Demarketing Minorities, And National Attachment

Amir Grinstein & Udi Nisan

Demarketing, Minorities, and National Attachment This study addresses two important global trends: protection of public goods, specifically the environment, and the emergence of multiethnic societies with influential minority groups. The study tests the effect of a government proenvironmental demarketing campaign on the deconsumption behavior of minority groups and the majority population. It suggests that minority consumers use consumption or deconsumption to manifest their social identity, beliefs, and goals as minorities in relation to the majority and that their motivation to respond positively to a government’s demarketing campaign is shaped by their national attachment levels. The study was conducted in Jerusalem, Israel, and involved a large data set (N = 66,272) containing household-level data on actual behavior of consumers. The studied groups involved the majority of non-ultra-Orthodox Jews and three minority groups: Israeli Arabs, ultra-Orthodox Jews, and Jewish Russian immigrants. The findings show that in a proenvironmental context, government demarketing is more effective on the majority group than on any of the three minority groups. In addition, minority groups with lower national attachment levels respond more negatively to the demarketing effort. However, higher education levels lead to a more positive response across the majority and minority groups. A central implication of the findings is related to the benefits derived from demarketing for different consumer groups (majorities versus minorities based on religion, ethnicity, and immigration status). The findings also provide a benchmark for demarketing efforts in an environmental context. Finally, they suggest that policy makers and marketers should consider using demarketing more to complement regulatory and economic tools.

Keywords: demarketing, minority groups, national attachment, public goods, public policy, environmental behavior

Furthermore, the use of proenvironmental demarketing is more complex in highly heterogeneous societies with influential minority groups and increased interethnic conflict. Although there have been calls to account for cultural diversity through minority targeting in marketing efforts (e.g., Deshpandé and Stayman 1994), an important open question is whether and under what conditions minority groups conform to the general norm of provisioning public goods and, specifically, how these groups react to government demarketing efforts in an environmental context. Knowledge of the ways minority groups behave in this context would help policy makers better cater to the heterogeneous needs of different ethnic groups, minimize tension and conflict among these groups, and implement policy more effectively, while developing and protecting the environment. We contend that minority consumers’ motivation to respond positively to a government’s demarketing campaign is often shaped by the tension between national and minority group attachment. National attachment reflects a person’s feelings of loyalty and patriotism to and sense of identification with a country (Sidanius et al. 1997). Minority group attachment reflects a person’s respect for the main symbols of the group, its rules, and leadership and the belief that the group is superior to others (Roccas, Klar, and Liviatan 2006). There is often a conflict between attachment to a minority group and attachment to the country because minorities and the majority may possess a different set of goals and priorities (Berger and Heath 2007; Triandis 1995). In turn, this may result in severe ethnic conflict and substantial economic losses to a country (Penaloza and Gilly 1999; Triandis 1995). Furthermore, because con-

n recent years, two global trends have dramatically changed cultures, consumption patterns, and marketers’ behaviors. The first, the protection and development of public goods, especially the environment, has become a major concern for societies (e.g., Baker and Sinkula 2005; Banerjee, Iyer, and Kashyap 2003). Second, global populations have become increasingly diverse and multiethnic, leading to an increase in conflict between different ethnic groups (e.g., Grier and Deshpandé 2001; Penaloza and Gilly 1999). In the wake of the first trend, public policy makers in many countries have tried to discourage people from consuming products or following behaviors that may have a negative effect on the environment, using economic, legal, technological, and demarketing approaches. However, the effectiveness of these efforts, particularly demarketing, which is of special interest to marketers and marketing scholars, is not clear (Pechmann et al. 2003; Wall 2005).

I

Amir Grinstein is Assistant Professor of Marketing, Guilford Glazer School of Business and Management, Ben-Gurion University of the Negev, BeerSheva (e-mail: [email protected]). Udi Nisan is Director General, Israel Government Companies Authority, and a lecturer, School of Public Policy, The Hebrew University, Jerusalem (e-mail: [email protected]). The authors acknowledge Liza Teper and Ori Sharf for their excellent research assistance. They are grateful to Gal Hershkovitz, Hagichon, and the Municipality of Jerusalem for providing the data for this research. They also express thanks to Rohit Deshpandé, William Baker, Danit Ein-Gar, and three anonymous JM reviewers for their helpful and constructive comments. The research was conducted while the authors were visiting scholars at Harvard University: the first author at the Harvard Business School and the second author at Kennedy School of Government.

© 2009, American Marketing Association ISSN: 0022-2429 (print), 1547-7185 (electronic)

105

Journal of Marketing Vol. 73 (March 2009), 105–122

sumers often use consumption as a way to express themselves (Aaker 1999; Belk 1988), particularly in product domains that are symbolic of identity (Berger and Heath 2007; Levy 1959), minority consumers may use consumption or deconsumption to manifest their social identity, beliefs, and goals as minorities, demonstrating their position in relation to the majority group and the government that represents it (Deshpandé and Stayman 1994; Ustuner and Holt 2007). This suggests that the greater the tension between national and group attachment, the less is minority consumers’ motivation to respond positively to national demarketing efforts. Our conceptual framework combines research on demarketing and environmental protection with literature on group attachment and consumption as self-expression. Drawing on this, we develop a set of hypotheses aimed at understanding the effect of a government proenvironmental demarketing campaign on minority groups versus the majority population, its effect on minority groups with different national attachment levels, and the moderating effect of a key consumer characteristic—namely, education—on the consumption pattern of the minority and majority groups. Specifically, we use household-level data on actual consumers’ behavior to understand the implications of a government campaign aimed at reducing water consumption. The depletion of water resources is a major concern to societies and constitutes a frequently studied research area in environmental studies (Potter 2001; Talukdar, Gulyani, and Salmen 2005). Furthermore, household water consumption provides an interesting case because of its extremely low demand elasticity (Deutsch and Liebermann 1985). Finally, water and the meaning of water play an important role in shaping national attachment and conflicts among groups (Alatout 2006; Menahem 1998; Morag 2001). We studied a government demarketing campaign that took place in Israel in the 1999–2001 period, following a few years of drought that, in turn, led to a crisis in the Israeli water sector. We draw our large and heterogeneous sample (N = 66,272) from Jerusalem, Israel’s capital and a highly heterogeneous metropolis. Jerusalem includes three main minority groups—Israeli Arabs, ultra-Orthodox Jews, and Jewish Russian immigrants—who live alongside the majority of non-ultra-Orthodox Jews. The Israeli case is also unique because it enables us to jointly study and compare three facets of minorities: religion (Jews versus Muslims and Christians), ethnic background (ultra-Orthodox Jews versus Jewish Russian immigrants), and immigration status (immigrants versus nonimmigrants). This article contributes to the marketing literature in several ways. First, it addresses and jointly studies two important global trends: environmental protection and the emergence of multiethnic societies composed of influential minority groups. It tests the effect of a government proenvironmental demarketing campaign on minority groups, thus contributing to research on ethnic and minority groups’ consumption patterns and reactions to marketers’ efforts (e.g., Deshpandé and Stayman 1994; Grier and Deshpandé 2001; Penaloza 1994) and to research on environmental marketing (e.g., Banerjee, Iyer, and Kashyap 2003; Pieters et al. 1998; Schlegelmilch, Bohlen, and Diamantopoulos 1996).

106 / Journal of Marketing, March 2009

Specifically, we find that the three minority groups studied were less responsive than the majority population in reducing their water consumption and that the less the national attachment of the minority group, the less responsive was the group. The findings suggest that higher education levels were related to a more positive response across both minority and majority consumers. Second, the study contributes to the debate in marketing about the effectiveness of demarketing (Pechmann et al. 2003), providing evidence of a general positive effect in the population. The validity of our results is enhanced because the “product” examined is household water consumption, an item with extremely low demand elasticity. To be regarded successful in such a product category, a demarketing campaign must overcome relatively strong consumer resistance (Deutsch and Liebermann 1985). Third, the study is one of the few available empirical demarketing and environmental marketing studies based on actual consumer behavior, not self-reported attitudes (Deutsch and Liebermann 1985). This has several advantages over previous research, which has typically examined consumer attitudes using self-reported measures (Schlegelmilch, Bohlen, and Diamantopoulos 1996). Although selfreports can provide valuable insights into consumer motivations and attitudes, they might be biased because of their subjective nature. In addition, previous research has demonstrated a weak linkage between consumer attitudes and actual behavior (Rangan, Karim, and Sandberg 1996; Schlegelmilch, Bohlen, and Diamantopoulos 1996). Finally, the sample we used is especially large and at the household level rather than the aggregate level. We organize the rest of this article as follows: We begin with a review of the relevant literature and present our hypotheses. Then, we describe the methods used and our main findings. We conclude with a discussion of the findings, a discussion of the study’s limitations, and suggestions for further research.

Conceptual Development Environmental Protection: The Consumer Perspective The conservation and protection of the natural environment is an area of growing interest for policy makers, marketers, consumers, and society as a whole (Baker and Sinkula 2005; Banerjee, Iyer, and Kashyap 2003; Granzin and Olsen 1991; Menon and Menon 1997). Research in this area discusses the ways governments, companies, and consumers can develop and protect the environment. It specifically suggests that 30%–40% of environmental degradation has been brought about by the consumption activities of private households (Grunert 1993). Thus, changing consumer behavior can have a substantial impact on the environment (McDougall et al. 1981; Stern 1999). Society’s growing awareness of the environment has led to consumers who demonstrate concern through their purchasing decisions, preferring environmentally friendly, “green” products (Menon and Menon 1997; Newell, Goldsmith, and Banzhaf 1998; Zinkhan and Carlson 1995). The

number of proenvironmental consumers is constantly rising, increasing the targeted population for proenvironmental marketing and demarketing efforts (Granzin and Olsen 1991; Menon and Menon 1997; Pieters et al. 1998). Nonetheless, this behavioral change is not easy for most consumers. There are many products and behaviors whose deconsumption would benefit the environment but whose personal benefits to consumers are only secondary; furthermore, such environmental benefits are often associated with the distant future (Kasulis, Huettner, and Dikeman 1981; Osterhus 1997). Although the secondary personal benefits (e.g., social) should not be ignored, they are likely to be less valued when deconsumption is considered in the context of environmental protection (Frederick, Loewenstein, and O’Donoghue 2002; Rangan, Karim, and Sandberg 1996). In addition, such behavioral changes are costly, involving, for example, switching and transportation costs (Osterhus 1997; Pieters et al. 1998). Overall, this suggests that the intangible, costly, indirect future benefits of deconsuming non-environmentally-friendly products raise concerns about the ability to elicit desired consumer behavior (Pechmann et al. 2003; Rangan, Karim, and Sandberg 1996; Zinkhan and Carlson 1995).

