“All employees are equal … but some are more equal than others”. Role identity and nonfamily member discrimination in family SMEs

Abstract

Purpose

This paper aims to investigate if, under which conditions, and with which consequences, nonfamily members have the perception of being discriminated against as a consequence of nepotism and adverse selection practices. This research also aims to investigate whether the carried-out role influences the perception of being discriminated against among nonfamily member employees.

Design/methodology/approach

A quantitative approach was carried out by adopting a structural equation model (SEM) analysis. The survey investigated a sample of Italian family SMEs (participating companies N = 186, total questionnaires collected N = 838).

Findings

Drawing on the multiple identities theory, findings show that role salience (RS) effectively contributes to reducing the unwanted effects of perceived discrimination (PD) among nonfamily member employees. In doing so, this study deepens the knowledge of nonfamily member employment conditions and their consequences on strategic outcomes such as organizational commitment (OC), organizational justice (OJ) and intention to quit (ITQ).

Research limitations/implications

By adopting a self-categorization approach, this study also advances current theoretical literature, as this methodological lens could help scholars further understand diversity in family business.

Practical implications

This study suggests it would be advisable to implement human resource management practices based on job rotation to promote cohesion and reduce perceived distances.

Social implications

SMEs are the most widespread type of firm in the world; as a consequence, avoiding PD among nonfamily member employees has general ethical relevance.

Originality/value

This study expands current literature by showing that RS plays an important role in determining levels of PD. This study also advances current literature by focusing on the impact of multiple identities on fairness and commitment at individual and group levels of analysis of family businesses.

Citation

Ferrari, F. (2024), "“All employees are equal … but some are more equal than others”. Role identity and nonfamily member discrimination in family SMEs", Journal of Family Business Management, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/JFBM-03-2024-0049

1. Introduction

Many researchers have suggested that family businesses exercise unfair practices in their workplace by offering preferential treatment to family employees (e.g. Chua et al., 2009; Cruz et al., 2014; Kellermanns et al., 2012; Samara and Arenas, 2017; Van der Heyden et al., 2005; Zientara, 2017). In addition, Chua et al. (2009) assert that family business owners are motivated to maintain more subjective and informal practices to protect the privileges of family members (Ferrari, 2019), i.e. to protect their economic goals but also their socio-emotional wealth (SEW) (Gomez-Mejía et al., 2007). Pursuing noneconomic goals can drive the owner family to generate nonfamily member discrimination, as previously highlighted (Chrisman et al., 2014), thus undermining fairness levels. In spite of this ample body of knowledge, however, until now discrimination of nonfamily members and its causes and consequences have not been adequately investigated (for a review, see Tabor et al., 2018; see also Rosecká and Machek, 2022).

Different theoretical approaches adopted to date (agency theory and SEW) have shown that family businesses largely adopt practices of nepotism and adverse selection (AS) (Barnett and Kellermanns, 2006; Lubatkin et al., 2007; Schulze et al., 2001). Such practices of favoritism of family members could explain the perceived discrimination (PD) among nonfamily member employees, but literature has not offered definitive results (Chrisman et al., 2007; De Massis, 2012; Ferrari, 2014), and a theoretical framework capable of effectively explaining and predicting favoritism towards family members has still not been developed. In their review, Tabor et al. (2018) highlight that injustice in family firms may not be as prevalent or problematic as assumed, depending on many situational and relational factors, and because fairness norms may function differently in family firms (Kidwell et al., 2012; Lubatkin et al., 2007). Other authors (e.g. Samara and Paul, 2019) conclude that scholars and practitioners face a complex situation where the preferential treatment given by family employers to family employees compared to their nonfamily employees is at times perceived as fair and at other times discriminative. What is currently missing is therefore an interpretative framework that explains under what conditions discrimination occurs.

Recent theoretical developments (Bettinelli et al., 2022) focus on identity as a key concept for analyzing and understanding interpersonal dynamics in family firms. This approach is promising and reinforces the idea that identity is a fundamental aspect of SEW which affects behaviors and interpersonal/intergroup relationships (Ferrari, 2020). In social psychology, identity is the basis of many cognitive and social processes, as well as mechanisms that drive human behavior in organizations (Brown, 2020). Considering for instance the behavior of the business owner family, social psychology warns that social categorization (i.e. the process by which people 'categorize' themselves as belonging to a certain group rather than another: see Tajfel and Turner, 1986; Turner, 1987) could trigger favoritism towards family members and discrimination against nonfamily members. Moreover, it is the specific context that may lead family and nonfamily members to pay attention to their carried out role (Chrisman et al., 2007; Lubatkin et al., 2007) rather than to the group they belong to. This complex scenario, due to the overlapping between family and business and between personal/social and professional issues, provides a unique opportunity to investigate identity (Bettinelli et al., 2022; Whetten et al., 2014). As multiple identities may coexist in individuals (Cunningham, 2020; Sundaramurthy and Kreiner, 2008), recent developments of the social identity theory (Deaux et al., 1995, 1999, Deaux, 2011), proposing the concept of role salience (RS), suggest that the category to which one has the perception of belonging depends on contextual factors. Consequently, the performed job may be more salient than membership to family/nonfamily group, thereby reducing PD. The concept of RS could therefore offer a good lens through which to identify the dynamics underlying discrimination and lay the foundations for appropriate managerial strategies.

Given the current theoretical scenario and its limitations, this research has two main purposes that are closely interconnected. Firstly, it aims to test if, and with which consequences, nonfamily members have the perception of being discriminated against as a consequence of nepotism and AS practices. In doing so, it deepens knowledge of nonfamily member employment conditions as advocated by Tabor et al. (2018) and their consequences on strategic outcomes such as organizational commitment (OC), organizational justice (OJ) and intention to quit (ITQ). Secondly, by applying a social identity theory approach (Deaux et al., 1995, 1999, Deaux, 2011) in explaining discriminatory dynamics towards nonfamily member employees, this study also aims to expand the current research agenda on group identity in family firms, as recently called for by Bettinelli et al. (2022).

2. Theoretical background

One of the main challenges facing family firms is ensuring fairness between family and nonfamily employees in the workplace (Samara and Arenas, 2017; Samara and Paul, 2019). The primary source of organizational injustice seems to be the overlap between family and business (Lansberg, 1989). Verbeke and Kano (2012) use the terminology “bifurcation bias” to describe the asymmetric treatment of family and nonfamily members in the family firm. The literature describes the negative impact phenomena have on fairness providing examples such as: nepotism (Padgett and Morris, 2005), authoritarianism (Tagiuri and Davis, 1992), management practices, acts of discrimination towards nonfamily members by family member employees (Barnett and Kellermanns, 2006; Lubatkin et al., 2007; Schulze et al., 2001), perception of nonfamily members as “foreign” (Barnett and Kellermanns, 2006; De Massis, 2012) and organizational culture centered on the founder (Schein, 1983). Family employees receive better performance appraisals (Verbeke and Kano, 2012), are overcompensated financially (Chua et al., 2009) and are provided with better leadership opportunities (Covin, 1994) compared to their nonfamily counterparts. Moreover, nonfamily employees are often considered as ineligible for stock option rewards (Gedajlovic and Carney, 2010) and are excluded from opportunities for succession (Lubatkin et al., 2005). In addition, literature shows that family members have better compensation packages (Jennings et al., 2018) and better training and development opportunities (Birdthistle, 2006). Furthermore, since parental altruism in literature is described as asymmetrical (Bergstrom, 1989), the family members can take advantage of this altruism (Ferrari, 2017, 2020), thus generating dissatisfaction in nonfamily members who cannot do the same (Klein and Bell, 2007).

2.1 Literature review and research hypotheses

Nepotism, AS (e.g. selection and career orientation of family members using criteria of belonging and not of expertise/merit) and other discriminative practices against nonfamily members are hence well documented by empirical literature (Ferrari, 2023a for a review, see Tabor et al., 2018). What is not especially clear are the conditions under which family member favoritism is perceived as discrimination by nonfamily member employees and its organizational consequences (Carsrud, 2006). For instance, Jennings et al. (2018) indicate that “bifurcation bias” is overstated because firm owners are unlikely to engage in practices that are perceived as unjust because they perceive nonfamily members as “almost-family”. Tabor et al. (2018) concluded although some family firms may engage in unfair treatment of nonfamily members, recent research questions whether these practices are widespread. In addition, it is not clear what are the factors that, given the same discriminatory practices adopted, cause these practices to be at times perceived as fair and at other times unfair (Samara and Paul, 2019).

In the family business field of research, the relationship between the employer and employees has been widely investigated by adopting different theoretical models. For instance, early work in Agency Theory (Jensen and Meckling, 1976; Ross, 1973) ascribed a favorable attitude towards family member employees, suggesting that the sense of belonging perceived by family members would protect the company from any opportunistic behavior by those members, differently from nonfamily member employees, who may be more prone to the risk of opportunistic behaviors (Kallmuenzer, 2015). This position has been challenged by the economy of the family approach, (Becker, 1981), which posits that family firms are characterized by the opportunistic behavior of the next generations and parental altruism, resulting in AS. Verbeke and Kano (2012) theorize that “bifurcation bias” exists when family owners and managers treat nonfamily members as short-term agents, but family members as long-term stewards. However, Chrisman et al. (2007) show that the family member employees in fact do act as agents, contrary to the expectations of stewardship theory. More recently, to overcome this inconclusive theoretical framework, it has been suggested that a peculiar characteristic of family firms is the desire of family business owners to preserve their SEW (Gomez-Mejía et al., 2007). At its core, SEW represents the stock of affect-related value that the family gains from its involvement in the business. It includes an emotional attachment to the firm, a close identification with its name, a desire for family influence and control, finally resulting in intra-family succession (Ferrari, 2019, 2020, 2023a, 2023b; Berrone et al., 2012), thus showing preference for family members as viable successors, regardless of their merit (Ferrari, 2017, 2020, 2023b).

In summary, theoretical approaches describe intra-family dynamics resulting in AS (Lansberg, 1983; Schulze et al., 2003; Chrisman et al., 2004; Schulze et al., 2001). AS can therefore be considered as a characterizing issue in the relationship between the employer and their children/relatives in family businesses. Facing or even simply witnessing AS dynamics that benefit family members should lead nonfamily members to perceive the work environment as discriminatory (Rupp and Spencer, 2006). In addition, AS, which by definition overlooks the principle of merit or competence in recruitment and career path, could in turn impact (OJ) (Colquitt et al., 2013; Samara and Arenas, 2017; Samara and Paul, 2019). As a further consequence, parental altruism can lead to assigning benefits (economic and otherwise) to family members regardless of their merit, consequently laying the groundwork for future deviant behavior (Eddleston and Kidwell, 2010; Ferrari, 2020, 2023b) and at the same time discriminating against nonfamily members. Based on the above considerations, it is possible to make the following hypothesis.

H1.

AS increases the PD level

In addition to being a consequence of AS, PD itself could negatively affect OJ (Enoksen, 2016). The members of an organization obviously want benefits, economic and otherwise, but in addition, they desire OJ, which is the perception of the moral quality of how they are treated by the organization and other members (Cropanzano et al., 2007). Ensuring adequate levels of OJ is a strategic goal for firms: an extensive amount of empirical research shows that employees' perceptions of OJ determine their positive or negative attitudes towards the organization and therefore their OC (Allen and Meyer, 1990; Imamoglu et al., 2019; Meyer et al., 2002; Meyer and Herscovitch, 2001; Strom et al., 2014).

Low levels of OJ may also increase the intention to quit (ITQ) in people who feel discriminated against (Colquitt et al., 2001, 2013). In the view of Tett and Meyer (1993), ITQ is defined as the desire of a person to leave a workplace or an organization or a decision to seek other alternative employment in other organizations (Krueger and Rouse, 1998). ITQ does not mean staff turnover automatically (Allen et al., 2005; Griffeth et al., 2000; Hom and Griffeth, 1995), however, it is strategic to consider the ITQ as an outcome to be avoided. Vardaman et al. (2012) show avoiding ITQ is important because it can affect OC negatively (see also Allen et al., 2005). Based on the above considerations, it is possible to make the following hypotheses.

H2.

PD decreases the OJ level

H3.

PD decreases the OC level

H4.

PD increases the ITQ level

Literature shows that the disparity between family and nonfamily employees, while is certainly present, does not necessarily indicate unfairness, as family and nonfamily employees have different sets of knowledge, skills, capabilities and sources of motivation (Block et al., 2015; Davis et al., 2010; Dawson, 2012). Hence, the available empirical evidence does not explain if and in which conditions AS and PD generate negative consequences. To this purpose, social psychology provides a further factor to be introduced into the model: the categorization of oneself as a member of a group (Tajfel and Turner, 1986). As suggested by Bettinelli et al. (2022), multiple identities' management and integration within family firms is an underdeveloped research area. In their research agenda, multiple identities is a useful approach for explaining complex phenomena, linked for example to the relationships between groups.

A large body of literature, almost exclusively psychosocial (with the exception of De Massis, 2012), described the mechanisms and conditions under which individual behavior is determined not by individual characteristics but by those of the group to which they believe to belong. This condition raises the risk for all family members of acting in ways favoring the in-group (the family itself) and at the same time discriminating against the out-group (nonfamily members) (Tajfel, 1978, 1981). In order to explain these intergroup dynamics, a promising stream of research was started some years ago by Deaux et al. (1995, 1999), Deaux (2011), which explores the different functions of relational processes and affective identification with the group by its members, processes which previously have been considered only cognitive. According to these authors in this context, it is not obvious that the outcome is the identification with the biological group membership (in this case, the owner family). If two colleagues, who perceive themselves as belonging to different groups (for instance family and nonfamily members), interact systematically due to the job they both perform, it is possible that mechanisms of conflict between groups are not activated: this is the so-called RS effect. Hence, in order to better explain the actual conditions in which PD could arise within a family small medium-sized enterprise (SME), in this study RS - is also considered as a potential independent variable.

Haar and Brougham (2022) showed the importance of workplace inclusion by targeting factors such as skills and goals that make the minority employees feel valued and part of the in-group. The prevalence of the performed role on the category to which it belongs could therefore reduce the perceived AS, by giving greater salience to professional issues rather than the in-group vs out-group discriminatory dynamics. Furthermore, due to daily professional cooperation/interaction, the salience of the role could decrease the impact of stereotypes and prejudices, in turn reducing the PD (Dovidio et al., 2017; Neubaum et al., 2020; Pettigrew and Tropp, 2008) and strengthening OJ. RS could also have a positive impact on OC: the psychosocial literature clearly shows how cohesiveness and a sense of belonging are above all a consequence of perceived similarity and shared goals (Hogg et al., 1995; Nasie and Stanescu, 2023); and OC in turn decreases the ITQ (Aggarwal et al., 2022; Imamoglu et al., 2019). Based on the above considerations, it is possible to make the following hypotheses (see Figure 1):

H5.

RS decreases the AS level

H6.

RS decreases the PD level

H7.

RS increases the OJ level

H8.

RS increases the OC level

H9.

RS decreases the Intention To Quit level

Figure 1 shows the theoretical model of the research.

3. The research

3.1 The sample and data collection

In order to select the participants, the research was carried out with the help of a young entrepreneurs association (“Giovani Imprenditori Confindustria”). The association was contacted personally by the researchers, and it made available its database containing the contacts of all associated companies freely. Starting from the association's database an exploratory mailing list was formed in order to gather the consensus to participate in the research. The invitation emails were sent to the institutional addresses of all the companies present in the database with a firm size up to 250 employees (N = 1,116). Since the research used questionnaires and measurement scales already widely validated in the Italian context, it was not necessary to undertake a pre-test study. Following this, a second more specific mailing list was created in order to find the firms with four specific characteristics:

1. Firm size, according to European Union (EU) parameters up to 250 employees, regardless of the gross turnover

2. A single family must share at least 50% of the ownership

3. The strategic decisions must be managed by the family

4. At least two generations belonging to the same family are involved directly in the firm

The second mailing generated a sample of 237 family firms presenting the desired requisites and available to participate to the survey. Finally, each company thus identified was contacted directly, to clarify the methods and timing of data collection. The sample definition process took approximately two months, from April to June 2022. For each participating company, only nonfamily members could fill in the questionnaire. In order that no single company was over-represented, a unique link was provided to each company that agreed to participate (in such a way as to be able to aggregate the responses of a single participating company into a single indicator). Each company, adopting a meta-analytic procedure, was treated statistically as a single case calculating for each variable the average derived from the analysis of all the questionnaires that came from the same company.

Data was collected with a questionnaire operating on Google Forms from July to September 2022. The nonrespondent bias was assessed by comparing the responses from early and last waves with a t-Test, such as the first and last quarter of responses. No significant difference emerged (p = 0.05).

3.2 Investigated variables and tools utilized

Adverse selection – AS: measured with two ad-hoc items (“Members of the owner family carry out tasks without being qualified”)

Perceived Discrimination – PD: measured with a 12-item questionnaire (Cronbach’s Alpha = 0.88) (“I am at a disadvantage compared to family members in work-life balance opportunities”)

Role salience – RS: measured with one ad-hoc item (“I perceive the members of the owner family as colleagues rather than counterparts”)

Organizational Justice – OJ: measured with the Italian version of the Colquitt Justice Scale, 2001 (Cronbach's Alpha = 0.86) (“Are the benefits/compensation you receive justified, considering your performance?”).

Intention to quit (ITQ): measured with three ad-hoc items (Cronbach’s Alpha = 0.70) (“For the past year, I have seriously considered resigning”)

Organizational Commitment – OC: measured with the Italian version of the OC scale of Allen and Meyer (1990) (Cronbach’s Alpha = 0.86) (“I don't feel ‘part of the family’ in this organization (R)”)

4. Findings

The survey was carried out on a sample of Italian family SMEs (participating companies N = 186, total questionnaires collected N = 838). A power analysis has been carried out to identify the sample size required in order to detect a given difference in a single mean with specified power (99) and significance (0.05): the sample dimension (n = 186) satisfies these methodological requirements.

Among the independent variables, AS shows average values (mean (M) = 3/6, standard deviation (SD) = 1.7): this suggests that there is a widespread perception that family members hold organizational positions for which they do not possess the appropriate characteristics/qualifications according to merit or expertise. The PD levels are quite low (M = 1.9/6, SD = 1.1): the out-group discrimination is not so high, but is undoubtedly present. The RS levels (M = 3.4/6, sd = 1.4) suggest that colleagues belonging to the owner family are seen equally as colleagues and as members of the out-group, confirming that social categorization is strong and widespread in the sample. Considering the dependent variables, the positive ones have acceptable or even good values (OJ M = 4.3/6, sd = 0.7; OC M = 4.4/6, sd = 1.7), ITQ is low but still present in the sample (M = 2.3/6, sd = 1.6). See Table 1 for descriptive statistics of the sample.

4.1 SEM analysis

In light of the evidence available in the literature, and given the complexity of the relationships between the hypothesized variables, the test of each hypothesis was carried out by calculating the regression coefficients by adopting the structural equation model (SEM) approach. The model shows a good fit (Akaike information criterion (AIC) = 2805.063; Bayesian information criterion (BIC) = 2876.029 with n = 186), confirmed by additional fit measures: Comparative fit index (CFI) = 0.948, goodness-of-fit index (GFI) = 0.999, root mean square error of approximation (RMSEA) = 0.198 and standardized root mean square residual (SRMR) = 0.047.

The factors considered in this study explain nearly 45% of the OC, approximately 49% of the OJ, 23% of the ITQ, nearly 28% of PD and finally 4% of AS, which is clearly due almost completely to exogenous factors. Given the large correlations and number of variables in the analysis, multicollinearity is possible. Therefore, a test for multicollinearity was carried out, and the variance inflation factors (VIFs) have been reported. Generally, a VIF above 4 indicates that multicollinearity might exist and further investigation is required. When VIF is higher than 10, there is significant multicollinearity that needs to be corrected. The result of the test shows multicollinearity among variables can be excluded (see Table 2).

Regarding the specific hypotheses tested by the study, AS has a strong and positive impact on PD; H1 is therefore supported. PD is confirmed as a factor with a strong and negative impact on OJ, OC and a positive effect on ITQ; findings thus support H2, H3 and H4. Findings also confirm that the introduction of the RS variable brings important insights into the model. The perception of members of the owner family as colleagues rather than as family members has a strong impact in mitigating both AS and PD, as well as ITQ. Furthermore, RS has a significant and positive impact on OJ and OC. Therefore, H5, H6, H7, H8 and H9 are supported. See Table 3 for a synopsis of the findings.

5. Discussion

Although the dynamics of in-group favoritism and out-group discrimination are well described in the literature (Samara and Arenas, 2017; Van der Heyden et al., 2005), the consequences for nonfamily members to date have been systematically under-investigated and they are not clear or definitive (Samara and Paul, 2019). This research brings new insights to the general debate on organizational fairness in family businesses and in particular on the role that AS, PD and RS play in influencing the levels of OJ, OC and ITQ among nonfamily members. These results support the value of a systemic approach to the topic of justice in family businesses, to overcome the limitations of previous approaches (Tabor et al., 2018), which often focused only on one factor at a time and described one-to-one relationships between variables. Findings stimulate some interesting streams of discussion examined below.

A large and detailed body of literature has demonstrated that family businesses often exercise unfair practices in their workplace favoring family members (e.g. Chua et al., 2009; Cruz et al., 2014; Kellermanns et al., 2012; Samara and Arenas, 2017; Zientara, 2017). However, the results of this treatment have not been clarified definitively. Findings show a moderate average level of PD, but at least 27% of this discrimination is due to AS. The results therefore confirm that the intra-family dynamics of selection and the careers of family member employees result in discrimination against nonfamily members, which should therefore become a strategic target for human resource management policies. These policies become even more important in light of the results that have emerged here: the impact of the here considered variables on OJ (approx. 48% of explained R-Squared), OC (approx. 45% of explained R-Squared) and ITQ (approx. 23% of explained R-Squared) could weaken trust in the organization and the reliability and continuity of the business.

Furthermore, the findings show that RS has positive effects directly on AS (−0.253): it therefore seems that the sharing of a professional identity potentially leads to a greater perception of the aspects related to the job and less of those related to the personal background which are instead the basis of the AS. By having a direct impact on the AS itself, RS can therefore mitigate the possible negative effects of the AS. In particular, findings suggest that RS can bring benefits by reducing the discrimination perceived by nonfamily members. According to an agency-based approach (Gedajlovic and Carney, 2010; Lubatkin et al., 2005), RS could reduce the probability of opportunistic behavior among nonfamily members. Instead, following a SEW approach (Berrone et al., 2012; Gomez-Mejía et al., 2007), the literature has repeatedly demonstrated that pursuing the enhancement of family identity can generate negative effects for family members (Eddleston and Kidwell, 2010; Ferrari, 2017, 2020, 2023b). This study suggests such negative effects may also affect the relationship with nonfamily member employees, due to the AS that comes from/as a result of pursuing and maintaining the family identity.

Findings also show that PD decreases if the job performed is more salient than family/nonfamily group membership. Identity therefore plays a fundamental role in explaining organizational dynamics, both at an individual and the group level: this evidence complements that hypothesized elsewhere (Bettinelli et al., 2022) and confirms the benefits of adopting an approach based on the model of multiple identities (Cunningham, 2020; Sundaramurthy and Kreiner, 2008; Whetten et al., 2014). Therefore, exploiting some recent theoretical and empirical developments (Deaux et al., 1995, 1999, Deaux, 2011), this study demonstrates that multiple identities do not only represent a ritualism to be managed (Kotlar et al., 2018, Kotlar and Sieger, 2019), but can become a source of benefits both for family and nonfamily members. In that specific case, the professional interaction, which gives greater salience to the performed job with respect to the social category to which it belongs, has a systematic and direct impact on mitigating the negative effects of PD. Indeed, RS seems to neutralize the effect of AS on ITQ: in a relationship that is not significant in itself, PD increases ITQ by 0.110, while RS decreases ITQ by −0.173. Literature suggests ITQ does not mean staff turnover automatically (Allen et al., 2005; Griffeth et al., 2000; Hom and Griffeth, 1995), but Vardaman et al. (2012) highlight that avoiding ITQ is also important because it can affect OC negatively (see also Allen et al., 2005). This study shows PD has a negative impact on sense of belonging (OC) and positive on ITQ, thus complementing that which is already known (Bayl-Smith and Griffin, 2014; Strom et al., 2014). Considering that in SMEs low staff turnover and high sense of belonging are usually very important goals, ensuring low levels of PD of nonfamily members becomes a strategic objective.

This study demonstrates that role identity has further important effects in moderating inter-group discrimination. Findings highlight that RS completely offsets the effects of AS on PD (0.245 vs −0.266); RS also a mitigates the negative impact of PD on OJ (−0.399), increasing the OJ level by 0.116. Finally, RS has a mitigating impact also the relationship between PD and OC: 0.300 vs −0.810. These findings support the suggestion that sense of belonging has a strong affective and interactional basis. This empirical evidence complements the traditional approach supporting the shared goal as the main trigger of sense of belonging (Hogg et al., 1995; Nasie and Stanescu, 2023). Although the impact of the RS is not sufficient to compensate for the (negative) influence that PD has on OC, the RS effect indeed always brings benefits in reducing negative effects of PD.

6. Conclusion

Research confirms that family businesses are characterized by dysfunctional dynamics (AS and PD) and highlights that such dynamics generate distortions and discrimination in the relationship between family and nonfamily members, with negative effects on OJ, OC and ITQ. Moreover, the literature has on other occasions demonstrated that pursuing SEW has side effects, (Ferrari, 2017, 2020, 2023b). This study brings further evidence to support this, demonstrating that, paradoxically, pursuing the objective of maintaining business continuity within the family generates AS, which in turn generates PD, to the ultimate detriment of reliability/continuity itself. Managerial practices could prevent the activation of this vicious circle, reducing the PD of nonfamily members. Such practices should prevent nepotism, i.e. a form of favoritism in business when family members are favored over nonrelatives/nonfamily members (Ferrari, 2020). Findings confirm the difficulty, especially in SMEs, of achieving a balance between informal relationships based on trust and daily collaboration and the formalization of human resource management processes (work organization, selection, compensation, career orientation). Informal relationships based on trust are an attractive factor for family businesses (Hauswald et al., 2016), but are prone to the 'bifurcation bias' (Verbeke and Kano, 2012). Conversely, Klein and Bell (2007) contend that formalization may discourage nonfamily members from joining the firm, as it is the informal and less bureaucratic structures that they often find attractive. This study also supports the opportunity for family SMEs to adopt a more formal and structured governance, in order to avoid nepotism and AS.

Furthermore, findings confirm the salience and interplay of various group-based identities can be understood in light of the performed job, (Bettinelli et al., 2022). Previous literature has highlighted that many aspects impact the levels of perceived OJ, such as; the size of the company (Barnett and Kellermanns, 2006), the level of managerialization (Carney, 2005), the bifurcation bias (Verbeke and Kano, 2012), the ways in which the agency relationship or stewardship is applied (Jennings et al., 2018). However to date no approach provides convincing reasons for the genesis of discrimination.

This study expands current literature by showing that RS plays an important role in determining levels of PD. Social identity theory thus provides the appropriate framework to deepen understanding of the dynamics involving the different group identities in a family business. Strengthening RS therefore becomes a strategic objective for family members who wish to reduce the unwanted effects of inter-group dynamics based on social identity.

Finally, Lubatkin et al. (2007) suggest that nonfamily members have a zone of indifference or tolerance toward the preferential treatment of family members. However, at a certain point, nonfamily members may become discouraged by family bias, activating perceptions of injustice (Tabor et al., 2018). This study suggests that the perception of injustice can be activated by the role that is most salient in the specific situation. The effect of RS in reducing negative consequences suggests that managerialization could well be a good antidote to nepotism. At the same time, this study highlights the side effects that pursuing SEW can have on the relationships between the owner family and nonfamily members, as highlighted elsewhere (Ferrari, 2020, 2023b). In particular, pursuing and defending the family identity, although it is a desirable noneconomic goal, makes the company itself vulnerable to discriminatory dynamics towards the out-group. Once again, family businesses must find a balance to manage the institutional overlapping between family and business.

7. Theoretical implications for family businesses literature

In their literature review, Bettinelli et al. (2022) affirmed it would be interesting to better understand the salience and interplay of various group-based identities. Following this suggestion, this study highlights that identity is a powerful framework to adopt in interpreting working conditions: identity is both a desirable noneconomic goal and a factor that makes family firms more vulnerable to discrimination. Furthermore, this study shows that shared identity mitigates PD. These findings complement that advocated by Bettinelli et al. (2022), who suggested a better investigation of/into whether different identity antecedents can lead to, or be influenced by, multiple and heterogeneous outcomes and contingencies.

This study advances current literature by focusing on the impact of multiple identities on fairness and commitment at individual and group levels of analysis of family businesses. Moreover, by adopting a self-categorization approach, this study also advances current theoretical literature as this methodological lens could help scholars to further understand diversity in family business. In doing so, this study responds to the recent call of Bannò et al. (2020) for deepening diversity management knowledge in family firms and advancing current knowledge regarding diversity research in family firms (e.g. Binacci et al., 2016; Chadwick and Dawson, 2018; Ling and Kellermanns, 2010).

Finally, this study brings new insights about nonfamily members, in addition to that already known in literature (Tabor et al., 2018; Rosecká and Machek, 2022). This paper particularly contributes to the available knowledge on OJ in family businesses, supporting literature that suggests preferential treatment of family members undermines the successful integration and assimilation of nonfamily members into a family firm. Daspit et al. (2017) show that cohesiveness in the business family may reduce bifurcated human resource practices: this study adds that by making the performed role salient it can also involve nonfamily members in this sense of cohesion, generating positive organizational outcomes.

8. Practical implications for practitioners and entrepreneurs/owners in family businesses

In addition to its theoretical contributions, this study also has practical implications. For example, it would be advisable to implement human resource management practices based on job rotation, to promote cohesion and reduce perceived distances. Self-categorization dynamics are based on a prototype, which becomes a stereotype and finally lead to prejudice toward the out-group (Tajfel, 1978, 1981). The goal for a family member therefore is to be perceived as a colleague, not as a usurper (Haar and Brougham, 2022). In line with this approach, Pettigrew and Tropp (2008) summarized that “contact reduces prejudice by (1) enhancing knowledge about the out-group, (2) reducing anxiety about intergroup contact, and (3) increasing empathy and perspective taking” (p. 922). In summary, at the heart of diversity management policies there is the aim to run the family business in a more professional way (Barnett and Kellermanns, 2006; Haar and Brougham, 2022).

9. Limitations and suggestions for future research

This study shows that among the considered variables, dynamics of mutual reinforcement and mitigation are also activated, which certainly have not been fully described in this study. The complexity of the scenario is therefore confirmed, and the opportunity to carefully consider any latent dynamics arises, for example by adopting a path analysis. Future research is hence called for to fill this gap.

In this study it was not possible to consider control variables such as, for example, the gender of the entrepreneur, which has an impact on the ethical dimension of the business, as suggested (e.g. Marín-Palacios, 2023). Future research will have to investigate the discriminatory dynamics also considering this important factor.

This study is limited to measuring the consequences of PD on psychological outcomes such as OJ and OC and shows that family businesses present critical issues. However, analysis of the consequences on the real performance of nonfamily members is now certainly required. Future research will have to clarify whether human resource management practices aimed to prevent PD are only appropriate for ethical and legal reasons or because, by reducing PD, they can actually improve the performance of nonfamily members.

Finally, although family firms are the most widespread type of business in the world (Samara and Arenas, 2017), diversity management in family business is still under-investigated (Bannò et al., 2020). This paper supports the evidence that in a family business, not belonging to the owner family is a kind/type of diversity (which can result in adversity), but nonfamily members have not until now been considered a target/subject for diversity management practices (Bannò et al., 2020). Future diversity research is called for/on to integrate nonfamily membership as at risk of discrimination and investigate more deeply the antecedents, behaviors and consequences of diversity management practices focused on nonfamily members.

Figures

Figure 1

Theoretical model

Table 1

Descriptive statistics

	Valid	Missing	Mean/6	Std. Deviation	Minimum	Maximum
Adverse selection - AS	183	3	3.049	1.717	1.000	6.000
Perceived discrimination - PD	186	0	1.952	1.173	1.000	5.000
Role salience - RS	186	0	3.419	1.401	1.000	6.000
Organizational justice - OJ	186	0	4.381	0.793	1.600	5.700
Intention to quit - ITQ	183	3	2.393	1.667	1.000	6.000
Organizational commitment - OC	183	3	4.475	1.757	1.000	7.000

Source(s): Created by author

Table 2

Explained R-squared and multicollinearity test

	R2	VIF
Organizational justice	0.485	1.941
Intention to quit	0.231	1.300
Organizational commitment	0.454	1.831
Perceived discrimination	0.277	1.383
Adverse selection	0.043	1.044

Source(s): Created by author

Table 3

SEM - Synopsis of the findings

Hypothesis	Predictor	Outcome	Estimate	Std. Error	z-value	p	Lower	Upper	Findings
H1	Adverse selection	Perceived discrimination	0.245	0.044	5.596	<0.001***	0.159	0.331	Supported
H2	Perceived discrimination	Organizational justice	−0.399	0.039	−10.312	<0.001***	−0.475	−0.323	Supported
H3	Perceived discrimination	Organizational commitment	−0.810	0.088	−3.684	<0.001***	−0.984	−0.637	Supported
H4	Perceived discrimination	Intention to quit	0.447	0.100	4.472	<0.001***	0.251	0.643	Supported
H5	Role salience	Adverse selection	−0.253	0.088	−2.854	0.004**	−0.426	−0.079	Supported
H6	Role salience	Perceived discrimination	−0.266	0.053	−4.985	<0.001***	−0.371	−0.162	Supported
H7	Role salience	Organizational justice	0.116	0.032	3.583	<0.001***	0.053	0.179	Supported
H8	Role salience	Organizational commitment	0.300	0.074	4.047	<0.001***	0.155	0.445	Supported
H9	Role salience	Intention to quit	−0.308	0.084	−3.684	<0.001***	−0.472	−0.144	Supported

Note(s): *p < 0.05, **p < 0.01, ***p < 0.001

Source(s): Created by author