The current issue and full text archive of this journal is available at www.emeraldinsight.com/0263-4503.

htm

MIP 27,1

146
Received November 2007 Revised June 2008 Accepted September 2008

A cross-cultural comparison of individualisms moderating effect on bonding and commitment in banking relationships
Satyabhusan Dash
Indian Institute of Management, Lucknow, India

Ed Bruning
I.H. Asper School of Business, University of Manitoba, Winnipeg, Canada, and

Kalyan Ku Guin
Vinod Gupta School of Management, Indian Institute of Technology, Kharagpur, India
Abstract
Purpose The purpose of this paper is to describe a cross-cultural study which examined individualisms moderating effect on the relationship between bonding and commitment between banks and their corporate clients. Design/methodology/approach Data were collected through surveys completed by corporate customers from 126 Canadian companies and 156 Indian companies. Multiple regression analysis was used to calculate relative effects of structural and social bond on commitment across the two samples. Hierarchical moderated regression analysis was used to examine individualisms moderating effect on the bonding-commitment relationship. Findings The papers ndings indicate that social and structural bonding are both antecedent to commitment, but that social bonding is given higher importance in the low individualism Indian society, while structural bonding is more important in the high individualism Canadian society. Individualism moderates the relationship between both social and structural bonding and commitment. Practical implications Bank relationships are dependent upon specic cultural contexts in which buyers and sellers interact. The type of bonding relationship (e.g. social or structural) determines the strength of commitment. Bank managers must understand the proper emphasis to place on developing social connections versus business transactional relationships with clients in individualistic versus collective cultures. Originality/value This paper dramatizes the importance of understanding ways in which bonding relates to commitment, particularly when societal values vary and thus alter the relative importance of forms of bonding that generate commitment. Through empirical analyses, the paper demonstrates the moderating effect of individualism on the social bonding-commitment and structural bonding-commitment linkages in the context of an important service sector. To date, these relationships have not been explored in either the Indian or Canadian context. Keywords National cultures, Banking, Customer relations, Consumer behaviour, Canada, India Paper type Research paper
Marketing Intelligence & Planning Vol. 27 No. 1, 2009 pp. 146-169 q Emerald Group Publishing Limited 0263-4503 DOI 10.1108/02634500910928380

The authors gratefully acknowledge the nancial support for this study from the Department of Foreign Affairs and International Trade, Government of Canada, and the Shastri Indo-Canadian Institute.

Introduction The importance of maintaining long-term relationships is an established fact in marketing theory and practice. A plethora of books (Sheth and Parvatiyar, 2000; Gummesson, 2002), special journal issues, and conference proceedings (both academic and managerial) in recent years exhibit testimony to the continuing interest in the relationship-marketing paradigm. In the past two decades marketers have shifted their thinking about transactions and relationships. Scholars have argued for a shift from a transaction to a relationship viewpoint (Gronroos, 1994), from a goods-dominated to a service dominated viewpoint (Vargo and Lusch, 2004), and for acceptance of the fact that relationships can enhance nancial returns (Gummesson, 2004) through creating customer and shareholder value (Payne and Frow, 2005). Furthermore, relationship marketing has been advocated as being the most relevant strategy for success in corporate banking (Moriarty et al., 1983; Perrian et al., 1991; Paulin et al., 1998). Hence, relationship marketing has become an important issue in the academic discipline and in banking practice. With the globalization of the world economy, cultures role in shaping business relationships has become a critical focus of academic researchers attention as well. Although considerable effort has been devoted to identifying factors that enhance the development of business relationships, few studies have explored the connection between cultural values and key elements of a business relationship particularly buyer-seller bonding and the development of a committed tie between them. Furthermore, most studies that address business relationships and cultural values focus on North America and Europe (Williams et al., 1998; Donthu and Yoo, 1998). It remains to be seen whether the ndings from these studies apply to other regions of the world. The search still continues among marketing researchers to develop a general theory of relationship marketing that is useful in explaining a wide range of buyer-seller relationships across cultures (Palmer, 1993). Culture values, beliefs and attitudes are important factors in consumers decision processes, and researchers consider culture an important determinant of consumer behavior (Aaker and Lee, 2001; Aaker and Maheswaran, 1997; Aaker and Williams, 1998; Desphande et al., 1986; Gurhan-Canli and Maheswaran, 2000; Han and Shavitt, 1994; Henry, 1976; Howard and Sheth, 1969; Klein et al., 1998; Ter Hofstede et al., 1999; Bruning, 1997). As antecedents of personal behaviors, different value orientations display different behavior patterns (Rokeach, 1973). Roth (1995), in a ten-country sample study, found signicant moderating effects of culture on the market share performance of brand image strategies. Studies by Donthu and Yoo (1998), Furrer et al. (2000) and Mattila (1999) have demonstrated that a signicant linkage exists between culture and service quality expectations. Hence, marketing efforts would be most effective when cultural value differences are considered, and would achieve better results when they match the cultural values of target consumers (Farley and Lehman, 1994). Most research to date that has examined antecedents of loyalty and commitment has focused on industrial markets, distribution channels and consumer goods. Professional service organizations such as legal, consulting and accounting rms, represent an under-researched area in terms of exchange relationships (Beaten and Beaton, 1995). For service rms, maintaining strong customer relationships is especially important due to the intangible, ephemeral and often interpersonal nature of the service delivery process. Commercial banking relationships are stronger cases than most service sectors for the

Bonding and commitment

147

MIP 27,1

148

development of ongoing relationships. In the face of increasing competition, the maintenance of relationships between banks and their corporate customers has been advocated as an excellent way for banks to establish a unique long-term relationship with their customers (Watson, 1986; Man So and Speece, 2000). Taken together, we believe business relationships and cultural values are among the more important issues facing international marketers. In this study, we focus on the manner in which one element of culture impacts upon business relationships in the banking sectors of two countries. More specically, we focus on individualisms moderating inuence on the bonding-commitment relationship in a commercial banking context. As we will demonstrate, our contribution not only adds to knowledge about cultures role in affecting business relationships, but it also provides evidence that shows cultural values impact the bonding-commitment relationship differently across two countries in an important service sector. Our paper has four main sections. First, the studys theoretical framework is reviewed in order to develop an understanding of the structure of the bonding-commitment-individualism relationship between a corporate customer and its commercial bank. Second, details of our survey methodology are discussed. We present our research ndings in third section. Finally, in the fourth and nal section we discuss our ndings, draw our conclusions, and discuss the implications for bank marketing as well as suggest avenues for future academic research. Theoretical background and hypotheses The foundation for this research was the extant literatures in relationship marketing in business-to business networks and national culture. Figure 1 shows the theoretical links between relational bonding and commitment, with the moderating effect of individualism, an important element of national culture. Commitment Relationship commitment as dened by Moorman et al. (1992) is an enduring desire to maintain a valued relationship. As per the authors, a valued relationship exits when the relationship is considered important. Similarly, their enduring desire to maintain basically means that a committed partner wants the relationship to endure indenitely and is willingly to work at maintaining it. Morgan and Hunt (1994, p. 23) dene
Conceptual Model Cultural Moderator Individualism

