Introduction

Building a deep and meaningful relationship between a consumer and a brand contributes to the long-term success and welfare of companies (Bhattacharya and Sen 2003). Over the last decade, consumer-brand identification (CBI) has emerged as an important construct that explains a consumer’s sense of belonging to a brand and positively influences consumer loyalty (Haumann et al. 2014; Lam et al. 2010; Popp and Woratschek 2017; Stokburger-Sauer et al. 2014). To date, scholars have established the sequential relationships between brand characteristics (e.g., prestige and distinctiveness), CBI, and consumer loyalty (Carlson et al. 2009; Stokburger-Sauer et al. 2014).

While the primary focus of CBI is on the vertical tie between a consumer and a brand, the horizontal ties between fellow brand users also exist and have been examined in the literature on brand community (Muniz and O’Guinn 2001) and consumer-to-consumer social capital (Mathwick et al. 2007). In the customer-based brand equity pyramid, Keller (2008) identifies sense of community as a dimension of brand resonance which is considered “the ultimate relationship and level of identification that the customer has with the brand” (p. 72). This view is consistent with self-determination theory (Ryan and Deci 2000) which includes the premise that a sense of community is associated with the basic human needs of autonomy, competence, and relatedness, which in turn foster intrinsic motivation and well-being. These theoretical perspectives together imply that CBI not only stems from brand characteristics (e.g., brand prestige and brand distinctiveness), but also entails a consumer’s sense of belonging to the brand as a social unit. In this study, we emphasize the importance of social capital among consumers, rather than prestigious and distinctive brand characteristics.

Another research void involves the role of consumer happiness in CBI research. Creating brands that make consumers happy is an effective marketing strategy because it produces a strong feeling of affective commitment and brand loyalty (Pansari and Kumar 2017; Schnebelen and Bruhn 2018). Consumer happiness is defined as a state of well-being related to a specific consumption activity (Mogilner et al. 2012; van Boven and Gilovich 2003) and can be seen as a useful comprehensive construct for explaining a sense of the larger picture of how brand consumption leads to people experiencing a happy life over time (Nicolao et al. 2009). Scholars have completed preliminary work exploring how consumer happiness contributes to a brand’s long-term success (Devezer et al. 2014; Zhong and Mitchell 2013), providing an opportunity to further advance our understanding of the effect of consumer happiness on CBI.

It is also important to note that most findings in previous CBI research are based on cross-sectional data collected at a single point in time and arise from the relationship between CBI and behavioral intentions (Lin et al. 2019; Millán and Díaz 2014; Popp and Woratschek 2017; Stokburger-Sauer et al. 2014; Wolter and Cronin 2016). Although a few researchers tested the dynamic changes of antecedents (e.g., perceived quality, self-brand congruity, and consumer innovativeness) and outcomes (e.g., behavioral intentions and willingness to pay more) of CBI over time (Haumann et al. 2014; Huang et al. 2017; Lam et al. 2013), it is still not clear how CBI influences future behavioral consequences, for example, in relation to behavioral loyalty (Garnefeld et al. 2013; Zeithaml et al. 1996) and purchase frequency (Seiders et al. 2005). Conducting a longitudinal study also allows us to examine the interaction effect of CBI and consumer happiness on future behavioral consequences. This analysis is important because consumer happiness is relatively stable and enduring (Devezer et al. 2014) and would be related to long-term brand consumption in people’s life (Zhong and Mitchell 2013).

Recognizing there is much left to understand, a more comprehensive model integrating CBI, social capital, consumer happiness, and future behavioral outcomes is warranted. The purposes of this study are to (1) develop a model that expands the traditional organizational identification framework (Ashforth and Mael 1989; Mael and Ashforth 1992) with the inclusion of organizational social capital (Nahapiet and Ghoshal 1998; Tsai and Ghoshal 1998) and consumer happiness (van Boven and Gilovich 2003) and (2) investigate the antecedents and consequences of CBI over time.

This study falls within the paradigm of transformative consumer research (TCR), a body of scholarship concerned with linking consumer research to consumer well-being (Anderson et al. 2013; Csikszentmihalyi 2000; Mick 2006). In the current marketplace which serves as a highly complicated, uncertain, and competitive environment, consumers are not likely to remain loyal to brands without achieving happiness in life. In fact, in the face of economic disturbances, brands such as Coca-Cola and McDonald’s attempt to connect with consumers by promising happiness with simple, but fundamental messages (e.g., “open happiness” and “happy meal”; Mogilner et al. 2012). This can also be achieved by appealing to consumer happiness through extraordinary experiences as evidenced by brands such as Disney and BMW (Bhattacharjee and Mogilner 2014). An integration of CBI and consumer happiness is both timely and practical.

