Henri Tajfel’s greatest contribution to psychology was the social identity theory. Social identity is an individual’s sense of who we are, based on affiliation(s) to one or more groups. Tajfel et al. (1979) proposed that all groups (e.g., social class, family, soccer team, etc.) to which people belong are an important source of pride and self-esteem. Groups give us a sense of social identity: a sense of belonging to the social world.

We usually divide the world into “them” and “us” through a process of social categorization (for example, we classify people into social groups). Tajfel et al. (1979) found that stereotyping (classifying people into groups and categories) is based on a natural human cognitive process. We have a natural tendency to group things together. In doing so, we often tend to exaggerate the differences among groups, and the similarities of characteristics in the same group.

This categorization is known as in-group (us) and out-group (them). The central hypothesis of social identity theory is that members of a group will seek to find negative aspects of an individual outside their group of belonging, improving their self-image (Islam, 2014).

Social identity theory claims that polarization of an individual (or group) is more an emotional characteristic than ideological. Moreover, self-identity, understood as being congruent with a particular group, may a produce positive sentiment for the group with which the person identifies, and a negative one, sometimes even fiercely so, for distinct and oppositional groups (Hetherington, 2009).

The relationship between social identity theory and consumption is already established identifying that ownership of a particular product can symbolically extend the consumer’s self (Belk, 1988). Thus, people may hold the belief that their identity is the sum of their possessions. The more consumers believe in and value their material possessions, the more part of the self these objects become (Belk, 1988). The greater the perceived congruence between aproduct's brand and the consumer’s self, the stronger the slef concept of the individual becomes (Santos & Santos, 2017). Consumers increasingly understand that the self is built through consumption, and that consumption expresses their true self (Ramalho & Ayrosa, 2009).

Often, the effect of personal ideologies on the politics of individuals reveal in turn their personality tendencies and psychological characteristics. Thus, individuals can manifest higher political ideological tendencies towards conservatism, being more resistant to change. Furthermore, more conservative people are more likely to find justifications for the system and the social order than individuals who are more liberal (Jung et al., 2017).

Regarding decision-making, such as purchase process, conservative individuals (politically right-wing) are more resistant to switching brands, for example. However, liberals (left-wing people) are more likely to complain about the brand when facing some dissatisfaction (Jost, 2017; Jung et al., 2017). Thus, the first hypothesis of this research is:

H1. Individuals who declare themselves to belong to left-wing ideology (versus that of the right) will present lesser (versus greater) purchase intention, in the case of brand support for an ideologically opposing politician.

Possibly, political polarization will bring about different results in brand evaluation, depending on the self-congruence of the individual and their involvement with the brand.

The notion of self-concept is defined as the sum of each person’s thoughts and feelings about themselves and can be taken in two forms: the “real self” and the “ideal self”. The real self describes how an individual perceives him/herself (“I as I am”), while the ideal self-concept explains how an individual wants to perceive him/herself (“the perfect self”) (Choi & Rifon, 2012; Jeong & Koo, 2015). Self-congruence occurs when there is a high level of consistency between the “ideal self” and the “real self”.

Individuals can be led to use a brand to express a self-concept or a concept that they consider ideal forthemselves (Japutra et al., 2019). For example, a person can use a brand that promotes environmental sustainability to show their own agreement with sustainable development (the real self-concept), or use another brand which is considered luxurious to express an ideal for their social life (ideal self-concept) (Islam et al., 2019). In this way, perceived self-congruence, whether real or ideal, leads the consumer to greater involvement with the brand, and to a tendency to defend and choose to purchase that brand.

The expectation of social belonging leads to the creation of greater affective bonds with a brand, when the consumer realizes that other individuals from groups to which he or she wants to belong also admire and consider themselves self-congruent with the same brand (Wallace et al., 2017). This phenomenon generates the most involvement with the brand and may lead to brand loyalty.