Social Bonding Relationship Commitment Structural Bonding

Figure 1. Conceptual model

commitment as an exchange partner believing that an ongoing relationship with another is so important as to warrant maximum effort to maintain it. They view relationship commitment as central to all relational exchanges between buyers and sellers. In services relationship marketing, Berry and Parasuraman (1991) make mention of the fact that relationships are built on the foundation of mutual commitment. Processes that build and nurture commitment are central areas of research. Similarly, the process through which consumers become loyal to specic brands has been widely discussed. Initially, loyalty was viewed as repeat buying. However, as the eld of consumer behavior matured researchers came to realize that repurchase is not sufcient evidence of brand loyalty. Assael (1969) denes brand loyalty as a commitment to a certain brand. Firms see brand loyalty as one of their main assets and make efforts to build and nurture it. Wang (2008) conducted a cross-departmental study of the nancial services industry based on three consumer samples from Taiwan. His empirical evidence suggests that commitment had a positive and signicant effect on behavioral loyalty. In summary, establishing commitment is a fundamental condition for building strong relationships. As we will argue in the following section, the type of bonding between two parties inuences commitment to the relationship, and the specic bonding-commitment relationship is dependent upon national cultural values. Bonding Bonding is dened as the dimension of a business relationship that results in two parties (customer and supplier) acting in a unied manner towards a desired goal. Various bonds exist between parties that indicate different levels of a relationship (Callaghan et al., 1994). Bonding has been successful in explaining within country buyer-seller relationships (IMP Group, Hakansson, 1982; Wilson and Moller, 1988). Bonds reect and cause commitment in business relationships (Hakanson and Snehota, 1995). Wang (2008) provided evidence that relationship bonding affected commitment and behavioral loyalty through trust. Bonds, or strong ties between business rms, are important aspects of exchange relationships in the network approach, and are classied under two broad categories: structural and social bonding. Structural bonding is the task orientation between buyer and seller. It is the degree to which certain ties link and hold a buyer and seller together in a relationship as a result of some mutually benecial economic, strategic, technological, or organizational objective (Williams et al., 1998). Social bonding is the bonding that takes place between individuals (i.e. the buyer and seller). During social bonding, individuals are bonded together via the organizational members personal and social relationship with their counterparts in a particular rm. Personal factors, such as trust or satisfaction with the relationship partner, play an important role in developing social bonding (Williams et al., 1998). Wilson (1995) denes social bonding as the degree of mutual personal friendship and liking shared by the buyer and seller. Recent ndings support the position that both structural and social bonding are important in terms of establishing committed customers. In their study of commercial banking relationships, Paulin et al. (2000) demonstrated that, despite the current emphasis on technology, key contact personnel remain important for the success of professional business-to-business services. When account managers are changed,

Bonding and commitment

149

MIP 27,1

150

the business clients feel that their commitment relationship with the bank is weaker and less client-oriented. This phenomenon is indicative of a lessening of both structural and social bonding structural bonding, in the sense that the client is dependent upon the ability of the banks representative to service his needs and does not receive the needed service, and social bonding, to the extent that the client feels a personal loss in the departure of the valued bank representative. Paulin et al. (2000) also provided evidence that changing account managers is negatively associated with a banks external effectiveness as measured by client satisfaction, purchase intention, service quality and by a willingness to recommend the bank to another party. In a similar study, Colgate and Danaher (2000) demonstrated that excellent personal bankers could increase overall customer satisfaction, commitment and loyalty compared to customers who do not have a personal banker. Moreover, personal bankers establish both structural as well as social bonds with clients. While studying members of the Purchasing Management Association of Canada, Smith (1998) reported that social, functional, and structural bonds provide the context from which relational outcomes, such as trust, satisfaction and commitment, are evaluated. Furthermore, he argued that communication, cooperation, and relationship investment are important predictors of social bonding, while relationship investment predicted structural bonding. Williams et al. (1998) assert that both social and structural bonding are positively related to commitment; however, structural bonding has a greater effect on commitment than social bonding. In sum, based on the research on bonding and commitment, we propose that relational bonding (both structural and social), from the view point of the corporate customer, will be positively related to a customers commitment to a bank. Thus, bonding is an important antecedent to commitment. Our formal hypotheses are as follows: H1. Social bonding between a corporate customer and its bank will be positively related to commitment in: H1a. India. H1b. Canada. H2. Structural bonding between a corporate customer and its bank will be positively related to commitment in: H2a. India. H2b. Canada. As Williams et al. (1998) have demonstrated, social bonding will be higher in countries where relationships between buyers and sellers are more important than transactional exchanges, while structural bonding will be more important in transaction-oriented countries. Given Malhotra et al.s (1994) nding that, compared to developed countries, developing countries assign more importance to social dimensions of business relationships, we hypothesize that: H3. Social bonding will be given higher importance by Indian corporate customers compared to their counterpart Canadian corporate customers.

H4. Structural bonding will be given higher importance by Canadian corporate customers compared to their counterpart Indian corporate customers. National culture The concept of culture is widely interpreted in the academic literature. Kroeber and Kluckhohn (1952) identied more than 160 ways culture can be dened. Kluckholn (1962, p. 25) dened culture as the part of human makeup which is learned by people as the result of belonging to a particular group, and is that part of learned behavior that is shared by others. It is our social legacy, as contrasted to our organic heredity. To study national culture between two countries we turn to the seminal work of Hofstede (1991, p. 5), who dened culture as the collective programming of the mind which distinguishes the members of one group or category of people from those of another. Hofstedes typology of culture is one of the more important and popular theories of culture types. While his theory is based on empirical data collected from IBM employees, scholars have relied on Hofstedes insights to study a nations cultural values (Sondergaard, 1994). In his theoretical system, countries are grouped into categories according to core cultural values. Samiee and Jeong (1994) make the point, however, that counties may not be homogeneous with regard to core values. Nonetheless, scholars have stated that Hofstedes (2001) core values not only reect beliefs of IBM employees, but also they have been found to pertain to individual citizens and consumers as well, although strong arguments have arisen claiming that Hofstedes generalizations apply to work-related values and are inappropriate in other contexts (Sondergaard, 1994). In this study, we accept Hofstedes notion that individualism is an important national value; however, we utilize an alternative measurement scale developed by Yoo et al. (2001) and Yoo and Donthu (2002) to operationalize the construct. Social bonding, structural bonding and national culture Malhotra et al. (1994) suggest that business respondents from developing countries give more importance to social interaction and personal connectivity than do their counterparts from developed countries. Similarly, Williams et al. (1998) characterize individualistic national cultures as those with less interpersonal orientation, while collectivist national cultures have high needs for interpersonal orientation. Based on an empirical study of buyer-seller relationships among business respondents from the United States, Germany, Costa Rica, and Jamaica, the authors conrmed their hypothesis that buyers from collectivist countries had the strongest desire for social bonding. Conversely, buyers from highly individualistic countries had the strongest desire for structural bonding. According to Williams et al. (1998), people from collectivist societies place greater importance on interaction and group-oriented relationships compared to those from individualist societies. Less relationship-oriented cultures are more likely to choose a bank primarily based on objective performance criteria, such as good pricing and delivery time. Conversely, buyers in more relationship-oriented cultures are more likely to prefer doing business with a bank that they can establish a personal and social relationship. Because collectivists value personal and social relationships, compared to buyers from individualistic-oriented societies, buyers from collectivistic national cultures are more likely to emphasize