Demarketing The need to change consumer behavior in the context of the environment leads public policy makers to employ a range of economic, legal, technological, and demarketing treatments (e.g., Andrews et al. 2004; Stern 1999; Wall 2005). Of special interest to marketers and marketing scholars is the effectiveness of demarketing. The concept of demarketing can be traced back to the 1970s. It was first defined by Kotler and Levy (1971, p. 75): “Demarketing is the aspect of marketing that deals with discouraging customers in general or a certain class of customers in particular on either a temporary or a permanent basis.” Demarketing campaigns typically involve traditional marketing efforts, such as advertising, public relations, and sponsorships (Deutsch and Liebermann 1985; Pechmann et al. 2003; Wall 2005). Empirical research on demarketing is scarce. Marketing scholars have studied demarketing primarily in the context of smoking (e.g., Andrews et al. 2004; Pechmann et al. 2003), drug use (e.g., Kelly, Swaim, and Wayman 1996), and energy conservation (e.g., Deutsch and Liebermann 1985; Kasulis, Huettner, and Dikeman 1981), often focusing on government, not business, demarketing (Gerstner, Hess, and Chu 1993). Although demarketing is viewed as more expensive than regulatory and economic tools, it is increasingly being employed (Andrews et al. 2004). Still, there is disagreement in the literature as to the effectiveness of these campaigns and whether they work better than alternative methods (Pechmann et al. 2003; Wall 2005). In the specific case of proenvironmental deconsumption, evidence is also inconclusive (e.g., Deutsch and Liebermann 1985; Newell, Goldsmith, and Banzhaf 1998; Osterhus 1997; Rangan, Karim, and Sandberg 1996). Overall, therefore, it is unclear whether demarketing works well in general and, in particular, for proenvironmental issues.

Group Attachment The growth of heterogeneous societies involving influential minorities may have significant consequences for consumer behavior in an environmental context. Specifically, the degree of tension between majority and minority groups may be important in explaining minority consumers’ willingness to protect the environment. Next, we introduce the concept of group attachment and discuss how tension between national attachment and minority group attachment may evolve. Across many disciplines, there is interest in examining the links people have to their groups. For example, social psychologists are interested in the way people identify with a group (e.g., Gaertner and Insko 2000), cross-cultural psychologists examine the tendency to extend the self to include groups (e.g., Triandis 1995), political scientists examine identification with political parties and the nation (e.g., Kosterman and Feshbach 1989), and marketing scholars study the impact of identification with a country on consumer behavior (e.g., Balabanis et al. 2001). Group attachment is evident when there is the emergence of a social pattern consisting of closely linked people who consider themselves a part of one or more collectives (family, coworkers, tribe, country). People who are highly attached to their social groups are primarily motivated by the norms of and duties imposed by those collectives, and they are willing to give priority to the goals of these collectives over their own personal goals and to emphasize their connectedness to the members of these collectives (Triandis 1995). In the specific case of national attachment, people define themselves in terms of their country membership. They extend their self-concept to include the country, respect the central symbols of the country (e.g., its flag, national anthem, rules, leadership), and believe that their country is better and more worthy than other countries (Roccas, Klar, and Liviatan 2006). National attachment is associated with patriotism, which is defined as the positive identification with the values of one’s country or simply as love for one’s country (Kosterman and Feshbach 1989; Schatz and Staub 1997). In the case of minorities, these groups often face the dilemma of balancing their in-group attachment with their attachment to their country (Sidanius et al. 1997). Indeed, the recent and dramatic increase in interethnic conflict and tension within multiethnic states (for examples, see Penaloza and Gilly 1999) raises the question whether it is possible to foster loyalty and identification with one’s own ethnic group and, at the same time, maintain shared national values and a sense of common national identification (Sidanius et al. 1997). There is an underlying assumption that national identity, or a sense of “belongingness” to the nation, is more strongly associated with membership in a dominant ethnic group (i.e., the majority) and less strongly associated with membership in subordinate groups (i.e., the minorities) (Sidanius et al. 1997). Furthermore, this state of national attachment is stable, not sensitive to contextual cues (Roccas et al. 2008). The tension between majority and minority groups and the associated differences in national attachment are driven

Demarketing, Minorities, and National Attachment / 107

by several forces. First, the groups often possess different sets of priorities and goals, and the tension increases when minority groups are not willing to sacrifice their goals for the sake of the majority (Triandis 1995). Second, dominant groups (i.e., the majority) often regard themselves as having a preeminent right to and ownership of the country, its resources, and its symbols, sometimes placing restrictions on the legal and social rights of subordinate groups (i.e., the minorities) (Sidanius et al. 1997). Indeed, in many countries, minorities lack political access and power and often experience discrimination, inequality, and alienation (e.g., Askegaard, Arnould, and Kjeldgaard 2005; Kretzmer 1990; Ling 2004; Ustuner and Holt 2007). Third, minorities often dissociate themselves from the majority to strengthen their social identity (Deshpandé and Stayman 1994; Ustuner and Holt 2007). Finally, tension is often apparent when people with affective attachment to the country show intolerance toward people and groups that do not conform to the country’s norms, such as minorities (Duckitt 1989).

Consumption as Self-Expression The degree of national attachment can be manifested through consumers’ consumption patterns. Consumers often use consumption as a way to express their identity and beliefs, and products are often consumed because of what they symbolize (Levy 1959). Objects, such as brands, and consumption patterns, such as eating out, can be associated with personality traits that provide self-expressive or symbolic benefits for the consumer (Aaker 1999; Berger and Heath 2007). These behaviors are important drivers of consumer preference and choice (Belk 1988). Consumers’ self-conception and the associated selfexpressive behaviors depend on the particular social role the consumer is playing, the degree of salience of this role, and the product domain (Aaker 1999; Belk 1988). Affiliation with a minority group is a salient social role, much more than belonging to a majority group (Deshpandé and Stayman 1994). Previous research in marketing has explored how affinity to a minority group affects consumption patterns and responsiveness to advertising. For example, Crockett and Wallendorf (2004) study how the political ideology of African American consumers structures their consumer choice as minorities, and Deshpandé and colleagues (e.g., Deshpandé and Stayman 1994; Grier and Deshpandé 2001) examine how marketers can leverage salient ethnicity among the Mexican American minority in the United States to gain positive responses from this group of consumers. Regarding the effect of product domains, previous research in anthropology, psychology, and marketing has suggested that consumption is used to express desired identities, especially when products are more symbolic (e.g., Belk 1988; Berger and Heath 2007; Cialdini et al. 1976; Levy 1959; Strang 2005; Ustuner and Holt 2007). Identityrelevant products involve, for example, a flag, a university sweatshirt, a national monument, or natural resources (Belk 1988; Berger and Heath 2007; Strang 2005). Natural resources, such as water and land, have often been studied as identity-relevant symbolic products (Alatout 2006; Strang 2005). Of special interest to this study is the link between water consumption and a person’s national iden-

108 / Journal of Marketing, March 2009

tity. The rich literature on water in anthropology, geography, policy, and environmental studies suggests that water, an essential resource to all human beings, can be viewed as a symbolic material that creates different national identities and boundaries (e.g., Alatout 2006; Blatter, Ingram, and Levesque 2001; Menahem 1998; Morag 2001; Strang 2005). In the Middle East, the context of this study, water is a component of economic development and national security and a source of regional conflict (Blatter, Ingram, and Levesque 2001; Le Billon 2001). Indeed, water often stands at the heart of national conflicts and disputes over borders and resources, which in turn emphasizes its centrality to the meaning of national identity (Blatter, Ingram, and Levesque 2001; Le Billon 2001; Morag 2001). Identity-relevant products, such as water, are highly relevant in the case of minorities who want to establish and communicate effectively their social identity (Deshpandé and Stayman 1994). Thus, minority consumers are likely to use consumption as a means to manifest their social identity, beliefs, and goals as minorities, demonstrating their position in relation to the majority group and the government that represents it. Indeed, as we discuss in detail in the next section, minority and majority groups in the Israeli case we study possess different meanings of water and water consumption that are likely to reflect on their national identity and attachment. Furthermore, because selfexpression should be congruent with the group’s beliefs and goals (Berger and Heath 2007; Sirgy 1982), in the case of an identity-relevant marketing effort, such as a government demarketing campaign, and the associated identity-relevant “product”—namely, the protection of national water resources—the self-expressed consumption pattern is most likely to reflect the group’s attitudes toward the government and the country.

Synthesis and Hypotheses We develop a set of hypotheses aimed at understanding the effect of a government proenvironmental demarketing campaign on minority groups versus the majority population, as well as its effect on minority groups with different national attachment levels. In addition, we study the effect of education levels on majority and minority consumers’ responses to demarketing efforts. Education is a central sociodemographic factor, important for understanding and characterizing different minority and proenvironmental consumers (Ergil 2000; Schlegelmilch, Bohlen, and Diamantopoulos 1996). Although our preceding discussion of demarketing and consumers’ proenvironmental behavior offers mixed evidence regarding the merit of demarketing campaigns, taken together with the trend toward environmentalism and the growing number of proenvironmental consumers, we hypothesize that, overall, a government proenvironmental demarketing campaign can positively affect the majority population. First, increasing concerns for the environment and the growing awareness of environmental issues suggest that consumers exposed to a proenvironmental campaign are more motivated than before to adopt the desirable consumption behavior (Pieters et al. 1998). Second, previous research has found that consumers are now more willing to

pay a premium for environmentally friendly products (Menon and Menon 1997; Newell, Goldsmith, and Banzhaf 1998). This implies that consumers may also be willing to pay the costs of sacrificing favored or customary behaviors for a common public goal by deconsuming products that have a negative environmental effect (Pieters et al. 1998; Rangan, Karim, and Sandberg 1996). In the specific case of government demarketing, the motivation underlying the behavior of the majority group can be altruistic and patriotic, involving a true sense of concern about the environment and the public good. Although the motivation may also be egoistic, involving a search for psychological outcomes, such as recognition or a sense of pride, these too represent a strong identification with the idea of contributing to the country (Belk 1988; Luo 2005). These motivations are less likely to be shared by minority groups. Minorities’ sense of national attachment is likely to be lower than that of the majority group, they often hold less positive attitudes toward the government, the country’s goals as reflected by the government campaign are often not congruent with the goals of the minority group, and minorities often dissociate themselves from the majority to strengthen their social identity (Deshpandé and Stayman 1994; Sidanius et al. 1997; Triandis 1995). Thus, we suggest that minorities are likely to manifest their consumption patterns to reflect a less positive response to the government campaign. This conclusion seems especially true in the case of water deconsumption in the Israeli context we study because water takes on different meanings by the majority and minority groups, causing it to be differently associated with the groups’ national attachment. For the majority group of non-ultra-Orthodox Jews, the meaning of water is closely associated with national attachment because it is linked to the process of nation building and establishment of its Jewish identity (Alatout 2006; Azaryahu 2001). Specifically, water symbolizes the ethos of the Halutzim, the Zionist pioneers who established agricultural settlements in Palestine during the nineteenth and early twentieth centuries and who are considered the founders of the Jewish state (Azaryahu 2001; Menahem 1998; Morag 2001). It also symbolizes the country’s economic and demographic development because water and water management played a major role in allowing the absorption of large-scale Jewish immigration and the creation of Jewish demographic presence in arid regions of the state (Menahem 1998; Morag 2001). Finally, water is associated with the country’s technological advancements that have led Israel to be a leader in modern agriculture, water research and management, and desert development (Morag 2001). Conversely, for the largest minority in Israel, the Israeli Arabs, though currently an equal partner for water consumption, water symbolizes the historic, ongoing conflict over water supply and sources between Israel and the surrounding Arab entities, as well as some inequalities in the allocation of water rations for agriculture (Beeri, Carmon, and Shamir 2005; Blatter, Ingram, and Levesque 2001; Ichilov 2003). The Jewish Russian immigrants also do not share the majority’s view of water because they all come from a country that is rich in this resource and are not as strongly attached to the ethos of the

Halutzim and nation building (Carmeli and Fadlon 1997; Leshem and Sicron 1998). Finally, for ultra-Orthodox Jews, some of which are hostile toward the secularist Zionist movement and state, water may take on religious symbolism, but its role in the foundation and development of modern Israel is less meaningful (Stadler 2002; Toft 2002). Thus: H1: Government demarketing of proenvironmental products is more positively related to the deconsumption behavior of majority group consumers than to the deconsumption behavior of minority group consumers.

Different minority groups may possess different national attachment levels because each ethnic group maintains its own particular goals and priorities, cultural characteristics, and attitudes toward the notion of full participation in the dominant society (Sidanius et al. 1997). A central stream of research on the emergence of minority groups and their national attachment discusses different motivations for immigrating. Specifically, immigration motivated by the desire to assimilate in a new country (e.g., immigration driven by political ideology) is expected to lead to higher national attachment than immigration motivated by other reasons (e.g., searching for economic opportunities, joining other family members, escaping economic recession) (Leshem and Sicron 1998; Penaloza 1994). It has also been suggested that some minority groups put more effort than others into increasing involvement in the society through political, economic, cultural, and educational activity. In turn, this is likely to increase national attachment. For example, De la Garza, Falcon, and Garcia (1996) find that Mexican Americans use ethnicity to create resources, such as group solidarity and political organizations, to facilitate their full participation in American society. Finally, different minority groups express their culture and beliefs in varying consumption patterns. For example, while Indian immigrants to the United States often consume products and items that are brought from India, Mexican American consumption patterns tend to be similar on many dimensions to those of Anglo-Americans in the southwestern United States (Penaloza 1994). In turn, consuming local brands and following local consumption patterns are likely to increase the sense of national attachment. The three minority groups in our study—Israeli Arabs, ultra-Orthodox Jews, and Russian immigrants—possess different national attachment levels.1 Israeli Arabs possess the lowest attachment level. First, they differ from both the majority and other minority groups in their religion (they are mostly Muslims and Christians, not Jews). This establishes their position as a subordinate group in Israel, a country that is defined as the “Jewish state” in its declaration of independence, national anthem, and laws (Kretzmer 1990; Toft 2002). Second, many Israeli Arabs have family relations with Palestinians, citizens of the Palestinian Authority in the West Bank and Gaza. Some even affiliate themselves 1Although we contend that, in general, there are significant differences between national attachment levels of the three minority groups and that each group is relatively homogeneous, it is important to note that within each minority group there is a range of opinions and behaviors and that exceptions can be found.

Demarketing, Minorities, and National Attachment / 109

with the Palestinian Authority. This leads to conflicting feelings of commitment to and identification with Israel because of the long-term violent territorial conflict between Palestinians and Israelis (Gordon and Arian 2001; Ichilov 2003). Third, fighting for one’s country is considered a measure of patriotism, and soldiers who fight and sacrifice themselves for the country are an example of the lengths to which people are willing to go in the name of the country and national attachment (Gibson and Abell 2004). In this respect, the vast majority of Israeli Arabs do not join the Israeli Defense Forces (IDF), though Israeli Jews must complete a mandatory military service of two to three years (Toft 2002). Military service is especially important in Israel, where it is perceived as a symbol of good citizenship, shapes civic identities, and reflects differential modes of participation in and belonging to the Israeli state (LomskyFeder and Ben-Ari 1999; Sasson-Levy 2002). Finally, Israeli Arabs reside in their own separate neighborhoods and have developed an education system that is independent of the state (Cohen 2004; Kop 2005). Ultra-Orthodox Jews are second to Israeli Arabs in their low national attachment levels. Although they belong to the Jewish majority and are better institutionalized and linked to the Israeli country and its resources than the Arab minority (Gordon and Arian 2001), this religious group is highly segregated and follows a different agenda than that of the secular Israeli state. Its main goal is the establishment of a religious Jewish state that will have religious education and lifestyle, regulated by this group’s interpretation of the Holy Scripture (Stadler 2002; Toft 2002). The fundamental conflict between secular and ultra-Orthodox Jews sometimes leads to violent acts by ultra-Orthodox extremists, which in turn increases the tension and animosity between these groups (Gordon and Arian 2001; Toft 2002). Many ultraOrthodox Jews do not support the existence of the secular Israeli state and do not fully participate in daily life. Specifically, a large share of ultra-Orthodox men do not work, opting to attend religious school (“yeshiva”) full time, and they typically do not serve in the IDF, despite the mandatory military service imposed on Israeli Jews (Ichilov 2003; Toft 2002). In addition, ultra-Orthodox Jews reside in their own discrete neighborhoods and have developed an education system that is independent of the state (Cohen 2004; Kop 2005). Russian immigrants possess the highest national attachment level among the three minority groups. First, although many of these immigrants migrated to Israel because of fear and economic reasons, escaping waves of anti-Semitism and economic recession in the former Soviet Union, substantial numbers over the years were Zionists and came out of a sense of identification with the Jewish state (Leshem and Sicron 1998). Second, this group has made efforts to participate fully in and assimilate into the Israeli society in the years since their immigration; they contribute to scientific work, entrepreneurial and business activities, the arts, sports, and political life, and they typically serve in the IDF (Carmeli and Fadlon 1997; Lerner and Menahem 2003; Leshem and Sicron 1998). Unlike ultra-Orthodox Jews and Israeli Arabs, they live in mixed neighborhoods and send their children to the state education system (Cohen 2004;

110 / Journal of Marketing, March 2009

Kop 2005). Third, compared with the other minority groups, this group of immigrants is relatively new to the country, the majority of whom arrived in the early 1990s. Thus, they have had less time to make negative inferences about the government and the country. In turn, this is likely to lead to less negative national attachment levels, as reflected through their consumption experiences (Belk 1988; Proshansky 1978). Still, throughout its assimilation process, this group has retained many of its traditions and social arrangements, including its language, establishing a social identity that is different from that of native Israelis (Leshem and Sicron 1998). In addition, compared with the majority of non-ultra-Orthodox Jews, Russian immigrants have more negative attitudes toward the IDF and thus are less willing to join it (Ben-Shalom and Horenczyk 2002; Carmeli and Fadlon 1997). In the context of water, this national resource is equally distributed among all Israeli households, and there are no conflicts over water among the different groups or evidence of significant difference in usage patterns (Beeri, Carmon, and Shamir 2005). However, there is a historic, ongoing conflict regarding water supply and sources between Israel and the surrounding Arab entities (Blatter, Ingram, and Levesque 2001), as well as some inequalities in allocation of water rations for agriculture among Israeli Arabs (Ichilov 2003). This may have an impact only on the Israeli Arabs, contributing to their low national attachment levels and increasing their antagonism toward the notion of protecting Israel’s water resources. Thus: H2: National attachment of the minority group moderates the relationship between government demarketing of proenvironmental products and deconsumption behavior of minority group consumers, such that the higher the national attachment, the stronger is the relationship between government demarketing and deconsumption behavior.

The relationship between minority group association and the response to a government proenvironmental campaign may be moderated by consumers’ level of education. There is evidence that more educated consumers are more likely to be aware of environmental problems and act in an environmentally friendly manner (Schlegelmilch, Bohlen, and Diamantopoulos 1996; Torras and Boyce 1998). These consumers are likely to seek knowledge, are capable of interpreting it and its implications, often critically assess government actions, and are more politically involved (Banerjee, Iyer, and Kashyap 2003; Schlegelmilch, Bohlen, and Diamantopoulos 1996; Torras and Boyce 1998). Educational levels have been linked to greater participation in environmental protection activities, energy conservation, and willingness to pay more for environmental cleanup (Granzin and Olsen 1991). This suggests that both majority and minority consumers with high education levels will be knowledgeable about environmental problems and solutions. Still, it is not clear whether both will be motivated to act in an environmentally friendly manner. Highly educated consumers belonging to the majority group will likely be motivated to contribute to the public good and respond positively to the government proenvironmental demarketing

campaign, but minority consumers are likely to face a dilemma. It has been suggested that the more educated the minority person, the more likely he or she is to be conscious and active in using consumption to manifest self- and social identity (Belk 1988; Sirgy 1982). Thus, these people are more knowledgeable not only about environmental issues but also about their social identity and social, economical, and political goals as a minority group, and they are often the leading activists (Pick and Dayaram 2006). This knowledge might make them the greatest opponents of the government, leading them not to follow the general norm of provisioning public goods. Still, in an environmental context, more educated minority consumers are likely to be aware of the long-term, negative, and sometimes irreversible consequences of a non-environmentally-friendly behavior and of its potential effect on the minority group. In turn, this is expected to lead to responsible behavior. Thus, we suggest that higher education levels, whether among majority or minority consumers, lead to more positive reactions to the demarketing effort: H3: Consumer education moderates the relationship between government demarketing of proenvironmental products and deconsumption behavior among consumer groups with different attachment levels, such that the higher the education, the stronger is the relationship between government demarketing and deconsumption behavior for all groups.

Methods Study Context The study was conducted in Jerusalem, Israel. Israel’s water resources depend mainly on two large equippers and on the Sea of Galilee, which are all interconnected. Natural water resource usage reached its maximum potential in the mid1990s. Both 1999 and 2000 were drought years, and to avoid supply difficulties, the Israeli government needed to find ways to reduce water usage. As the drought developed, the government decided to raise water prices and limit the water supply quotas for agricultural use only. Because of the low price elasticity of household water demand and the regressive effect of a potential rise in household water prices, the government decided to undertake a water savings campaign instead of using the price tool. As we report in greater detail subsequently, the price for household water usage was fixed for the period of interest (1999–2001), and there were no quotas for household water consumption. The large-scale government campaign included massmedia advertisements, an educational program for schools and youth movements, and campaigns aimed at both the majority population and the three central minority groups (i.e., Israeli Arabs, ultra-Orthodox Jews, and Russian immigrants). The government budget invested in the campaign was 11.25 million Israeli shekels ($2.5 million). The main theme was, “Think a drop” (in Hebrew, “drop” is also a small amount). The campaign was “instructional” (Rangan, Karim, and Sandberg 1996), involving instructions about easy things people could do at home to reduce water consumption and frequent reminders to them. The same theme

and spokesperson (a young boy) were used across the different consumer groups, and customization was mostly language based: The campaign aimed at the majority population was in Hebrew, the one aimed at Israeli Arabs was in Arabic, the one aimed at ultra-Orthodox Jews was in Hebrew, and the one aimed at Russian immigrants was in Russian. The campaign involved three intensive efforts, each lasting a month between 1999 and 2001 (winter 1999– 2000, summer 2000, and winter 2000–2001) (Beeri, Carmon, and Shamir 2005). In the natural-experiment setting we examined, it is difficult to conclude that changes in household water consumption are entirely the result of the demarketing campaign. Still, we present several arguments in support of this notion. First, the price of water in Jerusalem was fixed for the entire relevant period (January 1999–December 2001).2 Second, the campaign was widespread and reached the entire Israeli population. Specifically, demarketing efforts were distributed according to the relative size of the targeted groups; the majority of non-ultra-Orthodox Jews received most of the attention, followed by the Israeli Arabs, ultra-Orthodox Jews, and Russian immigrants (Beeri, Carmon, and Shamir 2005). Six exposure surveys of the entire population were undertaken during and after the campaign, confirming high pubic exposure rates (75% on average) (Beeri, Carmon, and Shamir 2005). Third, before the campaign, household water consumption was in a continuous upswing for a decade (Figure 1). Finally, the average temperatures in 1999–2001 were similar (Figure 2). 2Prices included two tariffs: water and sewage charges. Jerusalem used a three-block IBT (increasing block tariff) pricing structure set by the Israeli parliament. In 1999–2001, the average price in the first block, applying to the first 96 cubic meters (m3), was $1.2/m3 including a sewage surcharge. The price in the second block, for additional consumption up to 84m3, was $1.5/m3. The charge for all extra consumption was $1.9/m3. Water prices were not adjusted to inflation automatically. Thus, until it was adjusted in August 2001, the real price of water in Jerusalem actually had been moderately declining.

FIGURE 1 Yearly per Capita Household Water Consumption in Israel: 1992–2002

Source: Israel state budget for 2004.

Demarketing, Minorities, and National Attachment / 111

ment of the same households during the entire period, we focused only on households that remained in the same apartment for the entire period and retained their household size. In addition, we conducted an outlier analysis and removed from the sample observations with a deviation of 40% or more in water consumption (both down and up) between 1999 and 2001 because these can substantially inflate the sample’s standard deviation and affect overall results (Hadi 1992). To demonstrate the robustness of our results, we subsequently conducted an analysis that includes outliers and found that our results do not change. Thus, the original sample, which involves data on water consumption, is labeled Sample 1 and involves 66,272 observations. We then merged Sample 1 with data on household schooling to create Sample 2. The latter data set originates in the municipality’s education department and involves data on the schooling years of parents and their children. Unfortunately, this data set does not contain education data for all households. Rather, it includes only data on households with children in the school system in the last decade (1995–2004). Thus, Sample 2 involves 30,543 observations. Table 1 details the original data, the merging process, and the two samples we use.

FIGURE 2 Average Temperatures in Israel: 1999–2001

Source: The Israeli Central Bureau of Statistics (2005).

Measures We measured household water savings by water savings rate. We measured the decline in household water consumption following the demarketing campaign (in 2001) as a percentage of household water consumption before the campaign (in 1999). We also conducted all the analyses using an alternative dependent variable—namely, water quantity saved between 1999 and 2001 in cubic meters (m3); our results do not change. For minority status, our sample captures Jerusalem’s four main groups of population: the majority of non-ultraOrthodox Jews and the three minority groups: Israeli Arabs, ultra-Orthodox Jews, and Russian immigrants. Those minority groups have their own commercial centers and advertisement channels. Furthermore, the ultra-Orthodox Jewish and Israeli Arab communities have their own neighborhoods. An ultra-Orthodox Jewish household is defined

Sample and Data Collection This study uses disaggregate actual behavior data. The data include information on household-level water consumption and on several other household characteristics in Jerusalem. The city’s diversified household structure makes this data set attractive. It contains enough households to test the effect of demarketing on different minority groups (Israeli Arabs, ultra-Orthodox Jews, Russian immigrants) and on the majority of non-ultra-Orthodox Jews. Data came from three sources: Hagichon (the only water supply company in Jerusalem), the Municipality of Jerusalem, and the Ministry of the Interior. The original data cover water consumption for most households in Jerusalem for the 1999–2001 period. Because of data limitations and to facilitate a longitudinal assess-

TABLE 1 The Data Sets All Populations

Data set for 1999a Data set for 2000a Data set for 2001a Merged data set for 1999–2001 Merged data set without deviation of 20% or more Merged data set without deviation of 40% or more aWe

With Schooling Data

All

The Majority of Non-UltraOrthodox Jews

All

The Majority of Non-UltraOrthodox Jews

103,381 107,973 109,729 79,323 45,168 66,272

71,854 75,374 76,697 57,752 32,851 48,674

44,519 46,543 47,277 34,350 21,704 30,543

29,851 31,367 31,892 24,754 15,404 22,034

excluded observations for several reasons: households with shared meters, households larger than 12 people, and households that metered only part of the year.

112 / Journal of Marketing, March 2009

as such if it is located in an ultra-Orthodox neighborhood, as classified by the Jerusalem master plan (Cohen 2004). Similarly, a household is defined as Israeli Arab if it is located in an Arab neighborhood, as classified by the Jerusalem master plan. The Jerusalem master plan characterizes 200 neighborhoods in the city. This increases the accuracy of this measure. A new immigrant is entitled to reduced property tax for one year during his or her first five years in the country. We define a household as a new immigrant if it received this tax reduction for any year between 1999 and 2001. Most new immigrants in Israel come from the former Soviet Union and speak Russian (Leshem and Sicron 1998). Therefore, we define them as Russian immigrants. As we noted previously, the three minority groups possess different national attachment levels. Russian immigrants are expected to have the highest national attachment levels, followed by ultra-Orthodox Jews and then Israeli Arabs. To validate this ranking, we use data on level of patriotism. Patriotism reflects an important emotional bond with a country and is often used to measure national attachment and identification (Kosterman and Feshbach 1989; Schatz and Staub 1997). Specifically, we use secondary data obtained from a survey of a representative sample of Israeli citizens that was conducted by the Institute for Policy and Strategy at the Interdisciplinary Center of Herzliya (Institute for Policy and Strategy 2007). The survey, titled “Patriotism and National Strength in Israel,” was conducted for the first time in 2005 and then again in 2006. The measure of patriotism involved the question, “Do you consider yourself ‘very much patriotic,’ ‘very patriotic,’ ‘somewhat patriotic,’ or ‘not very patriotic’?” This measure has been widely used in both academic and applied research (e.g., American Enterprise Institute for Public Policy Research 2007; Bennet and Resnick 1990). The findings are consistent across 2005/2006 and suggest that the majority (72%) of non-ultra-Orthodox Jews possess the highest levels of patriotism (i.e., “very much patriotic” and “very patriotic”), followed by the Russian immigrants (64%), the ultraOrthodox Jews (49%), and finally the Israeli Arabs (28%). We measured education as the schooling years of the heads of the household. When there were two adults in the household, we took the maximum figure. In the analysis, we follow previous research in economics (e.g., Lahiri and Xing 2004) and use the dichotomous variable—academic versus nonacademic education—with 13 schooling years as the cutoff point. An alternative measure—average schooling years of the household—yields similar results. In addition, we control for wealth, household size, and lawn ownership. Previous research in marketing and environmental economics has suggested that wealthier consumers are more concerned about the environment and are more active about its preservation (McConnell 1997; Newell, Goldsmith, and Banzhaf 1998). We measured wealth as apartment size. This is a good proxy of household permanent income (Dahan and Nisan 2007). The use of an alternative measure—apartment size per person—yields similar results. We also control for household size as it appears in the Ministry of the Interior data. Finally, we control for lawn ownership because 12% of households in

Jerusalem have irrigated lawns. Lawn size can represent a wealth effect or irrigation decision following the campaign. We introduce two parameters in the regression: a dummy variable for having a lawn and a continuous lawn size variable.

Analysis and Results Descriptive Data Average per-capita water consumption in Jerusalem is 56m3/year, 7% lower than the average (60m3/year) in countries belonging to the Organisation for Economic Cooperation and Development (OECD) (see Figure 3). Table 2 reports descriptive statistics of water savings rates and quantities by different household characteristics for the sample without schooling data (Sample 1). Because the sample with schooling data (Sample 2) shows similar descriptive findings, we discuss only Sample 1 data in detail. The general demarketing effect we find is relatively high—an average of 5.78% household water savings for the entire population. Large differences are evident among the different groups we study: The water savings rate among the Israeli Arab population is very low (.7%), and the water savings rate among ultra-Orthodox Jews (4.4%) and Russian immigrants (5.2%) is less than the majority population of non ultra-Orthodox Jews (6.6%). There is a strong positive correlation between wealth and water savings; households with large apartments save more than households with small apartments (7% and 4.9%, respectively). It is also evident that household size has a negative effect: Large households save less than small households (4.1% and 6.9%, respectively). Lawn ownership and lawn size show a positive influence (2.3% more than an average household and 8.3% savings on average, respectively). An additional relevant finding, based on the descriptive analysis of Sample 2 (with schooling data), suggests that the water savings rate for households with academic education (6%) is higher than for households without academic education (4.8%). Additional descriptive data on the study’s constructs, including the correlation matrix, means, and standard deviations, appear in Table 3. Estimated Model To estimate the effect of the demarketing campaign, we compared household water consumption before and after the campaign. We ran a model using ordinary least squares regression for the dependent variable—rate of household water consumption decline between the years 1999 and 2001. The model is as follows: (1)

Δy i = α + βx i + ε i , y99

where Δyi is the difference in household i’s water consumption between 1999 and 2001 and xi denotes a vector of household characteristics as documented in 1999, which is constant until 2001. There are four groups of variables: minority status, education level, wealth indicators (apart-

Demarketing, Minorities, and National Attachment / 113

FIGURE 3 Average per Capita Water Consumption in Israel and Other OECD Countries

Source: OECD (1999).

TABLE 2 Descriptive Statistics: Water Savings by Household Characteristics (Data Set Without Schooling Data: Sample 1) Water Consumption 1999

Water Consumption 2001

Decrease in Consumption

Rate of Decrease

Number of Observations

All populations

227.94

212.34

15.61

5.78%

66,271

Consumer Groups Non-ultra-Orthodox Jews Israeli Arabs Ultra-Orthodox Jews Russian immigrants

212.39 322.81 254.48 164.82

195.40 317.03 240.79 153.86

16.99 5.78 13.69 10.96

6.60% .70% 4.40% 5.20%

48,674 4,242 13,355 299

Wealth (Apartment Size)a Large apartment size Small apartment size Apartment size: missing data

274.96 183.76 220.02

253.40 173.26 207.02

21.57 10.49 13.00

7.00% 4.94% 4.58%

28,622 28,886 8,763

Control Variables Large householdsb Small households Lawn owners Large lawnsc Small lawns

296.39 182.06 341.46 378.08 292.10

280.12 166.90 310.98 344.02 266.46

16.28 15.16 30.47 34.07 25.63

4.12% 6.89% 8.06% 8.29% 7.75%

26,595 39,676 9,588 5,504 4,084

aLarge bLarge cLarge

is defined as more than the median (78 square meters). is defined as four members or more. is defined as more than the median (100 square meters).

ment size, lawn ownership), and household size. Regarding minority status, the regression includes three dummy variables to represent the difference in the campaign effec114 / Journal of Marketing, March 2009

tiveness on the majority population (non-ultra-Orthodox Jews only) compared with the minority groups: Israeli Arabs, ultra-Orthodox Jews, and Russian immigrants.

TABLE 3 Correlation Matrix and Descriptive Statistics of the Study’s Constructs Constructs Water savings rate Household size Wealth (apartment size) Apartment size: missing data Lawn owners Lawn size Academic education Ultra-Orthodox Jews Israeli Arabs Russian immigrants M SD

1 1 –.12** .07** –.01* .07** .06** .04** –.05** –.06** –.01 .05 .17

2

3

4

5

1 –.03** –.01** .01** –.06** .01* .04**

1 .72** .06** –.10** .02** –.01*

.07 .26

.18 .39

6

7

8

9

.19** –.14** .02**

1 –.11** –.01

1 –.01

.55 .50

.24 .43

.04 .19

10

1 .19** –.13** .07** .03** .05** .34** .13** –.03** 4.43 2.09

1 –.62** .28** .26** .14** .06** –.05** –.03** 81.40 36.89

1 .03** –.11** .08** –.01

1

23.97 70.27

1 .00 .05

*p < .05. **p < .01.

Results The effect of demarketing on majority and minority groups. Table 4 reports the coefficients and p-values for the estimated model. We present the estimates for the model for both samples (Samples 1 and 2). Based on the analysis of Sample 1 (the data set without schooling data), the demarketing effect for an average nonultra-Orthodox Jewish household, with average household size and wealth and without a lawn, is approximately 6% of the yearly household water consumption.3 All the minority groups responded less positively to the campaign: Israeli Arabs were less likely to save water (–4.7%, p = .0001), followed by ultra-Orthodox Jews (–1%, p = .0001). Although the coefficient for Russian immigrants is also negative (–1.4%, p = .1831), it is not significant, suggesting that this group responded more positively to the campaign than the other minority groups. We also conduct a series of t-tests to confirm the differences between the estimators and find that

non-ultra-Orthodox Jewish households reacted significantly more positively to the campaign than the ultra-Orthodox Jews (t-value = 5.47, p < .0001), who responded more posi-

TABLE 4 Campaign Effectiveness Dependent Variable: Water Savings Rate Sample 1 Intercept Household size Wealth (apartment size) Apartment size: missing data Lawn owners

3To calculate the demarketing effect for the typical non-ultraOrthodox Jewish household, we undertake the following procedure: Our point of departure is the regression intercept, which represents the baseline effect on the average household in the entire population. We then add to the intercept two multiplicative terms that refer to the non–dummy variables included in the regression: household and apartment size (specifically, for non-ultra-Orthodox Jewish households, we multiply their average household size [three family members] and average apartment size [80 square meters] by their estimates and add these terms to the intercept). Overall, the calculation (see Table 4) yields a demarketing effect slightly above 6%: .054 + 3 × (–.008) + 80 × (.0004) ≈ 6%. To calculate the demarketing effect for all other populations (Israeli Arabs, ultra-Orthodox Jews, Russian immigrants), it is necessary to add to the intercept (5.4%) the two multiplicative terms that refer to household and apartment size (multiplying the relevant regression coefficients by the specific group’s average household size and average apartment size). To calculate the demarketing effect for households with lawns, whether related to the majority or minority groups, it is necessary to add to the overall demarketing effect of the group (e.g., approximately 6% in the case of nonultra-Orthodox Jewish households) the lawn size dummy coefficient (2.6%).

Lawn size Academic education Ultra-Orthodox Jews Ultra-Orthodox Jews × academic education Israeli Arabs Israeli Arabs × academic education Russian immigrants Russian immigrants × academic education F-value Adjusted R2 Number of observations

.054** (.0001) –.008** (.0001) .0004** (.0001) .022** (.0001) .026** (.0001) –.00003* (.0500) — –.010** (.0001) — –.047** (.0001) — –.014 (.1831) — 175.9** .02 66,271

Sample 2 .056** (.0001) –.010** (.0001) .0004** (.0001) .019** (.0001) .026** (.0001) –.00002 (.4347) .010** (.0001) .002 (.6954) –.009 (.0534) –.034** (.0001) –.001 (.9339) .021 (.5961) –.061 (.1901) 76.5** .03 30,543

*p < .05. **p < .01. Notes: p-values are in parentheses.

Demarketing, Minorities, and National Attachment / 115

tively than the Israeli Arabs (t-value = 12.02, p < .0001). Still, the Russian immigrants responded more positively to the campaign than the other minority groups because the estimator of this group is not significantly different from that of non-ultra-Orthodox Jews (t-value = 1.33, p > .1). In general, these findings support H1 and H2. Education. From our analysis of Sample 2, which incorporates schooling data, we conclude that households with academic education save an average of 1% more than households without academic education (p = .0001). This implies approximately 10% more water saving quantities or campaign effectiveness than households with lower education levels. In the case of the interaction of education with different minority groups, we find that the effects among Israeli Arabs (–.1%, p = .9339) and Russian immigrants (–6.1%, p = .1901) are nonsignificant, suggesting, as we expected, that the effect of education on these groups is similar to that of the entire population. That is, these minority groups and the majority population are affected similarly by education. The interaction effect of education with the ultra-Orthodox Jews was significant and negative (–.09%, p = .0534). The negative sign estimator implies that the education effect on a non-ultra-Orthodox Jewish household does not affect the ultra-Orthodox group. As we explain in greater detail in the “Discussion” section, this is likely the result of the unique, highly focused nature of education in this group (religious studies). This negative interaction effect, which contributes to the insignificant main effect of association with this population on the campaign effectiveness, demonstrates how this group differs from the entire population. Overall, therefore, these findings partially support H3. Controls. The estimate for apartment size, which reflects wealth, is .04% per square meter. This reflects a positive correlation between demarketing effects and level of wealth. Each 25 square meters adds 1% more campaign effectiveness, which implies 16% more than an average household. We also find that household size is significant and negative in all models, implying a negative effect on water savings, most likely because of the limited impact on children. Finally, the lawn ownership dummy variable is significant and positive in both models, reflecting a 2.6% additional demarketing wealth effect. Lawn size is significant only in the first model and can reflect either an additional wealth effect or irrigation needs in 2001 compared with 1999. Sensitivity analysis. We conducted several sensitivity analyses to demonstrate the robustness of our findings. We first attempted to rule out a main alternative explanation to our findings—namely, that the different responses to the demarketing effort were the result of variation in exposure to the campaign by different groups rather than the result of the groups’ different national attachment levels. Thus, we differentiated between two subgroups belonging to the same minority population on the basis of association with the education system—state versus independent. Because the two subgroups were likely to be similarly exposed to the demarketing campaign in terms of language and media exposure, we examined whether the different education

116 / Journal of Marketing, March 2009

associations affected their deconsumption patterns. We use association with the state education system (relative to an independent system) because this can explain higher national attachment levels following institutionalization and indoctrination of national values, traditions, and admiration (Ergil 2000). Because we have relevant and detailed schooling data only for Israeli Arabs, we added to our model a dummy representing whether Israeli Arabs send their children to the state education system or to an independent system. Our findings (Table 5) suggest that households with children in the Israeli education system were more likely to deconsume water following the campaign (–2.4%, p = .0047) than households with children in the independent education system (–4.3%, p = .0001), in support of the notion that national attachment plays an important role in TABLE 5 Ruling out Variations in Exposure to the Campaign as an Alternative Explanation Dependent Variable: Water Saving Rates (Sample 2) .056* Intercept .056* (.0001) (.0001) –.010* Household size –.010* (.0001) (.0001) .0004* Wealth (apartment size) .0004* (.0001) (.0001) .019* Apartment size: missing data .019* (.0001) (.0001) .026* Lawn owners .026* (.0001) (.0001) –.00002 Lawn size –.00002 (.4076) (.4347) .010* Academic education .010* (.0001) (.0001) .002 Ultra-Orthodox Jews .002 (.6909) (.6954) –.009 Ultra-Orthodox Jews × academic –.009 (.0534) education (.0534) — Israeli Arabs –.034* (.0001) — Israeli Arabs × academic –.001 education (.9339) –.024* Israeli Arabs in the Israeli — (.0047) education system –.017 Israeli Arabs in the Israeli education — (.3333) system × academic education –.043* Israeli Arabs not in the Israeli — (.0001) education system .015 Israeli Arabs not in the Israeli — (.4241) education system × academic .021 education Russian immigrants .021 (.5964) (.5961) –.061 Russian immigrants × academic –.061 (.1901) education (.1901) F-value Adjusted R2 Number of observations *p < .01. Notes: p-values are in parentheses.

76.5* .03 30,543

65.8* .03 30,543

explaining minority groups’ reaction to government demarketing.4 We also want to confirm that the different groups studied do not possess social networks that systematically differ in their structure and strength, because this may also explain variability in reception of the campaign. Thus, we collected data on utilization of social networks (i.e., frequency of communication with friends and family) of nonultra-Orthodox Jews, ultra-Orthodox Jews, and Israeli Arabs from the Israeli Central Bureau of Statistics (2002). Overall, the data do not indicate any systematic difference among the groups. For example, while the ultra-Orthodox Jews talk more frequently with friends than do the other groups, the Israeli Arabs more frequently meet with family members.5 Furthermore, the data suggest that the majority of non-ultra-Orthodox Jews’ utilization of social networks is not stronger than those of ultra-Orthodox Jews and Israeli Arabs. Because the former reacted more positively to the

campaign than the latter two groups, this provides some indication that the variability in reception of the campaign is not driven by social networks. Because our measure of wealth (i.e., apartment size) might not be satisfactory as a result of differences in household size, we also measured it using apartment size per person. In addition, we ran an ordinary least squares regression using the original measure, this time for households of four people only. In general, the estimates remain significant and similar (Table 6). Finally, in the early analysis, we excluded outlier observations (i.e., households that showed more than 40% change in water consumption in 2001 compared with 1999). Because this was an ad hoc cutoff point, we also ran the model after omitting outlier observations using a 20% cutoff point. Again, this analysis does not change the nature of our results (Table 6).

Conclusions and Implications 4The difference between the two Israeli Arab groups is significant at the 10% level (t-value = 1.73). 5Frequency of communication with friends and family involved the measures “I talk/meet with my friends/family every day/almost every day.” In the case of talking with friends, first in line were ultra-Orthodox Jews (49.8%), then non-ultra-Orthodox Jews (41.4%), and finally Israeli Arabs (28.6%). In the case of meeting with family, first in line were Israeli Arabs (36.8%), then ultraOrthodox Jews (24.8%), and finally non-ultra-Orthodox Jews (24.5%).

This study addresses two important trends: protection of public goods and the emergence of multiethnic societies that involve influential minority groups. We were able to accurately assess the effect of government demarketing on different minority groups because of our use of actual water consumption data instead of self-reported attitudes and because no other policy tool was employed during the examined period. The study offers implications for marketing scholars, policy makers, and marketers.

TABLE 6 Sensitivity Analysis Dependent Variable: Water Savings Rate Apartment Size per Person Intercept Household size Wealth (apartment size) Apartment size: missing data Lawn owners Lawn size Ultra-Orthodox Jews Israeli Arabs Russian immigrants F value Adjusted R2 Number of observations

.076** (.0001) –.006** (.0009) .000123** (.0013) –.006* (.0189) .032** (.0001) –.000 (.5437) –.011** (.0001) –.052** (.0001) –.013 (.2195) 147.4** .02 66,271

Apartment Size (Four People per Household) .013* (.0302) — .000069** (.0001) .023** (.0039) .029** (.0001) –.000 (.0515) –.009 (.0950) –.040** (.0001) –.050 (.1429) 25.0** .02 8968

Population After Removing 20% Deviation or Less .024** (.0001) –.003** (.0001) .00020** (.0001) .011** (.0001) .011** (.0001) –.000 (.4057) –.007** (.0001) –.027** (.0001) –.01 (.1431) 68.0** .01 45,168

*p < .05. **p < .01. Notes: p-values are in parentheses.

Demarketing, Minorities, and National Attachment / 117

The notion that minority consumers respond favorably to marketing efforts targeted at their minority status using ethnically congruent stimuli is intuitive and has been supported empirically (e.g., Deshpandé and Stayman 1994; Grier and Deshpandé 2001). However, we find that when the marketer is the government and the product is a public good, this notion does not apply. Specifically, we find that majority and minority consumers behave differently with a public good, especially one that is more symbolic. In this respect, the tension between national and group identification is likely to be an important force explaining behavior. Thus, policy makers and marketers should make an effort to identify the reference group of the different minorities: the country or their in-group (Roccas, Klar, and Liviatan 2006; Triandis 1995). When the reference group is not the country and minorities may believe that what is good for their country is not necessarily good for their group, it might prove useful to reconfigure a demarketing campaign so that it is not linked to the benefits of the country but rather to the personal benefits of the minority consumer. The finding that the reaction among Israeli Arabs to the campaign was more negative than the other two minority groups we studied emphasizes their divergence from the core Israeli society and its central goals. Although there are many differences between this group and the majority of the Jewish population, there are many common goals as well, including regional peace, economic prosperity, and protection of the natural environment. Policy makers and marketers should identify and then emphasize such similarities and mutual goals among different ethnic groups as a means to help decrease interethnic tension and improve the effectiveness of government marketing efforts. An important issue is that of generalizability. Although Israel represents a unique case in which ethnicity and religion are interrelated to some extent, we believe that our findings can be generalized to other countries in which minority groups play an important role. Note that though religion is central in shaping Israelis’ identity, the different minority groups in Israel are not classified only on the basis of their religion. Specifically, in our study we capture three important facets of minorities—religion, ethnicity, and immigration status—and compare Muslims (and some Christians) and Jews, groups of Jews with different ethnic backgrounds (Russian immigrants and ultra-Orthodox Jews), and immigrants and nonimmigrants. Our findings show that different minority characteristics affect demarketing effectiveness through national attachment. Thus, the findings can be related to minorities in different countries in which minority status is dictated by these forces and issues of national identity and attachment significantly influence minorities’ behavior in general and their consumption in particular. This would be the case for religious and ethnicbased minority immigrants, such as the North African Muslims in France and Mexican Americans in the United States, as well as for religious and ethnic-based minority nonimmigrants, such as the Basques in Spain and African Americans in the United States. Indeed, recent research on African Americans and their consumption patterns suggests that political ideology, in which issues of national attachment play an important role, is central to understanding shopping

118 / Journal of Marketing, March 2009

and consumption as an expression of social and political identities of different consumer groups (Crockett and Wallendorf 2004). The general demarketing effect we find is relatively high—an average of 5.78% household water savings for the entire population. This finding contributes to the debate in marketing regarding the effectiveness of demarketing efforts, at least in the context of proenvironmental products (Deutsch and Liebermann 1985; Pechmann et al. 2003). The validity of our results is enhanced because the “product” examined is household water consumption, an item with extremely low demand elasticity (Deutsch and Liebermann 1985). Furthermore, in an additional analysis, we examine the unique contribution of demarketing on the different studied groups beyond the various control variables. Similar to previous demarketing studies (e.g., Andrews et al. 2004), we find that demarketing has a significant but incremental contribution, explaining an additional 4.2% of the variance in household consumption. The effect of education on different groups’ responses to the campaign is noteworthy. First, it suggests that higher education levels positively affect environmentally friendly behavior. Second, it suggests that education can reduce the negative response of minority groups. In the case of ultraOrthodox Jews, we find a negative interaction effect. The reason is that high education levels in this group are associated with religious studies that are likely to limit this group’s exposure to other types of knowledge, specifically knowledge about environmental and ecological issues. We find that the less wealthy households reacted more negatively to the campaign. Although we cannot rule out that the special features of the proenvironmental campaign caused this difference, it may be a more general result of the response to the demarketing efforts. This is an additional indication of both the benefits and the limitations of demarketing on different sectors of the population. Another example is the finding that household size has a negative effect in all models, reflecting the difficulties in using demarketing to influence children’s behavior. Our results indicate that several groups, classified according to minority and socioeconomic status, reacted less positively to a national demarketing campaign. This raises important questions for policy makers and social marketers: How much investment, if any, should be directed toward customizing campaigns to these groups? What is the importance of targeting audiences on the basis of socioeconomic criteria (e.g., wealth, education) versus minority status? Our results emphasize the importance of minority targeting. For example, in the case of Israeli Arabs, more than 50% of their nonresponsive reaction to the campaign can be attributed to their minority status. Although some may argue that, based on our results, a viable option for campaign strategists is to ignore audiences that are less likely to respond to the marketing effort, we believe that from both sociopolitical and economic standpoints, a more appropriate approach would be to focus attention also on minorities and invest, at least to some degree, in campaign customization. Although demarketing campaigns are costly, the use of alternative approaches to reduce water consumption seems costlier. For example, employing the water price tool to

obtain the same water savings rate would have caused a significant regressive effect. To reduce water consumption by nearly 6%, water prices would have had to be raised by more than 35%, if we assume that the elasticity of demand in Jerusalem is –.17 (Dahan and Nisan 2007). Furthermore, investments in new desalination plants are high. Thus, the water savings potential from a demarketing effort should be taken into account when planning the water supply system and designing investment plans. The general picture that emerges from this research should encourage policy designers and regulators to make more use of demarketing, combining it with other policy tools. Although the relatively clear approach of forecasting economic effects would seem to encourage the use of direct economic tools (e.g., price), the demarketing effect we show herein can shed light on the potential contribution of demarketing to policy makers. Furthermore, a better understanding of the way different consumer groups are willing to participate in environmental protection activities will make it possible to develop and target appropriate policies and demarketing efforts at specific consumer segments to elicit behavioral change (Granzin and Olsen 1991). Finally, the results we report might provide a benchmark for demarketing effects in other environmental contexts and for other products.

Limitations and Future Research Directions Our findings elicit several possible alternative explanations that should be addressed. First, we need to rule out the explanation that the different responses to the demarketing effort were due to variation in exposure to the campaign by different groups rather than to the groups’ different national attachment levels. We first show that this is not the case by controlling for several factors that might create variability in reception of the campaign (e.g., household size). We then rule out this alternative explanation by studying one of the minority groups—Israeli Arabs—for which we had relevant and detailed education data. Specifically, although all Israeli Arabs were likely to be similarly exposed to the campaign, their reactions varied according to association with the education system. Indeed, Israeli Arabs who send their children to the state education system, and therefore are likely to have higher national attachment levels, responded more positively to the government campaign than those who send their children to the independent system. Finally, we collected data on utilization of social networks by the different minority groups and ensured that there was no significant systematic difference among the groups that might have affected reception of the campaign. Second, it is important to rule out the alternative explanation that minorities respond negatively to proenvironmental campaigns because they typically have lower levels of environmental beliefs and values than majority populations. After reviewing prior research on environmentalism and ethnicity, we contend that this may not be the case. Specifically, a review of this issue concludes that minority and majority populations have similar environmental beliefs and

values (Johnson, Bowker, and Cordell 2004). Furthermore, a 2006 survey on Israelis’ concerns about environmental issues conducted by the Pratt Foundation and Heschel Center for Environmental Learning and Leadership shows that Jews and Israeli Arabs share similar concerns about various environmental topics, such as air pollution, global warming, depletion of natural resources, noise, and water pollution.6 Overall, this provides some indication that national attachment may play a more important role than consumers’ environmental beliefs and values. This study suffers from additional limitations that might suggest future research opportunities. First, we studied actual consumers’ behavior, not their motivations and attitudes. Although there are substantial benefits to the use of behavioral data and though a growing number of researchers in marketing recommend studying actual behavior of consumers (e.g., Rangan, Karim, and Sandberg 1996; Schlegelmilch, Bohlen, and Diamantopoulos 1996), this approach also has its weaknesses in our context. Our behavioral data can inform us about what different consumer groups are actually doing but may be limited in explaining the reasons for such behavior. Indeed, we cannot firmly conclude that majority and minority consumers think differently about public goods, such as water, or that different consumers identify similarly with their minority or majority in-group. Some prior research on ethnic groups has indicated that there may be variation within minority groups regarding identification and attachment of individuals to their group (e.g., Deshpandé and Stayman 1994; Penaloza 1994) and that some minority individuals may simultaneously identify with their in-group and with the majority group (Sidanius et al. 1997). Further research might be valuable here, complementing behavioral data with data based on survey and ethnographic research that will generate deeper insights into the mechanism that drives the behavior of majority and minority consumers. For example, research efforts in these directions may benefit from prior ethnographic work on the underlying motivations of ethnic consumer groups (e.g., Penaloza 1994, 1995) and from survey research on the motivations driving environmental consumers (e.g., Pieters et al. 1998). Still, note that though these research methods can provide insights into the motivations and attitudes of consumers and consumer groups, their main weakness is that they may not be able to predict actual behavior. Second, the study was conducted in a specific country and product context. Although Israel is similar to many other developed countries and though water scarcity is a global environmental concern, the specific nature of the case might limit the general applicability of the findings. It might be useful to conduct cross-country research on demarketing for different environmental products and additional cultural groups. Third, we did not study the specific characteristics of the demarketing campaign in depth. Because different themes and framings can differently affect a demarketing campaign’s effectiveness (Pechmann et al. 2003), further research should also control for

6See

http://www.heschel.org.il/Media/seker06.pdf.

Demarketing, Minorities, and National Attachment / 119

these factors. Finally, the effect of demarketing over time is not clear. In this study, it was difficult to test for the longterm effect of the demarketing campaign because substantial price changes in late 2001 made it impossible to isolate

the campaign’s effect. Further research is warranted in this direction (Deutsch and Liebermann 1985)—for example, whether the less positive response among minority groups is stable over time.

REFERENCES Aaker, Jennifer (1999), “The Malleable Self: The Role of SelfExpression in Persuasion,” Journal of Marketing Research, 36 (February), 45–57. Alatout, Samer (2006), “Towards a Bio-Territorial Conception of Power: Territory, Population, and Environmental Narratives in Palestine and Israel,” Political Geography, 25 (6), 601–621. American Enterprise Institute for Public Policy Research (2007), “Polls on Patriotism and Military Service,” (accessed November 15, 2007), [available at http://www.aei.org/publications/ pubID.14889/pub_detail.asp]. Andrews, Craig, Richard Netemeyer, Scot Burton, Paul Moberg, and Ann Christiansen (2004), “Understanding Adolescent Intentions to Smoke: An Examination of Relationships Among Social Influence, Prior Trial Behavior, and Antitobacco Campaign Advertising,” Journal of Marketing, 68 (July), 110–23. Askegaard, Søren, Eric Arnould, and Dannie Kjeldgaard (2005), “Postassimilationist Ethnic Consumer Research: Qualifications and Extensions,” Journal of Consumer Research, 32 (1), 160–70. Azaryahu, Maoz (2001), “Water Towers: A Study in the Cultural Geography of Zionist Mythology,” Cultural Geographies, 8 (3), 317–39. Baker, William and James Sinkula (2005), “Environmental Marketing Strategy and Firm Performance: Effects on New Product Performance and Market Share,” Journal of the Academy of Marketing Science, 33 (4), 461–75. Balabanis, George, Adamantios Diamantopoulos, Rene Muller, and T.C. Melewar (2001), “The Impact of Nationalism, Patriotism, and Internationalism on Consumer Ethnocentric Tendencies,” Journal of International Business Studies, 32 (1), 157–75. Banerjee, Subhabrata, Easwar Iyer, and Rajiv Kashyap (2003), “Corporate Environmentalism: Antecedents and Influence of Industry Type,” Journal of Marketing, 67 (April), 106–122. Beeri, Shlomit, Naomi Carmon, and Uri Shamir (2005), Water Saving in the Urban Sector: A Feasibility Study. Haifa, Israel: The Technion Institute for Research and Development. Belk, Russel (1988), “Possessions and the Extended Self,” Journal of Consumer Research, 15 (September), 139–68. Bennet, Stephen and David Resnick (1990), “The Implications of Nonvoting for Democracy in the United States,” American Journal of Political Science, 34 (3), 771–802. Ben-Shalom, Uzi and Gabriel Horenczyk (2002), “Cultural Identity and Motivation to Become an Officer,” paper presented on the First Inter University Seminar of Armed Forces and Society, Kingtston, Canada (October 25–27). Berger, Jonah and Chip Heath (2007), “Where Consumers Diverge from Others: Identity Signaling and Product Domains,” Journal of Consumer Research, 34 (2), 121–34. Blatter, Joachim, Helen Ingram, and Suzanne Levesque (2001), “Expanding Perspectives on Transboundary Water,” in Reflection on Water, Joachim Blatter and Helen Ingram, eds. Cambridge, MA: The MIT Press, 31–53. Carmeli, Abraham and Judith Fadlon (1997), “Motivation to Serve in the Israeli Army: The Gap Between Cultural Involvement and Cultural Performance,” in Russian Jews in Three Continents: Migration and Resettlement, Noah Lewin-Epstein, Yaacov Roi, and Paul Ritterband, eds. London: Frank Cass, 389–405. Cialdini, Robert, Richard Borden, Avril Throne, Marcus Walker, Stephen Freeman, and Lloyd Sloan (1976), “Basking in

120 / Journal of Marketing, March 2009

Reflected Glory: Three (Football) Field Studies,” Journal of Personality and Social Psychology, 34 (September), 366–75. Cohen, Moshe (2004), Jerusalem Master Plan. Jerusalem: The Municipality of Jerusalem. Crockett, David and Melanie Wallendorf (2004), “The Role of Normative Political Ideology in Consumer Behavior,” Journal of Consumer Research, 31 (3), 511–28. Dahan, Momi and Udi Nisan (2007), “Unintended Consequences of Increasing Block Tariffs Pricing Policy in Urban Water,” Water Resources Research, 43 (March), W03402. De la Garza, Rodolfo, Angelo Falcon, and Chris Garcia (1996), “Will the Real American Please Stand Up: Anglo and MexicanAmerican Support of Core American Political Values,” American Journal of Political Science, 40 (2), 335–51. Deshpandé, Rohit and Douglas Stayman (1994), “A Tale of Two Cities: Distinctiveness Theory and Advertising Effectiveness,” Journal of Marketing Research, 31 (February), 57–64. Deutsch, Joseph and Yehoshua Liebermann (1985), “Effects of a Public Advertising Campaign on Consumer Behavior in a Demarketing Situation,” International Journal of Marketing Research, 2 (4), 287–96. Duckitt, John (1989), “Authoritarianism and Group Identification: A New View of an Old Construct,” Political Psychology, 10 (1), 63–84. Ergil, Dogu (2000), “Identity Crises and Political Instability in Turkey,” Journal of International Affairs, 54 (1), 43–62. Frederick, Shane, George Loewenstein, and Ted O’Donoghue (2002), “Time Discounting and Time Preference: A Critical Review,” in Time and Decision: Economic and Psychological Perspectives on Intertemporal Choice, George Loewenstein, Daniel Read, and Roy Baumeister, eds. New York: Russell Sage Foundation Press, 13–86. Gaertner, Lowell and Chester Insko (2000), “Intergroup Discrimination in the Minimal Group Paradigm: Categorization, Reciprocation, or Fear?” Journal of Personality and Social Psychology, 79 (1), 77–94. Gerstner, Eitan, James Hess, and Wujin Chu (1993), “Demarketing as a Differentiation Strategy,” Marketing Letters, 4 (1), 49–57. Gibson, Stephen and Jackie Abell (2004), “For Queen and Country? National Frames of Reference in the Talk of Soldiers in England,” Human Relations, 57 (7), 871–91. Gordon, Carol and Asher Arian (2001), “Threat and Decision Making,” Journal of Conflict Resolution, 45 (2), 196–215. Granzin, Kent and Janeen Olsen (1991), “Characterizing Participants in Activities Protecting the Environment: A Focus on Donating, Recycling, and Conservation Behaviors,” Journal of Public Policy & Marketing, 10 (Fall), 1–27. Grier, Sonya and Rohit Deshpandé (2001), “Social Dimensions of Consumer Distinctiveness: The Influence of Social Status on Group Identify and Advertising Persuasion,” Journal of Marketing Research, 38 (May), 216–24. Grunert, Suzanne (1993), “Everybody Seems Concerned About the Environment, but Is This Concern Reflected in (Danish) Consumers’ Food Choice?” European Advances in Consumer Research, 1, 428–33. Hadi, Ali (1992), “Identifying Multiple Outliers in Multivariate Data,” Journal of the Royal Statistical Society, 54 (3), 761–71. Ichilov, Orit (2003), “Teaching Civics in a Divided Society: The Case of Israel,” International Studies in Sociology of Education, 13 (3), 219–42.

Institute for Policy and Strategy (2007), “Patriotism and National Strength in Israel,” (accessed November 15, 2007), [available at http://www.herzliyaconference.org/Eng/_Articles/Article. asp?CategoryID=31&ArticleID=1892]. The Israeli Central Bureau of Statistics (2002), “Social Survey,” (accessed November 23, 2007), [available at http://www.cbs. gov.il/reader/?MIval=cw_usr_view_SHTML&ID=569]. ——— (2005), Statistical Abstract of Israel. Jerusalem: The Israeli Central Bureau of Statistics. Johnson, Cassandra, Michael Bowker, and Ken Cordell (2004), “Ethnic Variation in Environmental Belief and Behavior: An Examination of the New Ecological Paradigm in a Social Psychological Context,” Environment and Behavior, 36 (2), 157–85. Kasulis, Jack, David Huettner, and Neil Dikeman (1981), “The Feasibility of Changing Electricity Consumption Patterns,” Journal of Consumer Research, 8 (3), 279–90. Kelly, Kathleen, Randall Swaim, and Jeffery Wayman (1996), “The Impact of a Localized Antidrug Media Campaign on Targeted Variables Associated with Adolescent Drug Use,” Journal of Public Policy & Marketing, 15 (Fall), 238–53. Kop, Yaacov, ed. (2005), Israel Social Services. Jerusalem: Taub Center for Social Policy Studies. Kosterman, Rick and Seymour Feshbach (1989), “Towards a Measure of Patriotic and Nationalistic Attitudes,” Political Psychology, 10 (2), 257–74. Kotler, Philip and Sidney Levy (1971), “Demarketing, Yes, Demarketing,” Harvard Business Review, 49 (November– December), 74–80. Kretzmer, David (1990), The Legal Status of the Arabs in Israel. Boulder, CO: Westview. Lahiri, Kajal and Guibo Xing (2004), “An Econometric Analysis of Veterans’ Health Care Utilization Using Two-Part Models,” Empirical Economics, 29 (2), 431–49. Le Billon, Philippe (2001), “The Political Ecology of War: Natural Resources and Armed Conflicts,” Political Geography, 20 (5), 561–84. Lerner, Miri and Gila Menahem (2003), “Decredentialization and Recredentialization: The Role of Governmental Intervention in Enhancing Occupational Status of Russian Immigrants in Israel in the 1990s,” Work and Occupations, 30 (1), 3–29. Leshem, Elazar and Mosheh Sicron (1998), “The Absorption of Immigrants from the Former Soviet Union, 1990-1995: Main Findings,” in Profile of an Immigration Wave: The Absorption Process of Immigrants from the Former Soviet Union, 19901995, Elazar Leshem and Mosheh Sicron, eds. Jerusalem: The Magness Press, 442–62. Levy, Sidney (1959), “Symbols for Sale,” Harvard Business Review, 33 (May–June), 117–24. Ling, Yeh (2004), “Mexican Immigration and Its Potential Impact on the Political Future of the United States,” Journal of Social, Political, and Economic Studies, 29 (4), 409–431. Lomsky-Feder, Edna and Eyal Ben-Ari (1999), “From ‘The People in Uniform’ to ‘Different Uniforms for the People’: Professionalism, Diversity, and the Israeli Defense Forces,” in Managing Diversity in the Armed Forces: Experiences from Nine Countries, Joseph Soeters and Jan van der Meulen, eds. Tilburg, the Netherlands: Tilburg University Press, 157–86. Luo, Xueming (2005), “A Contingent Perspective on the Advantages of Stores’ Strategic Philanthropy for Influencing Consumer Behavior,” Journal of Consumer Behavior, 4 (5), 390–401. McConnell, Kenneth (1997), “Income and the Demand for Environmental Quality,” Environment and Development Economics, 2 (4), 383–93. McDougall, Gordon, John Claxton, Brent Ritchie, and Dennis Anderson (1981), “Consumer Energy Research,” Journal of Consumer Research, 8 (3), 343–53.

Menahem, Gila (1998), “Policy Paradigms, Policy Networks and Water Policy in Israel,” Journal of Public Policy, 18 (3), 283–310. Menon, Ajay and Anil Menon (1997), “Enviropreneurial Marketing Strategy: The Emergence of Corporate Environmentalism as Market Strategy,” Journal of Marketing, 61 (January), 51–67. Morag, Nadav (2001), “Water, Geopolitics and State Building: The Case of Israel,” Middle Eastern Studies, 37 (3), 179–98. Newell, Stephen, Ronald Goldsmith, and Edgar Banzhaf (1998), “The Effect of Misleading Environmental Claims on Consumer Perceptions of Advertising,” Journal of Marketing Theory and Practice, 6 (2), 48–60. OECD (1999), The Price of Water: Trends in OECD Countries. Paris: OECD. Osterhus, Thomas (1997), “Prosocial Consumer Influence Strategies: When and How Do They Work?” Journal of Marketing, 61 (October), 16–29. Pechmann, Cornelia, Guangzhi Zhao, Marvin Goldberg, and Ellen Reiblinget (2003), “What to Convey in Antismoking Advertisements for Adolescents: The Use of Protection Motivation Theory to Identify Effective Message Themes,” Journal of Marketing, 67 (April), 1–18. Penaloza, Lisa (1994), “Border Crossing: A Critical Ethnographic Exploration of the Consumer Acculturation of Mexican Immigrants,” Journal of Consumer Research, 21 (June), 32–54. ——— (1995), “Immigrant Consumers: Marketing and Public Policy Considerations in the Global Economy,” Journal of Public Policy & Marketing, 14 (Spring), 83–94. ——— and Mary Gilly (1999), “Marketer Acculturation: The Changer and the Changed,” Journal of Marketing, 63 (July), 84–104. Pick, David and Kandy Dayaram (2006), “Modernity and Tradition in a Global Era: The Re-invention of Caste in India,” International Journal of Sociology and Social Policy, 26 (7–8), 284–94. Pieters, Rik, Tammo Bijmolt, Fred van Raaij, and Mark de Kruijk (1998), “Consumers’ Attributions of Proenvironmental Behavior, Motivation, and Ability to Self and Others,” Journal of Public Policy & Marketing, 17 (Fall), 215–25. Potter, John (2001), “Editorial: Trends in Environmentalism,” The Environmentalist, 21 (3), 183–85. Proshansky, Harold (1978), “The City and Self-Identity,” Environment and Behavior, 10 (June), 147–69. Rangan, Kasturi, Sohel Karim, and Sheryl Sandberg (1996), “Do Better at Doing Good,” Harvard Business Review, 74 (3), 42–54. Roccas, Sonia, Yechiel Klar, and Ido Liviatan (2006), “The Paradox of Group-Based Guilt: Modes of National Identification, Conflict Vehemence, and Reactions to the In-Group’s Moral Violations,” Journal of Personality and Social Psychology, 91 (4), 698–711. ———, Lilach Sagiv, Shalom Schwartz, Nir Halevy, and Roy Eidelson (2008), “Towards a Unifying Model of Identification with Groups: Integrating Theoretical Perspectives,” Personality and Social Psychology Review, 12 (33), 280–306. Sasson-Levy, Orna (2001), “Constructing Identities at the Margin: Masculinities and Citizenship in the Israeli Army,” The Sociological Quarterly, 43 (3), 357–83. Schatz, Robert and Ervin Staub (1997), “Manifestations of Blind and Constructive Patriotism: Personality Correlates and Individual-Group Relations,” in Patriotism in the Lives of Individuals and Nations, Daniel Bar-Tal and Ervin Staub, eds. New York: Nelson-Hall, 229–46. Schlegelmilch, Bodo, Greg Bohlen, and Adamantios Diamantopoulos (1996), “The Link Between Green Purchasing Decisions and Measures of Environmental Consciousness,” European Journal of Marketing, 30 (5), 35–55.

Demarketing, Minorities, and National Attachment / 121

Sidanius, Jim, Seymour Feshbach, Shana Levin, and Felicia Pratto (1997), “The Interface Between Ethnic and National Attachment: Ethnic Pluralism or Ethnic Dominance?” Public Opinion Quarterly, 61 (1), 102–133. Sirgy, Joseph (1982), “Self-Concept in Consumer Behavior: A Critical Review,” Journal of Consumer Research, 9 (December), 287–300. Stadler, Nurit (2002), “Is Profane Work an Obstacle to Salvation? The Case of Ultra Orthodox (Haredi) Jews in Contemporary Israel,” Sociology of Religion, 63 (4), 455–74. Stern, Paul (1999), “Information, Incentives, and Proenvironmental Consumer Behavior,” Journal of Consumer Policy, 22 (4), 461–478. Strang, Veronica (2005), “Common Senses: Water, Sensory Experience and the Generation of Meaning,” Journal of Material Culture, 10 (1), 92–120. Talukdar, Debabrata, Sumila Gulyani, and Lawrence Salmen (2005), “Customer Orientation in the Context of Development

122 / Journal of Marketing, March 2009

Projects: Insights from the World Bank,” Journal of Public Policy & Marketing, 24 (Spring), 100–111. Toft, Monica (2002), “Differential Demographic Growth in Multinational States: Israel’s Two-Front War,” Journal of International Affairs, 56 (1), 71–94. Torras, Mariano and James Boyce (1998), “Income, Inequality, and Pollution: A Reassessment of the Environmental Kuznets Curve,” Ecological Economics, 25 (2), 147–60. Triandis, Harry (1995), Individualism and Collectivism. Boulder, CO: Westview Press. Ustuner, Tuba and Douglas Holt (2007), “Dominated Consumer Acculturation: The Social Construction of Poor Migrant Women’s Consumer Identity Projects in a Turkish Squatter,” Journal of Consumer Research, 34 (1), 41–56. Wall, Anthony (2005), “Government Demarketing: Different Approaches and Mixed Messages,” European Journal of Marketing, 39 (5–6), 421–27. Zinkhan, George and Les Carlson (1995), “Green Advertising and the Reluctant Consumer,” Journal of Advertising, 24 (2), 1–6.