Probably, consumers have favorable attitudes towards brands whose images are congruent with their self-images (Richins, 1994), so the greater the perceived self-congruence with the brand, the greater the consumer’s involvement with it (Xu, 2008). Based on this, we elaborated the second hypothesis of this study:

H2a. Self-congruence has a positive effect on brand involvement.

Literature shows that hatred towards the brand can be greater when there is a detachment in the perception of self-congruence, than in a person who did not identify as self-congruent to the brand (Islam et al., 2019). Because of this, there are hypotheses 2b and 2c.

H2b. Self-congruence has a positive effect on brand hatred in the case of brand support for an ideologically opposed politician.

H2c. Self-congruence has a negative effect on brand PI, in the case of brand support for an ideologically opposed politician.

Strong consumer self-congruence can generate greater involvement, which evokes involvement with a brand (Japutra et al., 2018). Managers of different brands hope their consumers establish affective and identificative bonds with their products and services, as these bonds tend to create longer term brand loyalty. However, in the case of transgression, as perceived by the consumer in relation to the brand with which he or she finds themselves self-congruent, the feeling of betrayal can generate brand hate (Hung & Lu, 2018).

Hate is essentially a primary emotion. However, on some occasions, it can be a secondary emotion (Zarantonello et al., 2016). Plutchik (1991), for example, describes hate as a secondary emotion, resulting from the combination of disgust and anger, which are considered primary emotions. According to Sternberg (2003), hate is a multifaceted emotion, based on three components: devaluation, anger, and denial of intimacy.

For most psychologists, hate results from the violation of moral codes (Zarantonello et al., 2016). Sternberg (2003) argues that all the emotions that combine to form hate (disgust, dislike, anger, fear, and contempt) can arise after the violation of individual or community rights. Therefore, these emotions can be perceived as imminent threats to people, to their freedom, their well-being, and their preservation.

Several aspects of this negative emotion have been discussed in marketing (Kucuk, 2016; Zarantonello et al., 2016; Hegner et al., 2017; Zarantonello et al., 2018). Brand hate can result in the desire for revenge, brand avoidance (Grégoire et al., 2009), and the non-purchase of products in the future (Banerjee et al., 2020). Considering these, we hypothesize that:

H3. Individuals with more (versus less) brand hate will declare a lower (versus greater) PI of the brand, in the case of support of the brand to an ideologically opposed politician.

Even though individual self-congruence has a direct effect on hate, we believe that involvement with the brand can also influence this relationship.

Consumers distinctly the same product and/or brand based on their levels of involvement (Zaichkowsky, 1985), and self-congruence can lead to greater consumer engagement with the brand. However, there are studies indicating that real congruence (which is related to individual self-concept and real personality characteristics), unlike ideal congruence (related to the social construction of the self), may not influence the generation of attachment and involvement (Huang et al., 2017; Japutra et al., 2019).

With that we propose:

H3a. Individuals more (versus less) involved with the brand will declare lesser (versus greater) brand hate, in the case of the brand supporting an ideologically opposed politician.

H3b. Individuals more (versus less) involved with the brand will declare lesser (versus greater) PI of the brand, in the case of the brand supporting an ideologically opposed politician.

Seminal studies have already established that the influence of personality traits affects consumer behavior (Kassarjian, 1971; Haugtvedt, Petty, & Cacioppo, 1992). Several studies reveal the regulatory focus structure (Higgins, 1997, 1998), which has gained increasing interest in academic literature, as it is related to an impressive diversity of consumption behaviors (see Higgins & Spiegel, 2004). In this study, we sought to verify the effect of the individual’s regulatory focus as a moderator of the process among self-congruence, involvement and hate in the purchase intention of a brand with political polarization antagonistic to consumer’s belief.

The regulatory focus theory is a theory of motivation and self-regulation with significant prominence in consumer research. This theory explains a variety of consumer decision behaviors (Silbiger et al., 2021). According to regulatory focus theory (Higgins 1997), consumers vary in how they see their goals and adopt different ways to try to achieve them (Brockner et al., 2004). Specifically, regulatory focus theory posits two separate types of regulatory guidance, promotion and prevention focus, which consumers adopt while pursuing goals.