The remaining content of this paper is organized as follows: first, an integrative framework of CBI, consumer-to-consumer social capital, and consumer happiness is proposed in order to identify the antecedents and consequences of CBI. The next section presents the methods we used. The results of hypothesis testing are then presented, followed by the discussion section.

Theoretical model and hypotheses

Organizational identification theory (Ashforth and Mael 1989; Mael and Asforth 1992) maintains that in order to identify with a brand, consumers need to perceive themselves as psychologically intertwined with the fate of a brand. In empirical work, researchers have used either Mael and Asforth’s (1992) or Bergami and Bagozzi’s (2000) cognitive identification scale to measure a person’s perceived identification with a brand (Ahearne et al. 2005; Bhattacharya et al. 1995; Haumann et al. 2014). Consistent with this view, we follow the cognitive approach and define CBI as an individual’s perceived oneness with a brand and the experience of the brand’s successes and failures as one’s own (Ashforth and Mael 1989; Stokburger-Sauer et al. 2014). In the following section, we first present research hypotheses regarding the relationships between social capital factors, CBI, and behavioral consequences. Then, research hypotheses pertaining to the relationships between CBI, behavioral outcomes, and consumer happiness are derived (see Fig. 1).

Fig. 1
figure 1

Theoretical framework and research hypotheses

Full size image

The impact of social capital

According to organizational social capital theory (Nahapiet and Ghoshal 1998; Tsai and Ghoshal 1998), social capital is defined as “the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit” (Nahapiet and Ghoshal 1998, p. 243) and includes three dimensions: structural (e.g., social interaction ties), cognitive (e.g., shared vision and culture), and relational (e.g., trust, identification, and norms of reciprocity).

In this study, we attempt to extend CBI research by examining the impact of social interaction ties (structural dimension) and shared vision (cognitive dimension) on CBI (relational dimension). As a proxy variable of the relational dimension, we select social identification and exclude other relational characteristics (e.g., reciprocity and trust). Social identification is similar to the idea of bridging (or inclusive) social capital, which refers to a sense of connection with heterogeneous people (Putnam 2000) and represents CBI in brand communities. On the other hand, the elements of reciprocity and trust are related to the idea of bonding (or exclusive) social capital, which is a sense of cohesion or solidarity among homogeneous people (Putnam 2000). In brand communities, activating the feelings of reciprocity and trust is applicable only to those who are socialized into homogeneous networks of consumers. In order to adequately conceptualize relational social capital among heterogeneous brand community members, social identification should be taken into account.

In the current marketplace, “brands are shifting away from single to shared ownership” (p. 25, Swaminathan et al. 2020). On the basis of the literature reviewed, CBI may evolve from the social interactions among consumers. As consumers interact with one another over time, their relationship becomes more intimate and stable. Close and continuous social interactions allow consumers to know each other, share consumption experiences, and develop a common point of view (Tsai and Ghoshal 1998), thus increasing a sense of belongingness and identification with their brand community. Harmonious and enjoyable social interactions in brand communities play a key role in creating brand value (Mingione et al. 2020) and building CBI (Augusto and Torres 2018) because peer-to-peer communications in brand communities facilitate the sharing of brand-related information (Meek et al. 2019), value creation practices (Mingione et al. 2020; Schau et al. 2009), and the development of long-term relationships not only with other consumers, but also with the focal brand (Augusto and Torres 2018; Stokburger-Sauer et al. 2014). In a hyperconnected society, brands act as catalysts of social interactions among consumers that offer social benefits such as a sense of belonging and identity expression in brand communities (Swaminathan et al. 2020). Combining these considerations, we hypothesize the following:

H1

The social interaction ties among fellow brand users have a positive impact on CBI.

A shared vision and a set of common values may also stimulate the development of CBI. Organizational social capital theory maintains that common goals, values, and beliefs promote the harmony of social relationships and facilitate the value congruence between an organization and its members (Tsai and Ghoshal 1998). In marketing, corporate vision can shape corporate brand identity (Balmer 2012), and the sharing of common visions acts as a significant driver of value co-creation processes among actors (Mingione and Leoni 2020). In brand communities, shared visions lead consumers to have common brand meanings and co-produce communal brand identity (Healy and McDonagh 2013). An implication from these theoretical explanations is that cognitively shared visions and values may play a significant role in developing and maintaining the relationship between a consumer and a brand. That is, as a consumer has mutual goals and values with fellow brand users as well as an interdependent self-concept, greater identification with his or her favorite brand arises and promotes the assimilation of the person’s self-concept into the image of the brand (Bagozzi and Dholakia 2006). Collectively, the pathway from shared vision to CBI can be hypothesized as follows:

H2

The shared vision among consumers in a brand community has a positive impact on CBI.