Bonding and commitment

151

MIP 27,1

152

interpersonal factors and will attach greater importance to social bonding in their relationships (Williams et al., 1998). From the above discussion, we conclude that buyers from collectivist countries, (e.g. India), will give more emphasis to interpersonal orientation and will attach greater importance to social bonding. Buyers from individualistic countries, (e.g. Canada), on the other hand, will place more emphasis on tasks to be performed and will therefore attach greater importance to structural bonding. Thus, individualism will tend to diminish social bondings role in enhancing commitment, but it will also enhance structural bondings effect on commitment. The following hypotheses organize our tests of these propositions: H5. Individualism will negatively moderate the relationship between social bonding and commitment. H6. Individualism will positively moderate the relationship between structural bonding and commitment. Methodology Survey Our study is based on a survey of Canadian and Indian banking business customers. Two versions of the survey, mail and internet, were developed and pre-tested in both countries prior to elding. The two versions were identical except for the manner in which they were presented to survey respondents. In developing the questionnaire, we were careful to insure the questionnaire items reected our study objectives. Thus, sections were included that captured respondents demographic details, attitudes, beliefs, and intentions. The instrument was subjected to several reviews by university colleagues and two groups of banking customers in order to remove obvious errors and to insure that the items were appropriate for the business context. Completion time for each instrument was approximately 15 minutes. For the internet survey, respondents were notied by the sponsor and given the internet account and password that would allow them access to the survey web page. Mail surveys were mailed rst class, postage paid, with a stamped, return envelope included with the instrument. Because different instruments were used to collect the data, it is possible that instrument bias may exist. We checked means and variances of the key study variables in the two questionnaire types and were unable to discern a signicant difference, which allowed us to pool the data for analyses. Follow-up telephone calls and personal contacts were made in an attempt to increase response rates. Sample For the purpose of our study, corporate customers are those bank customers who work in the business sector, maintain either a deposit or credit account for business purposes, and conduct transactions through a bank representative. In Canada, the Certied General Accountants Association sent an e-mail request with URL address for the web site to its members asking them to participate in the study. The web survey was provided online at a specied URL address. We also purchased a mailing list from a reputable Canadian listing service. Respondents were randomly selected based on Canadian province, sales volume and type of industry (i.e. manufacture, trade or service). For the Indian sample, company listings were provided by the Institute of Cost

and Works Accountants of Indian Chambers of Commerce located in different state capitals. In order to obtain comparability with the Canadian sample, the Indian sample was also stratied and randomly selected according to geographical region, sales volume and type of industry. Corporate customers from 1,000 Canadian and 1,000 Indian companies were contacted and asked to participate in the survey. Owing to wrong addresses, 148 mailed and internet-based questionnaires from Canada and 97 mailed and internet-based questionnaires from India were returned. Thus, 852 Canadian and 903 Indian corporate customers received the survey. After eliminating questionnaires with excessive amounts of missing data, the nal Canadian sample consisted of 126 responses and the Indian sample consisted of 156 responses. Our useable response rates were 14.8 percent in the Canadian sample and 17.3 percent in the Indian sample, which are both within the range typically reported for marketing research studies. A brief summery of the respondent sample is presented in Table I. As depicted in Table I, the demographic characteristics of Indian and Canadian respondents differed in several ways. First, female respondents were more prominent in the Canadian sample than the Indian sample (29.13 and 4.49 percent, respectively). Second, comparatively more manager-level respondents were observed in the Indian sample than in the Canada sample (60.90 and 21.25 percent, respectively), although there were more top-level executives in the Canadian sample relative to the Indian sample (78.74 and 21.25 percent, respectively). Third, the Indian sample was younger compared to the Canadian sample. Fourth, in terms of business sectors, there were more trading companies that participated in the Indian sample compared to the number reported in the Canadian sample (31.11 and 11.02 percent, respectively); however, the Canadian sample included more service companies than was found in the India sample (59.84 and 28.85 percent, respectively). Finally, the average length of the customer-bank relationship was 11 years in Indian companies while it was 17 years in Canadian companies. The differences identied across the two samples are as expected, given the cultural backgrounds of the two countries as well as their level of economic development. In summary, the data appear to be representative of the two populations. Measures All variables were measured with multiple-item scales and were derived from previous studies. A seven-point Likert-type scale was used for the sake of uniformity in measuring the variables. Commitment was measured using three items drawn from Morgan and Hunt (1994) that addressed how important the relationship was to the customer and how much effort the customer feels the relationship deserves. Social bonding was measured with four items and structural bonding was measured using ve items. Both social and structural bonding scale items were adapted from two previously tested scales (Rodriquez and Wilson, 1999; Smith, 1998). Smith (1998) argued that the items measuring structural bonding represent distinct sources of structure, not all of which would be expected in any given relationship, but when summed would indicate the extent of structural bonding. Since different measures of structural bonding may not covary, because they assess different aspects, we conceptualized structural bonding as a multidimensional composite index. Consistent with how such formative indicators are interpreted (Bollen and Lennox, 1991),

Bonding and commitment

153

MIP 27,1

154

Characteristics 149 (95.5) 7 (4.5) 95 (60.9) 61 (39.1) 28 66 47 15 62 49 45 91 54 11 Bhubaneswar, Calcutta, New Delhi, Mumbai, Hyderabad, Bangalore (17.9) (42.3) (30.1) (9.6) (39.7) (31.4) (28.9) (58.3) (34.6) (7.1) 11
a

Gender

Male Female Position Executive/manager Top level executive (director/VP/controller/GM) Age groups Under 30 30-40 41-50 .50 Nature of business Manufacturing Trading Servicing Sales turnover Under $10 million $10 million-100 million .$100 million Average years with the bank Business headquarters

Notes: aUnits: n (percent); samples n 282 (India 156; Canada 126)

Table I. Characteristics of multi-group sample India 89 37 26 100 4 27 42 53 37 13 76 73 38 15 Canada (70.9) (29.1) (21.3) (78.7) (3.1) (22.0) (33.1) (41.7) (29.1) (11.0) (59.8) (58.3) (29.9) (11.8) 17 Quebec, Alberta, Manitoba, Ontario, British Colombia, Nova Scotia, New Brunswick

