), our study is limited by the measures included in the database. For example, we lack measures of materialism with proven psychometric properties (Richins and Daw- son, 1992). In addition, we investigate primarily Chinese consumers residing in urban areas, and our cross-national comparison is highly restrictive, because our data do not allow for a direct comparison. Additional research should therefore examine the applicability of cohort segmentation in other countries. Although we include multiple products to enhance the reliability of our findings, our study also is limited to the categories of products that we surveyed. Finally, this study could be enriched with longitudinal datasets, as some exemplary work on cross-cultural consumption has shown (Ise et al., 1989). Acknowledgements ‘The authors thank the Departmental Editor and three reviewers for their constructive comments. The finan- Cial assistance provided by the Research Grant Council (RGC) competitive earmarked research grant is acknowledged. Note “The years of birth for baby boomers are based on Russell's (1999) categorization. ‘Achenreiner, C.8, (1997) ‘Materlistic values and susceptibilty to Influence in children’, Advances in Consumer Research 24: 82-88, Tura of erro Bins Sa* soca netetionst 52 ‘Alden, DL, Steenkamp, EM, and Batra, R. (1999) ‘Brand itioning through advertising in Asia, North America and urope: the fle of global consumer culture Jour of Marketing 63(1); 75-8: Anderson, 1.C. (1987) ‘An approach for confirmatory measure- ment and suctral equation modeling of organizational properties, Management Science 33(4): 525-541, satel, 1. (2002). The psychology of globalization’, The “American Psychologist 57010). 774-783, ‘ns, KY. (1998) “Inia-cultual variation: evidence and implica- tlons for international business, Jounal of Intemational Business Studies 304): 799-812 Bagozz, RP. and ¥i, Y. (1988) ‘On the evalustion of structural ‘equation models, Journal of the Academy of Marketing Science Yea): 74°94 Bagozz RP. Yi, Y. and Philips, LW. (1991) ‘Assessing construct ‘alcity in organizational research’, Administrative Science Quortery 36,3). 421-458. Batra, R. and. Tse, D.K. (2003) ‘Marketing in. transition economies, our of eraine| Marketing VIO 57. Batra, R, Ramaswamy, V, Alden, D.L, Steenkamp, | EM. and Ramachander, 5. (2600) ‘fects of brand local and nonlocal origin on consumer atitudes in developing countries, Journal {of Consumer Psychology 92): 83-95. Belk, RW. (1984) Three scales to measure constructs related to ‘Materialism: realty; valiity, and relationships to measures fof happiness, Advances n Consumer Reseorch 11: 291-297, Belk, RW. (1988) ‘Possessions and the extended sel, Journal of Consumer Reseorch 15(2): 138-168, Belk, RW. (1981) “Possessions and the Sense of Pas’ in RW. Belk (ed. Highways and Buyways: Naturalistic Research from the Consumer Behavior Odyssey, Association for Consumer Research: Valdosta, GA, pp: 114-130. Belk, RW, and Zhou, N. (2002) ‘A reader response analysis of global and local appeals in Chinese. consumer advertising’, Advances in Consumer Research 29: 339-341 Solsot, M. and Child, |. (1999) ‘Organizations as adaptive systems in complex environments: the case of China’ Organization Science 10(3). 237-253. Bollen, KA. (1989) Structural Equations with Latent Variables, Ie Wiley & Sons: New York ‘Campbell, €.(1987) The Consumer Ethic and the Spirit of Modern Hedonism, Basil Blackwell: Oxford. Chen, F. (2005) ‘Residential pattems of parents and their marred children in. contemporary China: le course approch, Population Reserch and Pacy Review 242): 125-148 chen, ¥.X. (1999) ‘Lost in revolution and reform: the socio- ‘economic pains of China's Red Guards generation, 1966 996%, Journal of Contemporary China 821): 219-239. China Advertising Association (2002) ’A 31 fold increase of CChina’s ad revenue in the past decade’, [www document] hittp:fAwwn.a.com.en (In Chinese) (accessed 22 November, 2004). Connerton, P. (1989) How Societies Remember, Cambridge University Pres Combe, ‘DiMaggio, P.. and Powell, W.W. (1983) "The iron cage revisited: institutional isomorphism and collective rationality in organi- ational fields, American Sociological Review 48(2): 147-160. Eastern, RA. and Crimmins, EM, (1991) ‘Private materialism, ‘personal se-ulilment, family life, and public interest: the hatute, effects, and causes of recent changes in the values of ‘American youth’, Pubic Opinion Quarterly 854): 499-533. Ear C.P. and Ralston, DA. (2004) ‘Generation cohorts and ‘personal values: a comparison of China and the United States, Srganization Science 15(2). 210-220. Fong, V.L. (2004) Only Hope: Coming of Age under China's ‘One-child Policy, Stanford University ress: Palo Alto, CA. Francese, . (2001) Big. spenders”, American’ Deriagraphis '23(9): 30-31. Gardner, C. and Sheppard |. (1989) Consuming Passion: The Rise ‘of Retail Culture, Unwin Hyman: London, urna fates Baines Seales to Identifying generation cohorts__kneta Mang at Gibson, 8. and Sanbonmatsy, OM, (2004) ‘Optimism, pes ‘mism, and gambling: the downside of optimism, Personality tnd Soca Psychology Buletin 302). 149-160, Goldberg, ME, Gorn, Gi, Peracchio, LA. and Bamossy, G. (2003) ‘Unde'standing matenasm among youth, Journ of nse hg $23). 278-708 Halbwacts, ML (1950) The Colective Memory, Harper: New York Holbrook, MB. and Schindler, RM. (1989) "Some exploratoy, findings on the development of musial tastes” Joural of Consumer Reseorch 16(15. 119-124 Hlfmany .,Ratneshway, , and Mick, D.G. (2000) ‘Consumer onl Structure and’ Goal-Determnation Processes An Inte grative Framework, in'S, Ratneshwar, D.c, Mick and © Hutfman (eds) The Why of Consumption, Roviledge: London, pp: 9-35. Hung, K and i, 8.¥, (2007) eWOM in a Vieual Community Social Embeddecnes, Consumer Leaming, and Reflex, Wrkang Paper Une of og Kong Hung, Ke Guy RF. and tse, D.K_ (2008) “Improving media decisions in China: ad targetabilty and cost-beneft analysis, Journal of Adverting 34(1) 49-64 tnglehar,'R (1977) The Silent Revolution: Changing Values and Pole Sites omarg Wester Publes, Princeton University Press: Princeton Ni Inglehart R984) Value Change inthe Uncertain 1970s, in G Blugos and K. Welermair (eds) Monegement Under bitering Value Systems, De Gruyter etn and New York, pp: 75-108 Inglehart 8 (1990) Colture Shit In evonced Idol Socety, Brinceton Univers Pes: Princeton, Ni Inglehart, R. and Norris, P. (2003) Rising Ride: Gender Equality ad Caltiol Change ori the Word, Camdige University Press Camby Jing, I. {e8) (2060) Feeding China's Litle Emperor: Food Ghitéen and Socio! Change, Sanford University Press: Pal Ato, CA Ioharison, JK and Thorei, H.8. (1985) international prod, postoning’, Journal of Intemational Busines Stadt 16(3): Soars, Johar 8, nd Sty, Mi, 1991) "akueorpresive vs tain advertsing appeais: when and why to use Which pea” Ioural of Adverting 300: 33-33. Pe Lau, EM, Tse, DAK and Zhou, N, (2002) “Instutional forces ‘and ‘organizational culture ia China. effzcts on. change Schemas, frm commitment and jab satiation’, Journal of Intemational Business Stacie 33(3). 933-550. Lipsita G. (1990) Time Possages:Collctne Memary and American Popdlor Cute, University of Minnesota Pres Minneapols, Lu iL, 2000) “To Be Relatively Conable In an Egan Serie’, D,. Dave (ed) The Consumer Revesion Uber Cin, “nvr “el cafem Pes Bec, "CA praca. Lunt, BAR and Uvngstone, $M. (1991) ‘Pychological, social ancagomi deen fing: compan eet 621-642. * ae Lut, Bi and Livingston, S34, (1992) Mess Contumtion and esol dent Brera foname Pperenc, Open Univer thy ress Sucktghar, PA McCrae, RR, Cote i, Tevaccano, A Parker, W.D, Mil Ei be Frye F end Mende, (2062) ersonaliy tra Gaveopment tem age 12 to Soe 48. tngitucnal, cose Secon) and eros itrl aati, furl! Pesoray Sd Soc! Payctology 83614361468 Meredith, Cl" and Schewe, CD. (2002) Defining Markets Deining Moments: Amerce’s 7” Generational Cotes, Ther Stored experonces, nd Why Business should Core Hungyy ings: New Yor Nee, V. (1982) “Organizational dynamics of market transtion ‘yl forms, propery rons and mneseconomy in Cine, Pamir cence Query 370. 12Socll institutional approach to Mentlfying generation cohorts North, D.C. (1990) Institutions, Institutional Change and Economic Performance, Cambridge University Press: Cambridge. Richins, -M. and Dawson, 5, (1992) A. consumer values ‘orientation for materialism and its measurement: scale development and validation, Journal of Consumer Research 1903). 303-316, Rogler, LH. (2002) ‘Historical generations and psychology: The ‘ase of the Great Depression ang World War Il’, The American Psychologist $7(12): 1013-1023, Rosen, 5, (2000) ‘Foreword’, in ¥. fang and D. Ashley (eds.) "Mao's Children in the New’ Chinas Voices from the Red Guard Generation, Routledge: New York, pp xi-xx. Russell, (1999) The Baby Boom Americans Aged 35 to 54, New ‘Strategist Publications: Ithaca, NY. Russell, C. and Mitchell, S. (1999) Best Customers: Demographics of Consumer Demand, New Strategist Publications: Ithaca, Nv. Scharinger, CM, and Danko, W.D. (1993) A conceptual and ‘npr’ Gamnpamon of atematve pouseele ie ie trode, Jur of Consumer Research 18(3) 380-594 schits 6. (1997) vino the Chinese eeosumer? Segment anh he People’s Republi of Chin’, Eurpeon Moregenent Fouret 3 bates sept Wk (1987) “The adolescence of Insutonl theory, “linia Sconce Quarter) 328) 495-311 ston S980) he role Gr attade objcs in atttude Tancion< euro! of Experiment Socio Polo 263) ar Sitgy, Mit, Gromal, ©. and Mangleburg, T. (2000) ‘Retail ironies selecongray, anata pabanage’ an itege, the model nha eset agenda, url of Bins Resetch Oy 87-138 silts Sasa! Bure (2002) Chino StottialYerboo, State Sinstal Barer People's hepuic of China, bling sisentomp, JEM, and Boungarnes, Hi. (1992) Me ole of ‘ptm titlaan leven explorsory consmer eho, Farol or Consumer hsoreh 193 334-448 sitenkamp, LEM. and Hofstede, F, (2002), “temstional mane Segrentau sues and perspective, ‘ntratonel Jourmot of Resch n Morcing 193) 195-213 sitertamp, [EM fat Rand Alen, Dt (2003) “How ered brand’ gobaess crests brand save, Jounal of Internal Busts Stues Sat) 33-6, siteskonp TEM Holtede, Find Wel, M. (199) “4 essnavoral ivesigaon info the Inca an abral atu antecedents of consumer mnovevenes ous o orig 650) 95-59 sul 20) Be You Le De Nian Qing Fang No Men’ {The ‘ug Hore Slaves) chino News Wook 29 hy. Tang.'W'nd Pathe WU. (2000) Chee Ubon ite under ‘lo: the changing Soi Conus, Cambrge Univers Press Combadge 130 Bice Bak RW. and. Zhou N. (1989) ‘Becoming. a onsrser society: 8 langtudinal and crosecutrl content stabs pant fas fon Hong Kong he peoples epubhc Br china ahd Tawa ural of Consume Resear 3) Spare vaughn, 8 (1986) "How advertising works: a panning moder, Journal of Navertsing Research 2603) 27-33. senna % Walder, A.G. (1995) ‘Local governments as industell fms: an ‘organizational analyse. of China's transitional economy’, ‘mencan Journal of Sociology 1OVC2) 263-301 Walters, .C.P. and Same, S. (2003) "Marketing strategy in ‘emerging markets: the cade of China’, Jounal of Internatlonal Marketing THC. 97-106. Wang, . (1994) Qualty e Life in Urban China: A Dato-Bosed ‘Study inthe Cy of Toni, University of Michigan Dissertation Services Ann Arbor, Rl Weiss, Mi, (2002) ‘Chasing youth’, American Demogrophic 2463): 34-41 Wotte, A: (2006) ‘Study: Americans coffee consumption ties sot inks, Gourmet News 713): 8. xu, ¥. (2000) Face the History [Zhi Mian Ui Shi, Zhongguo Wen lag Chu Ban She: Bein Yao, ¥.Z. and tu, H. (1993) Zhongguo Da Tou Sh (1956-1992), hong Yang Mn Zu Xue Yuan Chu Ban She: Being hang, ¥. and Huang, & (2003) Report on China's Iterstional ‘tots 2003, Yuandong: Shangha. Zou, Sand Cavusgi, "ST. (2002) The CMS: 4 broad ‘Conceptualizaton of global marketing stiategy and ts effec fn frm performance’ ournal of Marketing 664. 40-56. Zallow, Hiv (1991) ‘Bessie rumination im populat songs ‘and news magazines predict economic recession wa decreased consumer’ apinism’ and spending Jounal of Eeananc Psycroogy 123). 501-526. About the authors Kineta H Hung is Assistant Professor of Marketing, School of Business, University of Hong Kong. Her research interests include advertising, online com- munities, and consumption behavior in transi- tional economies, She has published in Joumal of Advertising and Journal of Retailing. Flora Fang Gu is Assistant Professor at the Depart- ment of Management and Marketing, The Hong, Kong Polytechnic University. Her current research interests include business-to-business relationship management, distribution channels, and mark issues in China, Chi Kin (Bennett) Vim is Associate Professor of Marketing, School of Business, University of Hong Kong. His research interests include services marketing, firm strategy, and marketing in transi- tional economies. He has published in Journal of Marketing, Journal of Marketing Research, Marketing Science, Journal of Retailing, and Journal of World Business. ‘Accepted by David K Tse, Departmental Eltr, 3 October 2006. Tis paper has beer with the authors fortwo revisions Tourna of nternatana Busines Sales