Individuals with a promotional focus tend to consider the gain objective, leaning towards risk-taking to achieve this objective (Higgins, 1998; Hüttermann et al., 2018). Contrary to this, , individuals with a preventive regulatory focus aim on “not to lose”, which can make them have a greater intention to stay in inertia or to make the safest possible decisions (Lin & Chen, 2015).

Despite being a personal characteristic, regulatory focus can be activated according to the circumstances (Boldero & Higgins, 2011; Westjohn et al., 2016). Even though Sengupta and Zhou (2007) state that regulatory focus cannot be considered a characteristic of the individual’s personality, there are indications (both theoretical and empirical) that the regulatory focus is a condition that can be both a personality trait, and, depending on a situation, induced into a momentary state (Pham & Avnet, 2004; Kim et al., 2021). Regarding this divergence, a current theoretical line believes that, depending on how the process of socialization and cognitive formation took place, there will be a predominance of one of the conditions. With that, the individual will tend to have a more promotional or a more preventive behavior (Higgins, 1997; Binswanger, 2001).

As already established, consumers can be classified into one of the types of regulatory focus. Accordingly, it is believed that the regulatory focus can be a good moderating variable regarding the purchasing decision relationships to which consumers are exposed (Dodoo & Wu, 2021). Thus, we built the following hypotheses:

H4. In a situation of brand support in preference to an ideologically opposed politician, the individual’s regulatory focus will moderate...

H4a: ...positively (versus negatively) the relationship between self-congruence and PI in promotional individuals (versus preventive).

H4b: ... positively (versus negatively) the relationship between involvement and PI in promotional individuals (versus preventive).

H4c: ... negatively (versus positively) the relationship between brand hatred and PI in promotional individuals (versus preventive).

Figure 1 presents the conceptual model of this research.

Figure 1. Source: theauthors

The stimulus of the scenario occurred in two moments. In the first, participants were asked to indicate the brand they most admired, and with which they had a good relationship, with the guidancee “Think of a brand. A brand you admire, with which you have a good relationship and which you buy whenever possible. Now write the name of this brand and two features you admire about it”.

Then, we asked the participants to assess self-congruence, after stimulating ideological polarization. The guidance was “Now, think about the following situation: you saw a [true] news story about your favorite brand. The news said that your favorite brand is funding a politician you don’t like, a politician who goes against all your ideals, meaning a politician who would never get your vote. This politician’s ideas are so contrary to your own that you would never vote for anyone in his or her party!). Finally, participants were asked to answer the remaining items of the scales and fill in their demographic profiles.

For all items on the scales of self-congruence (three items by Sirgy et al., 1997), hate (measured through six items by Hegner, 2017), purchase intention (three items also by Hegner, 2017) and involvement (10 items from the personal involvement inventory scale revised by Zaichkowsky, 1994), we used a seven-point Likert scale (anchored on 1=strongly disagree and 7=strongly agree) for measuring. These scales were adopted because: a) they are suitable for measuring the model’s constructs; b) they showed good psychometric consistency in the original studies; and c) they have already been replicated in later studies, with the same consistency levels.

The individual’s regulatory focus was measured through 18 items (proposed by Higgins, 1997), with nine items to measure the promotional regulatory focus and nine items to estimate the preventive regulatory focus. For the measurement, we used the sum of the options performed (-1 for preventive statements and +1 for promotional statements). The Higgins’ scale (1997) is the most used in academic studies in order to estimate the regulatory focus.

Finally, we measured political ideology with one item, using a 4-point categorical scale (1 = clearly left; 2 = more left than right; 3 = more right than left; 4 = clearly right). All items used can be seen in Table 1.

Table 1

Items
Self-congruence - Sirgy et al. (1997)
People similar to me use this same brand that I use.
This brand is consistent with the way I see myself.
This brand represents who I am.
Hate - Hegner et al. (2017)
I’m disgusted with this brand.
I do not tolerate this brand or your company.
The world would be a better place without this brand.
I am very angry at this brand.
This brand is horrible.
I hate this brand.
Purchase intention - Hegner et al. (2017)
I no longer buy products from this brand.
It is very likely that I will buy this brand in the future.
I will buy this brand the next time I need a product from this category.
Involvement - Zaichkowsky (1994)
This brand is important.
This brand is irrelevant.
This brand is useful.
This brand is beneficial.
This brand is interesting.
This brand is exciting.
This brand is fascinating.
This brand is essential.
This tag is unnecessary.
This brand is boring.
Regulatory focus - Higgins (1997)
In general, I care more about the success I hope to achieve in the future - Promotion
In general, I’m more concerned with preventing bad things from happening in my life - Prevention
I always think about what to do to succeed - Promotion
I feel anxious imagining that I might not be able to fulfill my responsibilities and obligations - Prevention
I always wonder what to do to make my dreams come true - Promotion
I always think about the person I wouldn’t want to become in the future - Prevention
My biggest priority now is to achieve my goals in life - Promotion
I always worry that maybe I can’t reach my subjective life - Prevention
I always think about the person I would ideally like to be in the future - Promotion
I always imagine myself experiencing bad things that could happen in my life - Prevention
I always picture myself experiencing the good things that could happen in my life - Promotion
I always think about what to do to avoid failure - Prevention
I see myself as someone who is trying to achieve the “ideal me” - fulfilling my hopes, desires and aspirations - Promotion
In general, I’m more concerned with avoiding negative outcomes for my life - Prevention
All in all, I’m much more focused on achieving success than preventing failure - Promotion
My biggest priority right now is to avoid failing - Prevention
In general, I’m more concerned with achieving positive results for my life - Promotion
I see myself as someone who is trying to achieve the "me that I should be" - fulfilling my tasks, responsibilities, and obligations - Prevention
Political ideology Ramos et al. (2020)
What is your current political position?

Source: Adapted by the authors

We sent the data collection instrument online to participants through the QuestionPro platform. The link to access the data collection instrument was sent through the research authors’ virtual social networks. The instrument did not ask for the respondent’s identification, which resulted in the voluntary participation of all. Respondents were also asked to send the link access to their network of contacts. Thus, the sample is considered as non-probabilistic for convenience, and collected using the snowball technique (Silva et al., 2014). Even though non-probabilistic samples do not allow the generalization of results, the collection strategy resulted in an adequate sample, as all participants are voters and are responsible for their purchasing decisions (100%) and most for the home purchases (77.18%).

Data collection occurred in February 2021.

We used descriptive statistics to analyze the distribution of sample characteristics, and exploratory factor analysis for the identification of the unidimensionality of the items and the multidimensionality of the constructs (Sousa & Herrero, 2021). Therefore, as recommended in the literature, we sought to identify a Kaiser-Meyer-Olkin (KMO) criteria greater than .60, a significant Bartlett’s sphericity test (p<.05), factor loadings greater than .50, and cross-loads less than .40 (Levin & Fox, 2004). As for the internal consistency of the scales, we used Cronbach’s Alpha, with a critical limit of .60 (Malhotra et al., 2014).

We performed the test of difference between the constructs using Student’s t test for independent samples. Finally, for the analysis of direct relationships, moderations, and mediations, we used model 89 of the PROCESS macro (Hayes, 2017). For all analyzes, we used the software SPSS for Windows version 23.

The main theoretical contribution of this study is the analysis of the effect of the individual’s regulatory profile - preventive versus promotional (Higgins, 1997; 1998) - as a moderator in the relationship between political polarization and the purchase intention declared by the consumer. Previous studies have shown that preventive individuals are more faithful to the brands they buy, while promotional individuals are more likely to adopt new consumption experiences and brand substitution (Lopes & Veiga, 2019). Thus, we verified whether the preventive (versus promotional) regulatory focus would increase (versus reduce) the PI of individuals declared to be left-wing (versus right), in situations of asymmetry between the political support of a brand and their beliefs individual (and vice versa).