Category

we assumed that our items result in structural bonding rather than vice versa. For example, we assert that structural bonding will be high when economic benets are high and/or when electronic ties or linkages will be more efcient; however, we do not expect the converse (i.e. that an increase in structural bonding causes high-economic benets through the relationship). As per Bollen and Lennox (1991), internal consistency is not applicable to such multidimensional composites. Support for our hypotheses will provide substantive validity of the structural bonding construct. Individualism was measured by administering the scale developed and validated by Yoo et al. (2001) and Yoo and Donthu (2002). The advantage of this scale over Hofstedes (1980, 1991, 2001) is that the instrument also applies to general consumer contexts and is not limited to work related situations. Items comprising each of the scales are presented in the Appendix. Reliability and validity Before using inferential statistics, we assessed the reliability and validity of our measurement scales by means of conrmatory factor analysis (CFA). Consistent with our conceptualisation, a three-factor measurement model was estimated using AMOS-4. The x 2 values for both country models were signicant: the Indian model, x 2 123.144, df 51, p 0.00 and the Canadian model, x 2 97.41, df 51, p 0.00). Other indices indicated a good t for both the Indian sample (CFI 0.92, TLI 0.89, RMSEA 0.07) and the Canadian sample (CFI 0.96, TLI 0.94, RMSEA 0.051). Furthermore, the x 2/df statistic is within acceptable levels (approximately two), which reinforces the strength of the models t to both data sets (Baumgartner and Homburg, 1996). In assessing measurement reliability, Fornell and Larcker (1981) stress the importance of examining construct reliability (CR) and average variance extracted (AVE). CR is calculated as the squared sum of the standardized individual item loadings divided by that squared sum of loadings plus the sum of the error variances for the measures. AVE is calculated as the sum of squared standardized loadings divided by that sum of squared standardized loadings plus the sum of the error variances for the measures. Bagozzi and Yi (1988) suggest two criteria in assessing reliability: the estimates of each measurement scale should equal or exceed a value of 0.60, and the average variance extracted should equal or exceed a value of 0.50. Table II presents the means, standard deviations, factor loadings, composite reliabilities and AVEs of our measurement scales. Structural bonding items are not included in the CFA analysis because they constitute a composite index. As indicated in Table II, requirements for CR and AVE are met as each multiple item scale exceeded the recommended cut-off criterion of 0.60 and 0.50, respectively, in both countrys data. Support for convergent validity is offered through the highly signicant loadings estimated for each individual item (Anderson and Gerbing, 1991). With the exception of one individualism item in the Indian sample, which was deleted from the nal analysis, all factor loadings were highly signicant and exceeded the p , 0.05 level, which is commonly considered meaningful in factor analytic investigations (Hair et al., 1995). As shown in Table II, factor loadings range between 0.65 and 0.91 for the Indian sample and 0.69-0.93 for the Canadian sample. For each sample, the estimates fell within the acceptable range and are statistically signicant ( p , 0.01). Therefore, the measures displayed adequate convergent validity.

Bonding and commitment

155

MIP 27,1

156

Measurement item Factor loading 0.69 0.69 0.80 0.81 0.76 0.82 0.67 0.75 0.80 0.72 0.77 0.69 0.83 0.65 5.77 (0.72) 5.90 (0.76) 5.84 (0.71) 0.53 00.88 0.91 0.86 5.68 5.28 5.14 5.26 (0.99) (1.17) (1.11) (1.19) 0.54 0.86 0.88 0.93 0.90 0.89 0.87 0.91 0.68 0.70 0.73 2.20 2.18 1.96 2.14 2.15 (1.19) (1.22) (1.08) (1.24) (1.21) 0.61

Individualism Self interest Self protection Individual rewards Individual welfare Individual loyalty Social bonding Friendship Personal relationship Advice sharing Belongingness Commitment Bank commitment Relationship maintenance Commitment effort

Notes: aCR construct reliability; bAVE average variance extracted

Table II. Summary of descriptive statistics and CFA results Factor loading Indian sample Mean (SD) CRa AVEb Canadian sample Mean (SD) 3.52 2.90 3.32 3.64 3.67 4.11 4.29 3.79 4.45 (1.37) (1.20) (1.37) (1.33) (1.42) 0.93 (1.67) (1.85) (1.66) (1.66) 0.91 4.95 (1.12) 5.03 (1.34) 5.10 (1.21) 0.78 0.80 CR 0.87 AVE 0.57

An assessment of discriminant validity was conducted for all the correlated constructs. A stringent criterion for testing discriminant validity, specied by Bagozzi and Phillips (1982), is to x the correlation parameter between two factors at 1.0 (as though perfectly correlated) and then employ a x 2 difference test on the values obtained for the constrained and unconstrained models. A signicantly lower x 2 value for the model in which the trait correlations are not constrained to unity would indicate that the traits are not perfectly correlated and that discriminant validity is achieved (Anderson and Gerbing, 1991; Bagozzi and Yi, 1988). For both the Indian and Canadian data, all x 2 differences were signicant at the p , 0.01 signicance level. In summary, the results indicate substantially improved model ts by separating the several constructs. The results support the conclusion that our measures displayed adequate discriminant validity. Cross-national measurement equivalence In cross-national research, instruments can only be compared when showing cross-national measurement invariance. In this study, measurement invariance is limited to metric invariance (i.e. invariance of factor loadings), which indicates that members in different groups interpret and respond to measures in an identical manner (Bagozzi and Yi, 1988; Steenkamp and Baumgartner, 1998). We tested for measurement invariance using a hierarchical ordering of two-nested models. The rst model tested whether the pattern of salient and non-salient factor loadings was equal across countries, also referred as congural invariance (Steenkamp and Baumgartner, 1998). Our results indicated that the data t well with the a priori hypothesized model: x 2 (102) 220.55 ( p , 0.001), CFI 0.94; TLI 0.92; RMSEA 0.051. In the second model, we tested whether the factor loadings were equal across countries a test for metric invariance (Steenkamp and Baumgartner, 1998). Our results indicated that the data t satisfactorily with the a priori hypothesized model: x 2 (111) 247.17 ( p , 0.001); CFI 0.93; TLI 0.91; RMSEA 0.055. The increase in x2 between the two models is signicant: Dx 2 (9) 26.52 ( p , 0.001). However, the remaining goodness-of-t indices (x 2/dof, TLI, CFI and RMSEA), which are less sensitive to sample size, show very considerably less marked decrease in t. In summary, the national level and cross-national level analysis indicated a high level of cross-national equivalence at the measurement level of social bonding, commitment and individualism constructs. Hence, we can compare mean scores as well as the strength of relationships among them and across national cultures. Analytical methods We employed hierarchical moderated regression analysis (HMRA) to examine whether the impact of social and structural bonding on commitment is moderated by individualism. From the correlation matrix (see Results section), it is clear that highly signicant correlations exist between individualism and the main-effect bonding variables. Therefore, HMRA is the appropriate analytical approach to eliminate possible multicollinearity problems in moderated regression analysis. To investigate the presence of a moderator effect on the construct relationships in our study, we follow Sharma et al.s (1981) procedure that involved constructing a regression model that included main effect terms for independent variables as well as product terms representing each moderator effect. Product terms were computed using our cultural

Bonding and commitment

157

MIP 27,1

158

moderator (e.g. individualism) and appropriate main effect variables (e.g. social and structural bonding). After accounting for main effects, each product term (i.e. moderator effect) was then entered individually. In terms of the moderator analysis, the key test was whether the product term accounted for a signicant amount of incremental variance. For our hypotheses to be supported, therefore, the signs of the main effect variables (i.e. social and structural bonding) needed to be positive (i.e. H1 and H2), the size of the social bonding-commitment coefcient must be higher for the Indian sample (H3), the size of the structural bonding-commitment coefcient must be higher for the Canadian sample (H4), and the values of the interaction terms must be in the predicted direction (i.e. negative for H5 and positive for H6). In order to further substantiate and validate the moderating effect of individualism in our model, we conducted an additional regression analysis at the group level. Since the moderating effect of individualism was tested with pooled data at the individual level, we expected the effect of social and structural bonding on commitment would also differ between groups within different Individualism categories. To test the relationship, we rst classied respondents into low and high individualism groups based on a median split, and followed with our regression analysis to examine the relative effects of social and structural bonding on commitment (i.e. H5 and H6). Results National value differences Using Yoo et al.s (2001) scale, we found that mean Individualism values for India and Canada were 2.13 and 3.41, respectively. One-way ANOVA results indicated that the mean difference was signicant ( p , 0.001, F 109.03) and that Canadian respondents were higher in Individualism compared to Indian respondents. Our ndings correspond with Hofstedes conclusions in the sense that the Indian Individualism index is lower than the Canadian index. Tables III and IV provide country-wise descriptive statistics and correlation matrices for our dependent and independent variables. Tests of hypotheses After achieving satisfactory results in the CFA model, we tested the studys specic research hypotheses. We conducted regression analysis to examine the hypothesized relationships between a countrys culture and its affects on the bonding-commitment relationship.
Social bonding Social bonding Structural bonding Commitment 1.00 0.43 (I) * 0.34 (C) * 0.30 (I) * 0.24 (C) * 2 0.14 (I)ns 2 0.13 (C)ns 5.34 (0.90) (I) 4.16 (1.57) (C)
ns