We also extend the brand management literature by examining the impact of consumer-to-consumer social capital on future behavioral loyalty. One way to explain this relationship is to assume that the more a consumer is integrated into a brand community, the more loyal the consumer is toward the brand (Muniz and O’Guinn 2001; Schau et al. 2009). The brand community literature offers explanations for this relationship. For example, previous researchers have found that consumers become loyal to their favorite brands by drawing values from peer-to-peer interactions in brand communities (Muniz and O’Guinn 2001; Schau et al. 2009). Consumers derive social and hedonic value from the consumption of branded offerings in brand communities (Schau et al. 2009). Such value-creating activities are socially constructed through consumer-to-consumer interactions and induce consumers to remain loyal to a brand as well as to the brand community (Muniz and O’Guinn 2001; Schau et al. 2009). Similarly, other researchers suggest that when consumers participate in and foster peer-to-peer interactions in a brand community, their brand involvement increases, thereby resulting in greater brand loyalty (Bagozzi and Dholakia 2006).

In this study we include two different types of behavioral loyalty: one assessing consumers’ subjective intentions and one measuring purchase frequency. Previous studies have tested subjective measures of brand loyalty (e.g., Lin et al. 2019; Stokburger-Sauer et al. 2014; Wolter and Cronin 2016). To better understand and more consistently predict behavioral loyalty and determine if social capital and CBI influence long-term brand loyalty, we believe it is important to examine two loyalty constructs: subjective intentions and purchase frequency. Taken together, it is suggested that the social interaction ties among fellow brand users are likely to increase behavioral loyalty and purchase frequency. The reasoning described above prompts the following hypotheses:

H3a

The social interaction ties among fellow brand users have a positive impact on behavioral loyalty.

H3b

The social interaction ties among fellow brand users have a positive impact on purchase frequency.

We predict a shared vision among consumers is directly related to behavioral loyalty. In brand community research, scholars suggest that shared consciousness of kind, a way of thinking about things in brand communities, leads to a number of favorable behavioral consequences such as socializing, customizing, advocating, and expressing (Muniz and O’Guinn 2001; Schau et al. 2009). In order to engage in these brand community practices, consumers must develop the collective understanding and knowledge of value creation shared with other brand community members (Schau et al. 2009). This cognitive social capital is developed through communal consumption experiences in brand communities and imposes a strong sense of obligation or responsibility on consumers to purchase the same brand over time (Muniz and O’Guinn 2001; Schau et al. 2009). Oliver (1999) provides additional support by suggesting that due to the same consumption value and camaraderie shared among consumers in some social organizations (e.g., Harley-Davidson’s Harley Owners’ Group and Green Bay Packers’ cheeseheads), they are willing to be socially integrated into the consumption village (e.g., brand community) that reinforces fully bonded brand loyalty. Combining these considerations leads to the following hypotheses:

H4a

The shared vision among consumers in a brand community has a positive impact on behavioral loyalty.

H4b

The shared vision among consumers in a brand community has a positive impact on purchase frequency.

The impact of CBI

Behavioral loyalty refers to an individual’s behavioral response to repurchase his or her favorite brand over time (Oliver 1999). According to organizational identification theory, consumers with high brand identification are more likely to return a favor to the brand and engage in various supportive behaviors than consumers with low brand identification (Mael and Ashforth 1992; Yoshida 2017). From the consumer’s perspective, such behaviors include repeat purchase, positive word-of-mouth, resistance to brand switching, and willing to pay a price premium (Ahearne et al. 2005; Haumann et al. 2014; Lam et al. 2010; Torres et al. 2018). Also, from the consumer-company identification framework (Bhattacharya and Sen 2003), highly identified consumers will exhibit effort-intensive behaviors (e.g., brand loyalty, brand promotion, customer recruitment, and resilience to negative information) in order to enhance their favorite brand’s long-term welfare. An implication from these theories is that once consumers identify with a particular brand, the connection plays a key role in increasing their behavioral engagement in the brand over time. Since we focus on behavioral loyalty and purchase frequency as behavioral consequences, we develop the following hypotheses:

H5a

CBI has a positive impact on behavioral loyalty.

H5b

CBI has a positive impact on purchase frequency.

The moderating effect of consumer happiness

The joint investigation of CBI and consumer happiness allows us to further explain the CBI-loyalty relationship. To this end, we propose that consumer happiness moderates the impact of CBI on behavioral consequences. This prediction is supported by the broaden-and-build theory (Fredrickson 2001) which suggests that positive emotions, including happiness, are vehicles for individuals’ personal growth and social well-being. According to this theory, psychologically well individuals are more likely to “broaden-and-build” the enduring resources necessary to achieve their goals. More specifically, receiving an increased supply of positive emotions, people are able to build physical (e.g., physical skills and health), intellectual (e.g., expert knowledge and problem solving skills), and social (e.g., social bonds and social support) resources and, as a result, become more healthy, knowledgeable, and socially connected with others. In consumer behavior, consistent positive feelings of happiness during brand consumption may help consumers build the enduring resources (physical, intellectual, and social resources) necessary to exhibit continual behavioral loyalty as a life-long endeavor (Fredrickson 2001). If brand consumption is not a pleasurable activity in life, the relationship between CBI and behavioral loyalty may not be significant. Conversely, the impact of CBI on behavioral loyalty will be stronger if consumers feel that brand consumption greatly contributes to their overall happiness (Guevarra and Howell 2015). Therefore, we expect that consumer happiness amplifies the extent to which highly identified brand consumers remain loyal to their favorite brands in a long-term perspective. These arguments lead us to the following hypotheses:

H6a

Consumer happiness moderates the relationship between CBI and behavioral loyalty, such that the impact of CBI on behavioral loyalty will be stronger among those who have higher levels of consumer happiness.

H6b

Consumer happiness moderates the relationship between CBI and purchase frequency, such that the impact of CBI on purchase frequency will be stronger among those who have higher levels of consumer happiness.

Replication effects

In addition to the hypothesized effects, the impact of brand prestige and brand distinctiveness on CBI will be positive and significant. Organizational identification theory allows for the proposition that individuals tend to (1) identify themselves with prestigious brands and (2) have a strong desire for brand identification by seeking positive differences (e.g., a sharper and more salient definition for brands) between themselves and reference groups (Ashforth and Mael 1989). In order to replicate the relationships reported in previous research (Bhattacharya et al. 1995; Stokburger-Sauer et al. 2014), we expect that brand prestige and brand distinctiveness will have a positive impact on CBI.

Methods

Research setting

The research context was professional football in Japan; this context was suitable for several reasons. First, professional football is a hedonic and experiential product. Hedonic and experiential consumption promotes consumer happiness due to the increased levels of social connectedness (Gilovich et al. 2015) and positive affect (Ryan and Deci 2001). The current setting is an ideal exemplar to examine how CBI and consumer happiness jointly influence consumer loyalty. Second, many football fans share their consumption experiences in their brand communities and tend to identify with their favorite team brands (Mazodier et al. 2018; Oliver 1999). Key variables such as social interaction ties, shared vision, and CBI are readily identified and assessed in this context.

In order to achieve our objectives, we used consumer survey data collected over six months. Data were collected from local residents who lived in the hometown area of a Division 1 team of the Japan Professional Football League (J. League). The key strengths in choosing this setting include (1) measuring football fans’ social capital, CBI, and happiness and (2) collecting data over the course of a season. The present study was conducted throughout the 2017 season.

Data collection

We gathered data from panel surveys conducted by Macromill, Inc., one of the major online research service firms in Japan. The research company randomly sent an invitation email to panel registrants who lived in six cities (Chofu, Mitaka, Fuchu, Koganei, Kodaira, and Nishi Tokyo) in Tokyo because these cities are the official franchise cities set by a professional football team based in Tokyo. One thousand and thirty respondents participated in the survey and answered questions on CBI and its predictor variables. Second, the research company asked the same subjects to rate two behavioral outcomes at the end of the season; data were collected from 618 subjects. Third, we eliminated 244 respondents because their length of time as a fan was zero. Overall, data were retained from 374 local residents, yielding a usable response rate of 36.3%.

In terms of the sample characteristics, 68.7% of the respondents were male. The average age of the subjects was 51.59 years old (standard deviation = 11.62). Age was also categorized into five groups. More than one-third of the subjects were in the 50 to 59 age range (37.2%), 28.9% were between 40 and 49 years old, 21.4% were 60 years old and above, 8.8% were between 30 and 39 years old, and 3.7% were 20 and 29 years old.

Non-response bias was addressed by comparing the respondents to the non-respondents and to the population (Miller and Smith 1983). First, we compared the mean scores of the Time 1 variables between our sample (n = 374) and those who did not participate in the second survey, but followed the club and responded to the first survey (n = 224). No significant difference was found on CBI, brand prestige, and brand distinctiveness, while social interaction ties, shared vision, and consumer happiness were slightly higher for the respondents than for the non-respondents (p < 0.05). Second, we compared sample characteristics with population characteristics. According to the J. League Annual Survey Report (League 2018) which was based on the data collected from 17,136 game attendees of all teams, the gender distribution (male = 61.9%, female = 38.1%) of the national survey was similar to that of our sample (male = 68.7%, female = 31.3%). However, the average age of our sample (M = 51.59) was higher than that of the national population (M = 41.70). We gathered data from local residents that included many elderly people. Collectively, we consider our sample as nearly representative of the overall population for this study.

Measures

We adopted items used to measure the constructs from previous research. Consistent with previous studies (e.g., Bhattacharya et al. 1995; Haumann et al. 2014), we used Mael and Ashforth’s (1992) scale to measure CBI. A three-item scale for measuring brand prestige was adopted from Hur, Kim, and Woo’s (2014) reputation scale. Brand distinctiveness was measured with a three-item scale adopted from the work of Jones and Volpe (2011) and Carlson et al. (2009). Social interaction ties were measured with a three-item scale adopted from the scale used by Chiu et al. (2006). Shared vision was also measured with a three-item scale adopted from Chiu et al.’s (2006) scale.

In order to measure behavioral loyalty, a five-item scale was adopted from two validated scales in the literature. Based on Oliver’s (1999) conceptualization of action loyalty, we selected five items from Zeithaml et al.’s (1996) customer loyalty scale and Garnefeld and colleagues’ (2013) behavioral loyalty scale. Also, behavioral loyalty can also be measured by summing the purchase frequency of consumers over a particular period (Seiders et al. 2005). In this study, we measured purchase frequency by asking the number of games attended by each respondent in the last nine months (Yoshida et al. 2015).

To test the proposed moderating effect, we identified consumer happiness which refers to a positive psychological state that results from a pleasurable consumption experience and contributes to the quality of life (van Boven and Gilovich 2003). We measured consumer happiness with a two-item scale adopted from van Boven and Gilovich (2003). In this study, the survey items for the seven latent constructs were measured on a 7-point Likert-type scale (see Table 1).

Table 1 Scale items and confirmatory factor analysis
Full size table

Back translation

A back-translation procedure (De Wulf et al. 2001; Yoshida et al. 2015) was used to translate the English-based survey items into Japanese. First, one of the authors who is fluent in both Japanese and English translated the English survey items into Japanese. Second, the translated items were then back-translated into English by another bilingual Japanese. To check the accuracy of the translation, a US-born American researcher assessed differences in meaning between the original and back translated instruments. From these procedures, we concluded that the two forms equally reflected each construct domain.

Results

Assessment of the measurement model

To examine the psychometric properties of our measures, we conducted a confirmatory factor analysis (CFA) using Mplus Version 7.31 (see Table 1). The comparative fit index (CFI) and Tucker Lewis index (TLI) were greater than the cutoff point of 0.90 (Hu and Bentler 1999). The values of the root mean square error of approximation (RMSEA) and standardized root mean square residual (SRMR) were smaller than the recommended value of 0.08 (Hu and Bentler 1999). While the ratio of chi-square to degrees of freedom (χ2/df) was higher than Hu and Bentler’s (1999) recommendation of 3.0, overall the fit is acceptable when looking at all the evidence.

Factor loadings (λ), composite reliability (CR) and average variance extracted (AVE) values are also shown in Table 1. Factor loadings ranged from 0.77 to 0.97. The CR values for the seven latent constructs exceeded the recommended level of 0.60 (Bagozzi and Yi 1988). Further, we assessed convergent and discriminant validity by estimating AVE values. The AVE values for the seven latent constructs ranged from 0.69 to 0.87, providing evidence of convergent validity (Fornell and Larcker 1981). Moreover, the AVE value of each construct was greater than the square of its correlation with any other constructs (Table 2). Thus, there was evidence of discriminant validity.

Table 2 Descriptive statistics and correlations (latent constructs)
Full size table

Model comparisons

Using Mplus Version 7.31, structural equation modeling (SEM) was next employed as a test of the predictive power of some of the Time 1 variables (social interaction ties, shared vision, and CBI) on the Time 2 variables (brand loyalty and purchase frequency). However, our hypothesized model does not allow us to examine possible direct effects of brand prestige and brand distinctiveness on the Time 2 outcomes. To fully test the direct impact of the predictor variables, we compared the hypothesized model (Model 1) with a competing nested SEM model (Model 2) which included the direct paths from brand prestige and brand distinctiveness to brand loyalty and purchase frequency. We performed a chi-square difference test and found that (1) Model 1 was a better fit to the data than Model 2 and (2) there was no significant improvement between Model 1 and Model 2: ∆χ2 (∆df) = 8.85(4), n.s (see Table 3). Moreover, none of the added paths from brand prestige and brand distinctiveness to the Time 2 variables were significant.

Table 3 Results of model comparison
Full size table

Furthermore, while we hypothesized social interaction ties influence CBI, this relationship can be reversed because CBI results in social networking behavior (Bhattacharya and Sen 2003). Thus, we conducted an additional chi-square difference test between the hypothesized model (social interaction ties → CBI) and an alternative model (CBI → social interaction ties). The results indicated that the model fit of the hypothesized model was significantly better than the alternative model: ∆χ2 (∆df) = 41.92(3), p < 0.01. Collectively, we concluded that our hypothesized model is adequate for assessing the relationships between the proposed constructs.

Assessment of the structural model

We examined the hypothesized relationships using SEM (see Model 1 in Table 3). The fit indices for the hypothesized model were χ2/df = 3.07, CFI = 0.94, TLI = 0.93, RMSEA = 0.074, and SRMR = 0.039, indicating that the fit measures were acceptable. With respect to hypothesis testing, social interaction ties (γ = 0.21, p < 0.01), shared vision (γ  = 0.54, p < 0.01), and brand distinctiveness (γ  = 0.13, p < 0.05) had positive effects on CBI in support of H1, and H2, whereas the replication effect of brand prestige on CBI was rejected. Also, social interaction ties were found to positively influence both behavioral loyalty (β = 0.38, p < 0.01) and purchase frequency (β  = 0.16, p < 0.05). Moreover, the effects of CBI on behavioral loyalty (β  = 0.37, p < 0.01) and purchase frequency (β = 0.28, p < 0.01) were positive and significant. These results provided support for H3a, H3b, H5a, and H5b, while H4a and H4b were not supported. The extent to which the exogenous variables provide an explanation of variations in the endogenous variables was assessed by R2 values (see Table 3). The R2 values for CBI, behavioral loyalty, and purchase frequency were 0.71, 0.55, and 0.21, respectively.

Our hypothesized model not only demonstrates the direct impact of CBI and its predictor variables on the two outcome variables, but also suggests the indirect impact of the predictor variables on the two outcomes through CBI. To test the indirect effects of the predictor variables on the two behavioral outcomes through CBI, we used the bootstrapping method recommended by Preacher and Hayes (2008). The mediation analysis was performed using Mplus Version 7.31 (see Table 4). A bootstrap estimation with 5000 resamples revealed the 95% confidence intervals (CIs) for the indirect effects of social interaction ties and shared vision on behavioral loyalty through CBI did not include zero. Similarly, the results showed the 95% CIs were greater than zero for the indirect effects of social interaction ties and shared vision on purchase frequency. These findings provide evidence that CBI mediated the effects of social interaction ties and shared vision on behavioral loyalty and purchase frequency.

Table 4 Indirect effects for the hypothesized structural model
Full size table

Moderating effects

To test the moderating impact of consumer happiness on the relationship between CBI and behavioral consequences (predict behavioral loyalty and purchase frequency), we performed a 2 (CBI) × 2 (consumer happiness) analysis of variance (ANOVA, see Fig. 2). The subjects were assigned into the low and high levels of CBI and into the low and high levels of consumer happiness using a median split (medianCBI = 3.0; medianconsumer-happiness = 3.0). Using behavioral loyalty as the dependent variable, we found significant main effects of CBI (F[1, 370] = 37.83, p < 0.01) and consumer happiness (F[1, 370] = 42.94, p < 0.01). Of greatest importance, a two-way ANOVA revealed a significant CBI × consumer-happiness interaction effect (F[1, 370] = 11.25, p < 0.01). Within the high consumer happiness condition, the respondents with high CBI reported higher behavioral loyalty (M = 3.59) than those with low CBI (M = 2.07). However, within the low consumer happiness condition, no difference was found between highly identified (M = 2.07) and lowly identified (M = 1.56) brand consumers. Thus, we found support for H6a.

Fig. 2
figure 2

Moderating effects of consumer happiness on the relationships between CBI and behavioral outcomes

Full size image

We also tested a two-way ANOVA using CBI and consumer happiness as predictor variables and purchase frequency as the dependent variable (see Fig. 2). While the main effects of CBI (F[1, 370] = 3.93, p < 0.05) and consumer happiness (F[1, 370] = 12.38, p < 0.01) on purchase frequency were positive and significant, the two-way interaction of CBI and consumer happiness did not reach statistical significance (F[1, 370] = 3.83, p = 0.051). Thus, H6b was rejected although the results showed a similar tendency to the findings of behavioral loyalty. Collectively, these results indicate that the impact of CBI on behavioral loyalty is contingent on consumer happiness.

Discussion

Theoretical implications

Whetten (1989) contends that providing new insights into previous theories can be considered as theoretical contributions. Heeding this guidance, in the following section, we discuss how this study suggests modifications in existing theories and why such modifications would work with alternative explanations.

First, we extend the brand identification literature by challenging the previous assumption that brand characteristics (e.g., prestige and distinctiveness) are the primary antecedents of CBI (Bhattacharya and Sen 2003; Stokburger-Sauer et al. 2014). We found that CBI was more strongly affected by the two social capital factors than by brand prestige and brand distinctiveness (see Table 3). Based on these results we suggest that CBI is more closely related to the cognitive (shared vision) and structural (social interaction ties) dimensions of social capital than brand prestige and brand distinctiveness. This is consistent with previous studies that suggest (1) building social bonds among consumers in brand communities fosters CBI (Bagozzi and Dholakia 2006) and (2) social benefits, such as the social interactions among fellow brand users, have a stronger effect on CBI than brand prestige and brand distinctiveness (Stokburger-Sauer et al. 2014). Although brand characteristics (e.g., prestige and distinctiveness) have helped to shape our understanding of an important route to CBI, the structural and cognitive dimensions of social capital that consumers possess in pursuit of their consumption goals in brand communities represent a new perspective on why people identify with brands.

Second, this work is unique because it investigated the mediating role of CBI in a predictive study. Since little effort has been made to examine the impact of CBI and its antecedents on behavioral consequences over time (Haumann et al. 2014; Huang et al. 2017), the current study contributes to the brand management literature with a more precise explanation of the hypothesized relationships. The results (Table 4) support the notion that the structural (social interaction ties) and cognitive (shared vision) dimensions of social capital first influence CBI which in turn predicts behavioral loyalty and purchase frequency. These findings are consistent with the theoretical perspective of Tsai and Ghoshal (1998) who suggest that social interaction ties and shared vision induce more loyalty to an organization through building relational social capital because close social interactions and common goals allow individuals to share important information, promote the harmony of social relationships, and eventually result in ongoing organizational support. Our study provided initial support for such sequential relationships in brand management research.

Our results also indicate that consumers’ behavioral consequences are more strongly associated with their social networking behavior (structural social capital) and CBI (relational social capital) than with shared vision (cognitive social capital). The results provide evidence for Oliver’s (1999) village concept, suggesting if a brand not only reflects a consumer’s personal identity, but also involves rich social interactions with other consumers, fully bonded loyalty emerges as a blend of self-concept with social bonds in the consumption village. Therefore, fostering CBI is not enough wherein providing opportunities to socialize with other consumers can also enhance behavioral brand loyalty.

In addition to the aforementioned points, our research contributes to the brand management literature by explaining how the impact of CBI on future behavioral loyalty is contingent on consumer happiness. Our moderator analysis indicated that CBI significantly influenced behavioral loyalty among those with high consumer happiness. Also, the moderating effect of consumer happiness on purchase frequency approached significance (p = 0.051), leading us to speculate happy consumers are more likely to have higher purchase frequency over time. These findings are consistent with the view that individuals with higher levels of happiness are more interested in and committed to the activity they perform (Frey 2017). A growing literature has emerged to explain how happiness is related to positive behavior (e.g., Fredrickson 2001; Luthans et al. 2008). For happy consumers, their brand consumption becomes more meaningful, aspirational, and sustainable (Huang and Rust 2011; Sheth et al. 2011) because such consumers can build and use their resources (e.g., physical, intellectual, and social resources) to achieve their desirable consumption goals (Fredrickson 2001). Thus, it seems reasonable to believe that consumer happiness improves the impact of CBI on consumers’ long-term pursuit of behavioral loyalty. Collectively, this study is a compelling extension of previous CBI research in which the role of consumer happiness was not explicitly discussed.

Although we did not expect that the direct impact of brand prestige on CBI would be insignificant, this finding is also consistent with previous research. Jones and Volpe (2011) did not find a significant effect of organizational prestige on organizational identification because when competitors have equivalent or better prestige, organizational prestige may not lead to organizational identification. While ranked in the top five at the beginning (March) of the 2017 season, the focal football team dropped to tenth in the standings (out of eighteen teams) during the first data collection (July) and finished thirteenth at the end of the season (December). Thus, a possible explanation is that the team’s poor performance damaged the prestigious reputation of the team and prevented the consumers from evaluating team prestige at a high level. While brand prestige is context-specific, consumers’ brand identification is stable and enduring as many sports fans’ identification with their teams is grounded in childhood (James 2001).

Managerial implications

Our conceptual framework and the study results signify that managers can enhance CBI and behavioral consequences through consumer-to-consumer social capital (i.e., social interaction ties and shared vision). For example, practitioners can expect that providing social interaction opportunities in both face-to-face and online social networks and communicating information about their mission, vision, and values will lead to higher levels of CBI and behavioral loyalty over time. In particular, service managers should recognize that consumer-to-consumer social capital helps ensure higher levels of CBI as well as consumer loyalty. From a practical standpoint, one implication from this study is that service managers need to make an effort not only to improve their brand characteristics (e.g., prestige and distinctiveness), but to also include consumer-to-consumer social interactions in their long-term loyalty strategies. Evidence of this can be found in the realm of sports where the inclusion of plazas at sports stadiums as social spaces for fans to gather in and around the stadium is a recent design trend to foster fan-to-fan connectivity as well as provide revenue opportunities year round (Muret 2016).

The results of this study also indicate that consumer happiness enhances the impact of CBI on future behavioral loyalty. Therefore, practitioners should place great emphasis on the importance of consumer happiness in their marketing actions. For instance, the utilization of marketing messages with an emphasis on well-being may foster a positive brand image among consumers. An example of such branding can be seen in the fact that many companies attempt to associate their marketing campaigns with Goal 3 (good health and well-being) of the United Nations’ sustainable development goals (SDGs). Also, another practical application of this study can be carried out as an advertising campaign that shows attractive photographs and graphics featuring happy brand consumers. It would be effective to send these consumers’ visual messages via websites, social media, and mobile platforms as well as traditional media including television, newspapers, and magazines. Furthermore, in order to achieve long-term brand loyalty, brand consumption must be meaningful and sustainable for consumers in the face of societal grand challenges such as climate change, aging society, international migration, racial discrimination, and the COVID-19 pandemic. Marketing communications must be authentic and credible to make a meaningful contribution to the lives of consumers even during adverse times.

Limitations and directions for future research

This study has several limitations that can be addressed in future research. First, this study was conducted in a professional football context. One limitation of the present study is the lack of generalizability of the findings to other settings. The consumption of professional football is hedonic (Funk and James 2001) and experiential (Yoshida 2017). With an increased emphasis on hedonic consumption (Hirschman and Holbrook 1982) and experiential consumption (Pine and Gilmore 1998; Schmitt et al. 2015) in developed countries, hedonic and experiential products are important in the pursuit of happiness among modern consumers (van Boven and Gilovich 2003; Zhong and Mitchell 2010). In future studies, researchers should carefully examine the interaction effect of CBI and consumer happiness on brand loyalty in various contexts of hedonic and experiential products.

Second, the proposed framework was tested in the context of Japanese professional football. Considering that Japan has a collectivist culture (Hofstede 2001), it should be acknowledged that this collectivism might strengthen the relationships between the proposed constructs. In particular, this cultural characteristic may have inflated the predictive power of social interaction ties and shared vision for CBI and behavioral consequences. It will be interesting to replicate this study in a different cultural context.

Third, another limitation is the omission of potentially important variables. While we integrated organizational identification theory and organization social capital theory in our research model, additional variables might influence behavioral loyalty and purchase frequency. For example, previous research has shown that consumer satisfaction, commitment, and engagement positively affect the behavioral consequences of consumer-brand relationships (Pansari and Kumar 2017). It will be interesting to see how future researchers examine the simultaneous effects of these attitudinal and behavioral constructs on consumers’ future consumption behavior.

Conclusion

This study represents an initial effort to provide evidence of the predictive validity of CBI and its antecedents in relation to consumer happiness and future behavioral consequences. Particularly, incorporating social capital and consumer happiness perspectives into CBI research extends scholars’ and managers’ ability to better understand and explain the predictive power of CBI and its antecedents over time.