Our findings go further. We found no evidence that consumer political ideology directly affects self-congruence. However, consumers who are self-congruent with the brands report greater involvement and greater hate. At first glance, this result may seem controversial. However, in situations of brand transgression, the individual with greater attachment may feel more betrayed (Hung et al., 2018), devalued (Sternberg, 2003), and may perceive symbolic incongruity, and ideological incompatibility (Hegner et al., 2017; Zarantonello et al., 2018), evoking emotions such as disgust, anger, and hatred (Hung et al., 2018; Plutchik, 1991; Zarantonello et al., 2018).

Previous research has linked individuals’ regulatory focus to their political ideology, linking the focus of prevention with liberalism (Janoff-Bulman, 2009), or conservatism (Chan et al., 2019), which, in our research, it makes more sense. Although there is no direct effect between regulatory focus and political ideology of individuals, our results reinforce the theory that people who are self-righteous, as well as individuals with a preventive focus, tend to need stability, avoiding uncertainties (Ashford & Cummings, 1983; Lanaj, Chang, & Johnson, 2012). Following this theoretical perspective, our study shows that the regulatory focus moderates both the relationship between involvement and purchase intention and the relationship between brand hate and purchase intention. Therefore, individuals with a focus on promotion, in a context of brand transgression, declared greater intention to abandon the brand.

Again, through the results obtained, the arduous task of managers and marketing practitioners becomes evident. Electoral sponsorship, or financing, can be a possible form of democratic participation for an organization, or even a form of “social support” given to a politician or public manager, through public relations actions and other marketing communications. However, as there is a strong rivalry in the ideological polarization in Brazil (and in much of the world, as can be seen in Ezzamel et al., 2007), these actions result in a managerial risk. Even though the individual’s regulatory focus can ease brand hate, aroused by an ideological asymmetry, there is no evidence that focus moderation can fully mitigate this emotion. With that, it is possible to recommend extreme caution in positioning and political support by a brand.

A brand's caution towards its political positioningis already a contribution that can help marketing managers and administrators in general. Furthermore, even if the moderation of the individual's regulatory focus has not resulted in a process of total mitigation of the hate effect in the PI process in the scenario we stimulated, we consider that the literature indicates the possibility of regulatory focus manipulation (adjustment regulatory framework, as presented by Cesario et al., 2004; Aaker & Lee, 2006) as a way to change, even temporarily, the individual regulatory focus. With this in mind, managers and practitioners can use environmental or communication stimuli as ways to influence regulatory adjustments, in order to enhance the moderating effects played by this variable to reduce the hate aroused by the ideological asymmetry of a possible action of political support.

We adopted procedures to guarantee the validity of this study. However, there are limits that must be recognized such as the characteristic of the non-probabilistic sample. Even if non-probabilistic samples are widely used in studies from the field of applied social sciences (Heckathorn & Cameron, 2017; Rahi, 2017), the observed results cannot be generalized. Even though our sample is adequate for analyzing the proposed model, the concentration of young individuals (≤ 25 years old) may have influenced the magnitude of the results. Therefore, one recommendation is the use of probabilistic samples or, if these are not possible, the analysis of data from an older sample as a way to verify any possible difference regarding our findings.

Evidently, hate may not be the only possible mediator of the relationship between involvement and PI. It would be interesting for future studies to analyze the effect of other emotional manifestations (such as frustration, disappointment, or contempt) on the same process.

Finally, with the maturity of the theoretical fields of applied social sciences, the dependent variables of the models studied must increasingly have behavioral characteristics. Hulland and Houstou (2021) reinforced the importance of behavior for academic studies in order to gain external validity. With that in mind, a recommendation for future studies is to replace the purchase intention with a behavioral variable resulting from the moderate mediation process analyzed in this research.