Structural bonding 1.00 0.06ns(I) 0.26 *(C) 2 0.28 * (I) 0.02ns(C) 5.60 (0.89) (I) 4.96 (0.75) (C)

Commitment

Individualism

1.00 2 0.06ns 2 0.10ns 5.83 (0.60) (I) 5.03 (1.15) (C) 1.00 2.13 (0.98) (I) 3.14 (1.08) (C)

Table III. Variable correlations, means and standard deviations: Indian and Canadian samples

Individualism Mean (SD) Notes: *p , 0.01.

Not signicant at p , 0.05 level. aIndian data (I) and Canadian data (C)

Variable

Social bonding

Structural bonding

Commitment

Individualism

Social bonding individualism

Structural bonding individualism

1.00 0.45 * 0.39 * 20.33 * 1.00 0.29 * 20.31 * 1.00 2 0.28 * 1.00 0.69 * 0.93 * 1.00 0.74 * 2 0.05ns 2 0.17 * 20.02ns 0.03ns 0.39 * 20.17 *

Social bonding Structural bonding Commitment Individualism Social bonding individualism Structural bonding individualism

1.00

Notes: *p , 0.01.

ns

Non-signicant at p , 0.05 level. aVariables included independent, dependent, moderator and product terms

Bonding and commitment

159

Table IV. Variable correlations a (pooled data) : main and interaction terms

MIP 27,1

160

The basis for the specic tests of the hypotheses and their results are presented in Tables V and VI. As recommended by Malhotra (1996), we evaluate the unstandardised slope coefcients in conducting our cross-cultural comparison between respondents from the two countries because unstandardised data reect an ethic standard, as they were unadjusted for within-sample variability: . Social bonding and commitment. The results reported in Table V indicate that social bonding was positively related to commitment in both countries, which supports our H1. A further review of the individual items in Table V indicate that the effect of social bonding on commitment was much higher in India than in Canada slope coefcients were b 0.22 ( p , 0.01) and b 0.13 ( p , 0.05) for the Indian and Canadian samples, respectively, thus supporting H3. . Structural bonding and commitment. With respect to H2, the results reported in Table V support the structural bonding-commitment linkage for the Canadian but not the Indian sample. Thus, H2 is only partially supported. Table V also indicates that the effect of structural bonding on commitment was higher in the Canadian compared to the Indian sample. Slope coefcients in the Canadian and Indian samples were b 0.31 ( p , 0.01) and b 2 0.05 (ns), respectively. Therefore, H4 was strongly supported. . Individualisms inuence on commitment, structural bonding and social bonding. Table VI presents the results of the HMRA that includes main and interaction (i.e. moderator) effects. All main effects were signicant and of the expected signs: social bonding (b 0.33; p , 0.01) and structural bonding (b 0.14; p , 0.05) were positively related, and individualism (b 2 0.15; p , 0.01) was inversely related, to commitment. The signicant incremental R 2 results reported in Table VI also indicated that the impact of social and structural bonding on commitment was moderated by individualism. The signs of the

Variable Table V. Country level effect of structural bonding and social bonding on commitment Social bonding Structural bonding R2 Notes: *p , 0.05; * *p , 0.01;

Unstandardized regression coefcients Pooled sample India 0.23 * * 0.15 * 0.17

ns

Canada 0.13 * 0.31 * 0.10

0.22 * * 2 0.05 0.09

Not signicant at p , 0.05 level

Variable(s) Table VI. Results of hierarchical moderator regression analysis (pooled data): main and moderator effects Social bonding Structural bonding Individualism Social bonding individualism Structural bonding individualism Notes: *p , 0.05; * *p , 0.01

R2 0.17 0.19 0.23 0.30

R 2 diff. 0.02 0.04 0.07

b (standardized regression coefcients)

0.33 * * 0.14 * 20.15 * * 20.67 * * 2.03 * *

F value 27.78 * * 21.18 * * 20.32 * * 23.74 * *

interaction terms (b 2 0.67 and b 2.03) were statistically signicant and in the predicted direction. Thus, H5 and H6 were supported by the data. The two interaction terms accounted for 11 percent of the incremental variance of the model, an amount that has signicant practical as well as theoretical implications. Based on the test developed by Sharma et al. (1981), we conclude that individualism serves as a signicant moderator variable in the bonding-commitment relationship. Bonding and commitment low vs high individualism group analysis. In order to corroborate individualisms moderating effect on commitment, we separated the study samples into two groups low and high individualism. The results reported in Table VII indicate that the effect of social bonding on commitment was much higher in the group measuring low in individualism compared to the group with the higher individualism score. Slope coefcients were b 0.34 ( p , 0.001) and b 0.06 (ns) for the low- and high-individualism groups, respectively. Furthermore, the effect of structural bonding on commitment was higher in the high (b 0.38 ( p , 0.001)) relative to the low (b 2 0.06 (ns)) individualism group. These grouped data ndings corroborated the results exhibited in Table VI and add additional support for H5 and H6.

Bonding and commitment

161

Discussion and conclusions We have learned from the relationship marketing literature that bonding is related to commitment. We also have learned from the cross-cultural marketing literature that groups differ with respect to the values they hold dear and that impact upon group members judgments and choices. Our research questions in this study emanate from the intersection of these two bodies of thought: do cultural values inuence the bonding-commitment relationship? More particularly, does individualism moderate the relationship between social bonding and commitment, and between structural bonding and commitment? Secondly, do the social bonding-commitment and structural bonding-commitment relationships differ across cultures? The objective of this study was to test several hypotheses related to these questions. Social and structural bonding are hypothesized as antecedents to commitment. Using OLS regression analysis at the country level, we found that social bonding was antecedent to commitment in both the Indian and Canadian samples, but that structural bonding led to commitment only in the Canadian sample. Hence, Canadian corporate customers commit to their banks through both channels: by establishing personal relationships with bank representatives as well as through mutually shared expectations and mutually benecial objectives. They utilize both personal (i.e. social bonding)

Variable Social bonding Structural bonding R2

Unstandardized regression coefcients Pooled sample Low individualism group 0.23 * * 0.15 * 0.17
ns

High individualism group 0.06ns 0.38 * * 0.12

20.06ns 0.24

0.34 * *

Notes: *p , 0.05; * *p , 0.01;

Not signicant at p , 0.05 level

Table VII. Effect of structural bonding and social bonding on commitment: low versus high individualism groups

MIP 27,1

162

and transactional specic means (i.e. structural bonding) in creating commitment to a commercial bank. Indian corporate customers, on the other hand, depend on personal relationships when developing commitment to their banks. Apparently, neither mutual economic benets nor the explicit tasks performed by bank staff appear to inuence the degree to which commitment is formed. Thus, the Indian customers relationship with the bank representative determines the strength of the Indian corporate customers commitment to the bank. Our results from the country-specic analyses dramatize the strong effect social bonding has on generating commitment in the Indian sample, and the equally strong effect structural bonding has on inuencing commitment in the Canadian sample. When the data were pooled for HRMA (i.e. Indian and Canadian respondents data were combined), both social and structural bonding became important commitment antecedents, although the social bonding effect was stronger than the structural bonding effect. Several explanations for this difference in ndings come to mind. First, the increased sample size of pooling the data may have created sufcient statistical power to detect a signicant effect for the structural bonding measure. Second, the strength of the Canadian reliance on structural bonding may have dominated the Indian indifference to structural bonding as a means of generating commitment. In either case, it appears from the two analyses that social bonding is an important source of commitment, but that structural bonding is less likely to be important to the Indian corporate customer. This nding corroborates previous study results reported by Williams et al. (1998), Rodriguez et al. (1999), Paulin et al. (2000) and Colgate and Danaher (2000). We also discovered that individualism inuenced commitment directly as well as moderating the strength of the relationship between the two forms of bonding and commitment. In the HMRA, individualism was negatively related to commitment, meaning that higher levels of individualism resulted in lower levels of commitment. Individualism, as a main effect, exerts a negative impact upon commitment, which is consistent with ndings from cross-cultural research (Hofstede, 1980, 2001). Self-centered people tend to rely on their own devices to achieve ends and place less responsibility on others, and commitment to others will likely be less. Thus, corporate customers who are highly individualistic will likely show less commitment to their banks. The more important nding, however, is that higher levels of individualism negatively affected the social bonding-commitment relationship. In the pooled HRMA analysis, the social bonding by individualism interaction term (i.e. the individualism moderator) was negative, which signied that the strength of the social bonding effect was lessened when individuals reported strong individualistic tendencies. Not only does high individualism negatively inuence the degree of commitment directly, it also dampens reliance on social bonding as a means of generating commitment. On the other hand, high individualism exerts a strong positive effect on the structural bonding-commitment relationship. In the pooled analysis, the structural bonding by individualism interaction term was positive and very strong. A similar effect was discovered in the grouped analysis structural bonding was highly signicant for the high-individualistic group but social bonding was not. The results suggest, therefore, that highly individualistic customers are more concerned with objective measures of success in their relationships with banks and place less emphasis on the personal relationship itself.

Implications In studying the perceptions of the Indian and Canadian banking customers, we have not only established valuable empirical benchmarks, but we have also provided insights as to how and why relational models work in eastern and western cultures. A major question addressed in our study is whether cultural differences affect the strength of social or structural bonds on commitment in Indian and Canadian banking. Our evidence indicates that it does. In this respect the study has contributed to the advancement of knowledge regarding banking customer-provider relationships in both countries. Furthermore, our effort represents one of the few marketing studies to empirically verify the moderating role of cultural values in customer-client, business-to-business relationships in a service sector. Both pooled and country-level analyses suggest that social bonding (affective and personal) is crucial to building commitment, a nding that supports previous studies by Williams et al. (1998), Rodriquez and Wilson (1999), Paulin et al. (2000) and Colgate and Danaher (2000). However, our ndings demonstrate that the inuence of social bonding on commitment is comparatively higher for Indian (lower individualism) compared to Canadian (higher individualism) customers, while structural bonding is more important for Canadian than for Indian customers. Our ndings have signicant implications for banking service providers with regard to establishing and maintaining relationships with international and immigrant customers. From a managerial point of view, country and cultural group level differences should be considered carefully when developing international service strategies. Our results show that understanding the linkage between individualism and aspects of a bank-customer business relationship will provide useful insights for how rms should allocate marketing resources to different cultural service markets. To ignore the impacts of cultural differences is to run the risk that marketing practices in one country may prove to be insufcient, or inappropriate, in another. Corporate customers steeped in traditional Indian culture will place signicant importance on the bank representative with which they work. While transaction-specic services and concrete displays of efciency will matter to them, our data indicate the personal relationship is most important and will matter more in the long run. It is also important that managers bear in mind that societies undergo cultural changes: as members age and pass on they are replaced with newer members whose values may not reect those of their parents or grandparents, or of their traditional business institutions. Hence, a national culture is always in a state of transition. In the current era, individualism may be on the rise as Indians, particularly young Indians, increasingly attempt to emulate the values and behaviors of western societies. With the emergence of a highly integrated world, where information, location and products are much more accessible, it is quite likely todays Indian youth will gravitate closer to westernized values and expectations. If the trend continues, todays youths children will be even more likely to acculturate western values individualism in particular. More over, these people will work in Indian businesses and make decisions about banks and banking services. In brief, they will become corporate customers eventually. Clearly, the implication for banks is to remain abreast of the shifting cultural values and adjust their customer service strategies to accommodate appropriately the cultural needs of each group they service. In conclusion, because cultural conditions differ among countries, and because relationships between customers and service providers (e.g. banks) differ as well,

Bonding and commitment

163

MIP 27,1

164

approaches to relationship marketing may vary and cannot be treated as homogeneous. Our ndings dramatize that cultural values have a strong inuence on how relationship marketing is practiced. We believe banks and other nancial institutions can benet by managing relationships more in tune with the cultural imperatives of the society while, at the same time, maintaining a vigilant eye on dynamic changes occurring in that society. Our ndings introduce a number of questions that are suggestive of future research directions. First, we have discussed Indian and Canadian cultures as if they were homogeneous. Are they in fact homogeneous or do sub-sets exist, which would therefore require banks to identify each sub-set in devising effective customer service strategies? Do other cultural values (e.g. uncertainty avoidance, power distance, etc.) moderate the bonding-commitment relationship? If so, what is the direction and strength of the relationship? Does individualism also moderate the trust-commitment relationship? Do other national cultural values moderate the trust-commitment relationship? Our foray into the intersection of relationship marketing and cross-cultural marketing research is ripe with possibilities to explore. In this study, we have garnered support for the argument that individualism inuences bonding and commitment in the Indian and Canadian banking context. The questions posed above will direct our future inquiries. Limitations and future research Our results clearly indicate that individualism moderates the bonding-commitment linkage at the customer level. Social rather than structural bonding processes inuence Indian corporate customers, who are characterized by low individualism, whereas Canadian corporate customers place higher preference on structural bonding processes. Our conclusions, however, may assume more homogeneity within cultural groups than actually occurs. Samiee and Jeong (1994) argue that sub-cultures may be present within the geographic boundaries of different countries, and that variations in relationship expectations between sub-cultures in the same country may exist as well. Similarly, in their research in relationship marketing, Donthu and Yoo (1998) discovered individual level variations in service quality expectations across several countries. Although we demonstrate individual level effects of bonding on commitment between groups measuring high and low on individualism, other within-group moderators may be operative as well. We suggest therefore that future studies focus on evaluating sub-cultures within a dominant national culture in order to conrm our generalizations. Secondly, we have investigated the moderating effect of a single cultural value individualism. Would a similar relationship hold for risk avoidance, long-term orientation, or other cultural or national values? It is expected that additional patterns exist, which, if true, suggest that culture might play a more signicant role in relationship building and maintenance than originally thought. Third, our hypotheses are posed in the context of a specic business service relationship commercial banking. Our study ndings would be generalized to a broader class of buyer-seller relationships if additional service sectors were studied as well. Fourth, the studys focus is directed to the effect of an aspect of national culture on the bonding-commitment relationship. Undoubtedly, several other variables inuence commitment in buyer-seller relationships, as suggested by Wilson (1995). If the study were expanded to link other factors, in addition to national culture, conclusions regarding the nature of the

bank-customer relationship would be more enlightening. Finally, in this study, our research focus was directed to only the buyers perspective. In a dyadic relationship analysis, however, both buyers and sellers observations are very much important for making a meaningful conclusions. For this reason we recommend future research work be based on a dyadic perspective, which may provide a more accurate assessment of cross-cultural buyer-seller relationships.
References Aaker, J.L. and Lee, A.Y. (2001), I seek pleasures and We avoid pains: the role of self-regulatory goals in information processing and persuasion, Journal of Consumer Research, Vol. 28, pp. 33-49. Aaker, J.L. and Maheswaran, D. (1997), The effect of cultural orientation on persuasion, Journal of Consumer Research, Vol. 24, pp. 315-28. Aaker, J.L. and Williams, P. (1998), Empathy versus pride: the inuence of emotional appeals across cultures, Journal of Consumer Research, Vol. 25, pp. 241-61. Anderson, J. and Gerbing, D.W. (1991), Predicting the performance of measures in a conrmatory factor analysis with a pretest assessment of their substantive validities, Journal of Applied Psychology, Vol. 76 No. 5, pp. 732-40. Assael, H. (1969), The constructive role of inter-organizational conict, Administrative Science Quarterly, Vol. 14, pp. 573-82. Bagozzi, R.P. and Phillips, L.W. (1982), Representing and testing organizational theories: a holistic construal, Administrative Science Quarterly, Vol. 27 No. 3, pp. 459-89. Bagozzi, R.P. and Yi, Y. (1988), On the evaluation of structural equation models, Journal of the Academy of Marketing Science, Vol. 16 No. 1, pp. 74-94. Baumgartner, H. and Homburg, C. (1996), Applications of structural equation modeling in marketing and consumer research: a review, International Journal of Research in Marketing, Vol. 13 No. 1, pp. 139-61. Beaten, M. and Beaton, C. (1995), Marrying service providers and their clients: a relationship approach to service management, Journal of Marketing Management, Vol. 11 No. 55, pp. 55-70. Berry, L.L. and Parasuraman, A. (1991), Marketing Services, The Free Press, New York, NY. Bollen, K. and Lennox, R. (1991), Conventional wisdom on measurement: a structural equation perspective, Psychological Bulletin, Vol. 110 No. 2, pp. 305-11. Bruning, E. (1997), Country of origin, national loyalty and product choice: the case of international air travel, International Marketing Review, Vol. 14 No. 1, pp. 59-74. Callaghan, M.B., McPhil, J. and Yau, O.H.M. (1994), The theoretical development of dimensions of a relationship marketing orientation, in Feng, Y.C. and Wee, C.H. (Eds), Asia Pacic Business in the Year 2000 Conference Proceedings, Academy of International Business, Beijing, pp. 240-9. Colgate, M. and Danaher, P. (2000), Implementing a customer relationship strategy: the asymmetric impact of poor versus excellent execution, Journal of Academy of Marketing Science, Vol. 28 No. 3, pp. 375-87. Desphande, R., Hoyer, W.D. and Donthu, N. (1986), The intensity of ethnic afliation: a study of the sociology of Hispanic consumption, Journal of Consumer Research, Vol. 13, pp. 214-20. Donthu, N. and Yoo, B. (1998), Cultural inuences on service quality expectations, Journal of Service Research, Vol. 1 No. 2, pp. 178-85.

Bonding and commitment

165

MIP 27,1

166

Farley, J.U. and Lehman, D.R. (1994), Cross-cultural laws and differences in market response, Management Science, Vol. 40 No. 1, pp. 111-22. Fornell, C. and Larcker, D. (1981), Evaluating structural equation models and unobservable variables and measurement error, Journal of Marketing Research, Vol. 18, pp. 39-50. Furrer, O., Shaw-Ching Liu, B. and Sudharshan, D. (2000), The relationships between culture and service quality perceptions: basis for cross-cultural market segmentation and resource allocation, Journal of Service Research, Vol. 2 No. 4, pp. 355-71. Gronroos, C. (1994), From marketing mix to relationship marketing: towards a paradigm shift in marketing, Management Decision, Vol. 32 No. 2, pp. 4-20. Gummesson, E. (2002), Total Relationship Marketing: Rethinking Marketing Managements From 4Ps to 30Rs, 2nd ed., Butterworth-Heinemann, Oxford. Gummesson, E. (2004), Return on relationships (ROR): the value of relationship marketing and CRM in business-to-business contexts, Journal of Business & Industrial Marketing, Vol. 19 No. 2, pp. 136-48. Gurhan-Canli, Z. and Maheswaran, D. (2000), Cultural variations in country of origin effects, Journal of Marketing Research, Vol. 37, pp. 309-17. Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1995), Multivariate Data Analysis, 5th ed., Pearson Education, Singapore, Indian Reprint. Hakanson, H. and Snehota, I. (Eds) (1995), Developing Relationships in Business Networks, Routledge, London. Hakansson, H. (1982), International Marketing and Purchasing of Industrial Goods, Wiley, Chichester. Han, S. and Shavitt, S. (1994), Persuasion and culture: advertising appeals in individualistic and collectivistic societies, Journal of Experimental and Social Psychology, Vol. 30, pp. 326-50. Henry, W.A. (1976), Cultural values do correlate with consumer behavior, Journal of Marketing Research, Vol. 13, pp. 121-7. Hofstede, G. (1980), Cultures Consequences: International Differences in Work Related Values, Sage, Beverly Hills, CA. Hofstede, G. (1991), Cultures and Organizations: Software of the Mind, McGraw-Hill, Berkshire. Hofstede, G. (2001), Cultures Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations, 2nd ed., Sage, Thousand Oaks, CA. Howard, J.A. and Sheth, J.N. (1969), The Theory of Buyer Behavior, Wiley, New York, NY. Klein, J., Ettenson, R. and Morris, M. (1998), The animosity model of foreign product purchase: an empirical test in the Peoples Republic of China, Journal of Marketing, Vol. 62 No. 1, pp. 89-100. Kluckhohn, R. (Ed.) (1962), Culture and Behavior: Collected Essays of Clyde Kluckhohn, The Free Press, New York, NY. Kroeber, A.L. and Kluckhohn, C. (1952), Culture: a critical review of concepts and denitions, Paper of the Peabody Museum of American Archaeology and Ethnology, Vol. 47 No. 1, pp. 1-223. Malhotra, N.K. (1996), Methodological issues in cross-cultural marketing research: a state-of-the-art review, International Marketing Review, Vol. 13 No. 5, pp. 7-43. Malhotra, N.K., Ulgado, F.M., Agarwal, R. and Baalbaki, I.B. (1994), A comparative evaluation of dimensions of service quality between developed and developing countries, International Marketing Review, Vol. 11 No. 2, pp. 5-15.

Man So, S. and Speece, M. (2000), Perceptions of relationship marketing among account managers of commercial banks in the Chinese environment, International Journal of Bank Marketing, Vol. 18 No. 7, pp. 315-27. Mattila, A. (1999), The role of culture in the service evaluation processes, Journal of Service Research, Vol. 1 No. 3, pp. 250-61. Moorman, C., Zaltman, G. and Deshpande, R. (1992), Relationships between providers and users of market research: the dynamics of trust within and between organizations, Journal of Marketing Research, Vol. 29, pp. 314-29. Morgan, R.M. and Hunt, S.D. (1994), The commitment-trust theory of relationship marketing, Journal of Marketing, Vol. 58 No. 3, pp. 20-38. Moriarty, R.T., Kimball, R.C. and Gay, J.H. (1983), The management of corporate banking relationships, Sloan Management Review, Spring, pp. 3-15. Palmer, A. (1993), Cultural inuences on relationship marketing, in Hennig-Thurau, T. and Hansen, U. (Eds), Relationship Marketing, Gaining Competitive Advantage Through Customer Satisfaction and Customer Retention, Springer, New York, NY, pp. 271-6. Paulin, M., Perrien, J., Ferguson, R. and Payaud, M. (2000), Business effectiveness and professional service personnel: relational or transactional managers, European Journal of Marketing, Vol. 34 Nos 3/4, pp. 453-71. Paulin, M., Perrien, J., Ferguson, R., Alvarez, S. and Seruya, L.M. (1998), Relational norms and client retention: external effectiveness of commercial banking in Canada and Mexico, International Journal of Bank Marketing, Vol. 16 No. 1, pp. 24-31. Payne, A. and Frow, P. (2005), A strategic framework for customer relationship management, Journal of Marketing, Vol. 69, pp. 167-76. Perrian, J., Filiatrault, P. and Ricard, L. (1991), The implementation of relationship marketing in commercial banking, Industrial Marketing Management, Vol. 22, pp. 141-8. Rodriquez, C.M. and Wilson, T.D. (1999), Relationship bonding and trust as a foundation for commitment in US-Mexican strategic alliances: a structural equation modeling approach, ISBM Report No. 21, Pennsylvania State University, University Park, PA. Rokeach, M. (1973), The Nature of Human Values, The Free Press, New York, NY. Roth, M.S. (1995), The effects of culture and socioeconomic on the performance of global brand image strategies, Journal of Marketing Research, Vol. 32, pp. 163-75. Samiee, S. and Jeong, J. (1994), Cross-cultural research in advertising: an assessment of methodologies, Journal of the Academy of Marketing Science, Vol. 22 No. 3, pp. 205-17. Sharma, S., Durand, R.M. and Gur-Arie, O. (1981), Identication and analysis of moderator variables, Journal of Marketing Research, Vol. 18 No. 2, pp. 291-300. Sheth, J. and Parvatiyar, A. (2000), Handbook of Relationship Marketing, Response Book, Sage, New Delhi. Smith, B. (1998), Buyer-seller relationships: bonds, relationship management, and sex-type, Canadian Journal of Administrative Science, Vol. 15 No. 1, pp. 76-92. Sondergaard, M. (1994), Hofstedes consequences: a study of reviews, citations and replications, Organization Studies, Vol. 15 No. 3, pp. 447-56. Steenkamp, E.M. and Baumgartner, H. (1998), Assessing measurement invariance in cross-national research, Journal of Consumer Research, Vol. 25, pp. 78-90. Ter Hofstede, F., Steenkamp, F.M. and Wedel, M. (1999), International market segmentation based on consumer-product relations, Journal of Marketing Research, Vol. 36, pp. 1-17.

Bonding and commitment

167

MIP 27,1

168

Vargo, S. and Lusch, R. (2004), Evolving a new dominant logic for marketing, Journal of Marketing, Vol. 68, pp. 1-17. Wang, W-H. (2008), The interrelationship of retailers relationship efforts and consumers attitude and behavior, Measuring Business Excellence, Vol. 12 No. 1, pp. 13-28. Watson, I. (1986), Managing the relationships with corporate customers, International Journal of Bank Marketing, Vol. 4 No. 1, pp. 19-34. Williams, J.D., Han, S.L. and Qualls, W.J. (1998), A conceptual model and study of cross-cultural business relationships, Journal of Business Research, Vol. 42 No. 1, pp. 135-43. Wilson, D.T. (1995), An integrated model of buyer-seller relationship, Journal of the Academy of Marketing Science, Vol. 23 No. 4, pp. 335-45. Wilson, D.T. and Moller, K.K. (1988), Buyer-seller relationships: alternative conceptualizations, Working Paper No. 10-1988, Institute for the Study of Business Markets, Pennsylvania State University, University Park, PA. Yoo, B. and Donthu, N. (2002), The effects of marketing education and individual cultural values on marketing ethics of students, Journal of Marketing Education, Vol. 24, pp. 92-103. Yoo, B., Donthu, N. and Lenartowicz, T. (2001), Measuring cultural values: development and validation of the CVSCALE, working paper, Georgia State University, Atlanta, GA. Further reading Nunnally, J.C. (1988), Psychometric Theory, McGraw-Hill, Englewood-Cliffs, NJ.

Appendix
Commitment adapted from Morgan and Hunt (1994) Please rate your agreement with each of the 1. Our business rm is committed to our bank following statements. 2. We intend to maintain the relationship with our bank indenitely 3. The relationship our rm has with our bank is deserving of our rms maximum efforts to maintain Structural bonding adapted from Rodriquez and Wilson (1999) and Smith (1998) How important are all the following characteristics to 1. Formal contracts or agreements between your rm you in terms of continuing your and the bank relationship with your bank? 2. Sharing of industry or competitive information between your rm and the bank 3. Electronic ties or linkages between your business rm and bank (such as electronic mail, access to computerized banking transaction system, etc.) 4. Economic benets associated with your relationship with your bank 5. Exchange cards, gifts and the like on birthdays, holidays or on special events between your rm and bank Social bonding adapted from Rodriquez and Wilson (1999) and Smith (1998) How important are all the following characteristics to 1. Friendship between you and your bank you in terms of continuing your representative relationship with your bank? 2. Establishing personal relationship with you and your bank representative 3. Sharing of personal advice or support with your bank representative 4. Feelings of belongingness or acceptance between you and your bank representative Individualism adapted from Yoo et al. (2001) Please rate your agreement with each of the following 1. Individual should sacrice self-interest for the group statements (either at such school or the work place)w 2. Individuals should stick with the group even through difcultiesw 3. Group welfare is more important than individual rewardsw 4. Group success is more important than individual successw 5. Individuals should only pursue their goals after considering the welfare of the groupw 6. Group loyalty should be encouraged even if individual goals sufferw Notes: Social bonding and structural bonding scale item: 1 very important, 7 very unimportant. Individualism and commitment scale items: 1 strongly agree, 7 strongly disagree. wDenotes reverse coded item

Bonding and commitment

169

Table AI. Measurement scale items

Corresponding author Satyabhusan Dash can be contacted at: satya@iiml.ac.in

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints