Introduction

Family firms (FF) are the most common type of firms in the world (Tsoutsoura, 2021). However, in spite of the great boom that its research has experienced over the last decades, there are still many issues that remain unsolved (Chrisman et al., 2009; Dyer, 2018; O’Boyle et al., 2012). Whether FF are better or worse performers is one of the key issues that are not clear (Dyer, 2018). Meta-analyses of the performance of family versus non-family firms have not offered conclusive results either (O’Boyle et al., 2012) and some authors suggest that there is a dark and a bright side to family involvement (Chirico & Bau, 2014). What is clear is that FF represent a heterogeneous set of firms that do not allow a generalization of many assertions (Cruz et al., 2012).

The Resource-Based View (RBV) of the firm (Barney, 1991; Barney et al., 2001), which explains how internal business factors lead companies to obtain sustainable competitive advantages, highlights the role of intangible resources and capabilities (Kidwell et al., 2019) which might help to explain these conflicting discussions. Among such intangible resources, this paper emphasizes the role of Human Resources (HRs), including both family and non-family members. Specifically, we focus on family involvement together with family and non-family human capital (HC) and altruism and we analyze their effect on firm performance. According to the FF literature, HC represents one of the three key forms of “family capital”: social, human, and financial (Hoffman et al., 2006). Human resources have been recognized to be of the utmost importance in FF because family members are characterized by being rich in trust and are highly committed to the firm thanks to the intermingling of the family and business system and also because this extends to the whole organization, including non-family members (Gast et al., 2018; Herrero, 2018). Additionally, HC has not only been acknowledged as being one of the most relevant resources of FF to gain performance (Ahamad et al., 2022; Eddleston et al., 2008; Sirmon & Hitt, 2003), but also several recent works (Campos-García et al., 2021; Mallén et al., 2020) have empirically demonstrated its importance in other outputs such as internationalization (Almodóvar et al., 2016; Campos-García et al., 2021; Javalgi & Todd, 2011; Onkelinx et al., 2016b, 2016a) and firm performance (Hitt et al., 2006). This may be of even higher relevance in SMEs because their reduced size makes the interaction of the family and business systems more relevant (Wright, Roper, Hart, & Carter, 2015).

HC has been conceptualized as the education, experience, knowledge, and skills individuals bring to an organization, contributing to value creation (Bogdanowicz & Bailey, 2002). The literature has broadly emphasized the relevance of these resources stemming from HC in the organization (Stoyanov, 2016; Zula & Chermack, 2007). Altruism is a voluntary act that may be conceptualized as the willingness to help others without considering the effect that this action may have on oneself and without expecting a reward (Mallén et al., 2020). Previous research has found positive relationships between organizational citizenship behaviors, such as altruism and organizational success (Podsakoff et al., 2009).

While organizational HC and altruism have been proven to lead to the generation of competitive advantages (Dyer, 2006; Lopez-Cabrales et al., 2009; White, 2006), the literature is not clear about whether family involvement enhances the impact of HC and employees’ attitudes on SMEs’ performance when compared to non-family SMEs. Therefore, there have been calls in the literature (Dyer, 2006) to analyze these aspects comparing FF and non-family firms (Zhang et al., 2022). Due to their distinct features in terms of governance, ownership, structure, and management succession, FF are likely to possess different resources and behave differently from non-family firms (Acquaah et al., 2011). Researchers agree that family involvement in management is what makes FF unique (Chua et al., 1999). Therefore, we claim that it is this family involvement, rather than mere ownership, what makes differences in terms of strategy, HC, etc. (Revilla et al., 2016) and, consequently, by FF we will refer to a firm whose administration the owning family is involved in.

The characteristics of FFs can be seen as reinforcers or harmers of the effect of HC and altruism on performance. That is, the literature has found arguments in both directions. On the one hand, FFs can reinforce the positive effect of HC and altruism, because they are said to be good at transmitting knowledge, especially tacit knowledge, since family members often get involved in the firm at an early age and they have some resources that facilitate knowledge absorption (Salvato & Melin, 2008). Therefore, family involvement could favor the dissemination of the valuable and unique knowledge of HC among the SMEs, reinforcing its influence on performance. Furthermore, family members are highly experienced in firm-specific issues due to shared experiences since birth (Herrero & Hughes, 2019; Sirmon & Hitt, 2003), strengthening the influence of HC and altruism on performance.

On the other hand, FFs can be seen as buffers of the positive effect of HC and altruism on firm performance, because they may not be so good at attracting and retaining external talent (Arijs et al., 2018), and they may not be able to find a sufficient number of competent family members for occupying critical positions (Mehrotra et al., 2011; Muttakin et al., 2014). Furthermore, nepotism may assign unsuitable family members to high management positions, leading to inefficiencies in performance (Dyer, 2006) and these unskilled family members could harm the positive influence of HC and altruism on the company's success.

Given the different arguments that can be provided for and against the hypothesis of FFs as a contingent factor moderating the influence of organizational HC and altruism on SMEs’ performance, in this work, we try to answer the following research question: Does family involvement in management enhance or buffer the positive impact of HC and altruism on SMEs’ performance? By doing so, we fill the gap that exists with regard to the contingent effect that family involvement exerts on the relationship between HC, altruism, and firm performance. Furthermore, there is a void with regard to the inclusion of non-family members in the HC concept of FF as noted by some authors (Campos-García et al., 2021) who maintain that most of the FF literature focuses almost exclusively on the role of the HC of the founder/owner, managers or family members, ignoring the potential of the value of the HC associated with the non-family members of FF. In addition, we follow the advice of the recent FF literature that suggests that being an FF should not be a dichotomous concept, but ought to be measured over a continuum (Gómez-Mejía & Herrero, 2022; Herrero, 2018), such as the level of family members involved in management.

Theory and Hypotheses

Human Resources in Family Firms

Human resources must be available and, at the same time, relevant to organizational purposes if they are to represent a real source of competitive advantages for organizations (Ployhart et al., 2014). Using the RBV (Barney, 1991), Lepak and Snell (1999) focused on the strategic aspect of HC as the principal driver of employment modes and HR configurations by specifying two dimensions: value and uniqueness. They conceived the strategic value of organizational HC as its potential to improve the efficiency and effectiveness of the firm, exploiting market opportunities, and/or neutralizing potential threats, and defined the uniqueness of organizational HC as the degree to which such a value is rare, specialized and firm specific.

Some Human Resources Management studies focus on HC considering people’s skills, ability, knowledge, intelligence, creativity, judgment, know-how, and attitudes (Brueller et al., 2016; Carmeli & Schaubroeck, 2005; Kidwell et al., 2019; Nyberg & Wright, 2015; P. M. Wright et al., 2014). However, to better focus our research, we view HC as a means to convert personnel-level capacity and effort into organizational-level outcomes, and we follow the definition of Lepak and Snell (1999) and a large stream of authors who consider knowledge, skills, and abilities—but not attitudes—in the definition of HC (Brueller et al., 2016; Gutierrez-Gutierrez et al., 2018; Kloutsiniotis & Mihail, 2017; Sung & Choi, 2018). This view of HC is in line with the vast majority of the works in the FF literature (Campos-García et al., 2021; Cerrato & Piva, 2012; Dyer et al., 2014; Fernández-Olmos & Díez-Vial, 2015; Hitt et al., 2006; Javalgi & Todd, 2011; Onkelinx et al., 2016a; Sorenson & Bierman, 2009), that consider HC as the know-how and abilities of personnel, but not their attitudes or values (Dawson, 2012). In his review on family firm’s HC, Dawson (2012) notes that the majority of FF scholars focus on Knowledge, Skills, and Abilities (KSA) to explain HC (Wright, Dunford, & Snell, 2001), what he considers as head and hands. However, he also notes that head and hands are necessary conditions but not sufficient to align employees with firm goals nor to leverage firm resources leading to performance increases. Attitude (willingness) is the missing link to a productive behavior. This is what Dawson (2012) refers to as heart. Consequently, and given that KSA and attitudes do not have to go in the same direction, as an individual may have strong KSA (head and hands) but a poor attitude, or vice versa, this paper analyzes HC and attitude (treated as altruism), separately in their effect on performance.

The KSA included in HC are not exclusive of family employees as a separate unique resource, they also include non-family members’ influence (Huybrechts et al., 2011). In most family SMEs, non-family employees tend to be well treated and are frequently considered part of the extended family. According to Gómez-Mejía and Herrero (2022), they are often called as “compadre” or “comadre.” This situation is much less likely to happen in a large firm where the family and the business system have clearer boundaries and where relations are more impersonal (Cruz et al., 2019). In a similar vein, positive behaviors such as altruism could be shown by family and non-family members of the SMEs (Gast et al., 2018). Consequently, in this work, by HC/altruism, we refer to the HC/altruism of managers and employees who may directly impact the FF’s performance, regardless of if they refer to employees of family or non-family firms and within the former, and irrespective of if they are family members or not.

The Contingent Influence of Family Involvement in Management on the HC and Firm Performance Relationship

Barney (1991) claimed that sustained competitive advantage derives from the resources and capabilities (tangible and intangible assets) controlled by a firm that are valuable, rare, imperfectly imitable, and not substitutable. Family business scholars have frequently cited this resource-based approach (RBV) (Chrisman, Chua, & Sharma, 2003; Chua, Chrisman, & Steier, 2003; Habbershon, Williams, & MacMillan, 2003; López-Fernández, Serrano-Bedia, & Pérez-Pérez, 2016; Tokarczyk, Hansen, Green, & Down, 2007) to analyze the unique resources of FFs and to explain how FF may differ from non-family ones (Chrisman et al., 2005).

The set of resources specific to a particular company that results from family involvement is known as “familiness” (Habbershon & Williams, 1999; Mani & Lakhal, 2015). Frank et al. (2010) also refer to “familiness” as a bundle of resources and capabilities shown by family businesses that make them different from non-family firms (Carnes & Ireland, 2013; Cucculelli, 2018; Danes et al., 2009). This has proven to represent a peculiarity of family businesses with respect to other types of companies (Carnes & Ireland, 2013; Cucculelli, 2018; T. Habbershon & Williams, 1999). Familiness is difficult to measure and has often been done so by the level of family involvement in the firm. While family involvement is a necessary but not sufficient condition for familiness, it has often been used as an indicator of familiness because it implies, to a greater or lesser extent, a bundle of intangible resources such as trust, reciprocity, or commitment (Chrisman et al., 2005; Zellweger et al., 2010). For example, comparing FF attributes and resources’ uniqueness, Sirmon and Hitt (2003) found that family involvement usually leads to extraordinary warmth and commitment among family members, favoring friendly and close relationships.

Family firms possess a long-term orientation and commonly offer a more stable work environment using long-term contracts (Campos-García et al., 2021), so when key employees’ skills and abilities are low, family members will try to compensate it by using other resources based on reciprocity feelings and family commitment. For example, by increasing social relations with agents inside and outside the firm to benefit from better-aligned interests that can lead to the creation of competitive advantages (Dawson, 2012). Some other unique advantages can compensate for low-skilled managers in family firms because, across generations, they typically cultivate long-standing relationships with stakeholders, so the latter are likely to develop deeper personal attachments to the family that owns and operates the business (Cennamo et al., 2012). FF also have certain advantages for attracting customers because of the goodwill and trustworthiness generated by the family name and its commitment, which eventually adds value to the “family brand name” (Dyer, 2006).

Consequently, even if nepotism could mean that incompetent family members attain key positions (Zellweger et al., 2013), because of a tacit norm of reciprocity, these members will be highly committed to the firm (Pearson & Marler, 2010). Therefore, they are expected to make an extra effort in these aspects to compensate for their low talent because family members share common goals and behave altruistically for the good of the family and the business. Identifying reciprocity and commitment as a family-specific resource demonstrates the positive impact that family involvement in management can have on a business (Eddleston et al., 2008; Sirmon & Hitt, 2003), particularly when family managers lack other characteristics such as the necessary skills and abilities needed for the position. Furthermore, the family environment can facilitate specific training by strengthening individual skills, enhancing the value of HC (Dyer et al., 2014). Family members keep comparatively closer relationships with employees, as compared with non-family managers, which help family firm managers to identify the weak points of their HC and, consequently, facilitate the way to find a solution (Almodóvar et al., 2016). Besides, family influence can also be exerted by family members who do not work for the firm by injecting family values into the firm’s goals, decision systems, and ways of interacting at the family-firm interface (Danes et al., 2009).

However, family involvement does not always lead to an automatic improvement of HC’s influence on performance (Zellweger et al., 2010). There is also a less positive side associated with family involvement, which can buffer HC’s potential and place the family firm at a disadvantage. Family involvement can lead to valuable assets and impose costs and liabilities for firm performance (Dyer, 2006). Some aspects of family involvement in management can diminish HC’s effect on performance, mainly when the firm’s HC is high, as we argue in what follows.

High HC is associated with high firm economic performance, but does family involvement moderate such a relationship? Even if economic gain is a primary human goal, other purposes bear a comparable weight which arises from the organizational identifications and loyalties (Simon, 1993). A firm’s economic performance will benefit from high talented HC, but in family firms, the economic goal becomes less important beyond a certain threshold. In contrast, once a certain economic level is achieved, other family goals acquire more relevance (Chrisman et al., 2012), such as preserving family independence, conservatism, or avoidance of high-risk activities. We expect that if HC is high in firms with no family involvement, they will try to get the most out of their HC, at least in terms of benefits for the firm’s economic performance, given that they have a higher focus on financial goals with no upper threshold, whereas firms with high family involvement would focus on other family goals, once a certain economic level is achieved.

Additionally, when HC is high, we expect that talented managers of non-family firms will get involved in innovation and entrepreneurial activities (Naldi et al., 2007), which lead to enhanced competitive advantages. Conversely, high family involvement in management would likely interfere negatively in these activities, as family firms are risk-averse and, consequently, tend to innovate less because of uncertainty about the potential success of the new business (Carney et al., 2015). Furthermore, as innovating or developing entrepreneurial activities often require the skills of outsiders and as family members tend to avoid depending on outsiders (Gomez-Mejia et al., 2010; Herrero, 2017), firms with high family involvement in management will be reluctant to innovate or to develop entrepreneurial activities. Consequently, they will probably not take advantage of the full potential of talented HC. This does not represent a problem when HC is low, as low talented HC will not likely innovate nor develop entrepreneurial activities.

All previous arguments lead us to think that the relationship between HC and firm performance in contingent on the level of family involvement.

All the above leads us to hypothesize that:

  • H1: Family involvement in management moderates the positive relationship between HC and firm performance.

The Contingent Influence of Family Involvement in Management on the Altruism and Firm Performance Relationship

Altruism is considered helping conduct that involves a concern for others (Guinot et al., 2016). When the workforce is motivated by prosocial behaviors, it goes beyond its responsibilities and improves its productivity (Grant et al., 2008). By perceiving that their work is valuable for others, employees are more engaged, dedicated, and develop extra-role behaviors (Mallén et al., 2020). Previous research has found positive relationships between organizational citizenship behaviors, such as altruism, and firm’s performance and success (Podsakoff et al., 2009). The mechanisms by which these behaviors affect performance are increased managerial productivity and coworkers’ freeing up resources for other important purposes. Besides, they help to coordinate activities, strengthening the organization’s ability to attract and retain the best employees, increasing the stability of the organization’s performance, and empowering the organization to more effectively adapt to changes in the environment.

To help others, the altruistic firm member needs to know who is in need. Family members frequently share a common language that favors others’ understanding and needs (Chirico & Salvato, 2008), facilitating altruistic members’ roles and altruism increases communication and cooperation, developing a supportive collective culture and norms of reciprocity and commitment that extend across an entire organization (Eddleston & Kellermanns, 2006; Madison & Kellermanns, 2013). Family involvement, through family identity and strong family ties, favors communication, not only among family members but also among non-family members (Herrero, 2018; Pearson & Marler, 2010), facilitating the collaboration between altruistic members and those firm members in need of help (Zellweger et al., 2013), ultimately enhancing the effect of altruism on firm performance. These resources can enhance not only the benefits of altruism in FFs but can also buffer its negative effect when altruism is absent. For example, a family member may not be altruistic; that is, he/she may not behave without expecting something in return, but even though that family member may behave positively with others for other reasons, such as FF identification or for his/her good, as the fate of the family firm is associated with his/her own life, because there is a skinny layer between the family and the firm (Gast et al., 2018). Stewardship Theory maintains that individuals feel satisfaction through pro-organizational and collectivistic behavior and they aim at aligning their own interest and those of the organization (Campos-García et al., 2021). This applies particularly to the case of family members because aspects such as firm identification or strong ties help to align their own interests with those of the family firm and, consequently, compensate the lack of altruism of other non-family members by their own stewardship behavior because their interests are typically aligned—by nature—with those of the family firm (Campos-García et al., 2021). Additionally, family involvement leads to the firm’s emotional attachments and strong feelings that can enhance family members’ altruism. Consequently, even if altruism is not particularly relevant in a specific family firm, many other resources can enhance performance because family members often possess ethical norms that invite social support and stewardship behaviors not only from non-family members but also from family members (Sorenson, Goodpaster, Hedberg, & Yu, 2009), which ultimately extract more value from the extant altruism favoring firm performance. Furthermore, firms with a higher level of family involvement in management commonly leads to greater shared vision and common values within the management team (Cerrato & Piva, 2012) and, consequently, the altruism behavior of these managers will multiply the effect of other firm members’ altruism based on a reciprocal commitment and interest alignment.

However, altruism also has a dark side in cases of high family involvement. The literature has demonstrated that the preservation of SEW is more relevant the higher the level of family involvement (Gomez-Mejia et al., 2010; Herrero, 2017, Campos-García et al., 2021) and the need to preserve affect-based resources, such as SEW, may lead family members to favor family well-being over firm aspects (Campos-García et al., 2021), giving both parents and children incentives to take actions that can threaten the welfare of the firm (Schulze et al., 2003). In addition, even if family involvement in management allows a greater appreciation of the benefits and risks in decision-making, at high levels of family involvement, the weight assigned to SEW preservation results in a higher risk aversion which may hinder firm growth (Cerrato & Piva, 2012).

Consequently, we theorized that even if family involvement can enhance the positive effect of altruism, when altruism is very high, its positive benefits can be hampered by the biased attitude toward the family due to a strong family involvement. All the above leads us to hypothesize that:

  • H2: Family involvement in management moderates the positive relationship between altruism and firm performance.

Methods

Sample and Data

We tested our hypotheses in a sample of Spanish wineries. This industry was chosen because of its relevance in economic, social, and environmental terms (OeMv, 2014). This influence is also clear in other areas of the world, since the global wine world trade has increased to over 100 million hectoliters. In volume, Spain is the world’s largest exporter of wine, with a growth of almost 50% in the last decade to nearly 23 million hectoliters. Spain also has the largest area of land under vines in the world. Moreover, the Spanish wine industry employs around 25,000 people, generating sales of more than 7 billion euros (OeMv 2018a). On January 1, 2018, there were 4373 wineries in Spain, of which only 290 had more than 20 employees (OeMv, 2018b). The market is thus largely made up of small firms competing in a hostile environment. Therefore, their HRs seem to play a decisive role.

We used the following research process to obtain the data. First, we utilized the SABI/AMADEUS database to identify all Spanish wineries with annual sales above 100,000 Euros and, at least, 5 employees. This research philosophy and design gave us an initial sample of 948 wineries. The reason why we included these two restrictions was that we wanted to ensure that the winery had a minimal structure to guarantee the role of HRs. Second, in 2014, these companies were contacted by telephone to ensure they fell within the sample framework. This resulted in 520 active wineries manufacturing wine. Out of them, we obtained 149 useable SMEs responses, a rate of approximately 29%. We designed the questionnaires focusing on the aspects we wanted to analyze, based on well-known scales. Once we had the questionnaire, we submitted them to two managers of wineries and to two academic experts for feedback and to ensure that all the questions were clearly understood. After a few minor comments were revised, we addressed the questionnaires to winery managers. To safeguard against bias and verify the quality of the responses obtained, we also surveyed secondary respondents (oenologists). Twenty-three oenologists of the 149 firms responded, which enabled us to establish the inter-rater reliability agreement for our final population sample. Objective information about performance was obtained from the SABI/AMADEUS database, reducing the risk of common method bias. In order to check for non-response bias, we compared mean differences between respondents and non-respondents for industry membership, number of employees, and revenue. No significant differences were found, suggesting that non-response bias was not present. When we considered the performance measure, there were some missing data, resulting in a reduction of the sample size to 105 observations.

Our sample is composed of family and non-family SMEs. Firms are on average 24 years old (standard deviation of 16.66) and have an average number of employees of 20.8 (standard deviation of 79.4). The percentage of wineries with family members involved in management was around 60% (standard deviation of 0.40). They are on average in their second generation, ranging from zero (those that are not FFs) to the fifth generation (see Table 1).

Table 1 Descriptive statistics of our sample
Full size table

We conducted the following process to ensure data validity and reliability. First, all the measures were pretested in interviews with 25 managers and two academics who are experts in the field. We asked them to review the survey, to ensure the clarity of the questions, and to ascertain whether the scales captured the desired information. We then revised any potentially confusing items before submitting the questionnaire. Given that the output variable we focused on was performance at the firm level and that the contribution of each person to this output was not possible to be distinguished from the input of other individuals, all the variables were measured at an organizational level.

Measures

Independent Variables

Human capital: we adapted the scales proposed by Lepak and Snell (2002) and Lopez-Cabrales et al. (2009) to measure HC. We considered two dimensions: value and uniqueness of organizational human resources. This measure has been widely used in the literature and has been previously validated (Brueller et al., 2016; Gutierrez-Gutierrez et al., 2018; Kloutsiniotis & Mihail, 2017; Sung & Choi, 2018). Consistent with the definition of HC, we measured HC with respect to the firm members who could exert influence in the firm, whether or not they belonged to the owner family. By doing so, in addition to specifying the HC that the firm possesses, it becomes unnecessary to distinguish between family or non-family firms to measure their HC (see Table 2), note that we have preferred the use of this formal measure of HC, which includes knowledge, skills, and abilities, than those proxies that measure HC through just one indicator, such as the proportion of employees with a certain level of university education (Fernández-Olmos & Díez-Vial, 2015), the ratio of R&D staff (Almodóvar et al., 2016; Campos-García et al., 2021), average salary (Gourlay & Seaton, 2004), or the value added per worker (Onkelinx et al., 2016b) because these latter measures are focused just on specific aspects, hence not being able to include a wider measure of KSA.

Table 2 Factor loadings for human capital value and uniqueness items
Full size table

Altruism: we used the scale proposed by Guinot et al. (2016) to measure Altruism (see Table 2).

Family involvement in management: we measure this variable as the percentage of family members in the management team, allowing a continuum measure of family firms as has been recently requested by many FF researchers (Gómez-Mejía & Herrero, 2022; Herrero, 2018). We preferred this to a dichotomous classification of family and non-family firms in order to classify firms according to a continuous measure ranging from “very non-family firm” (no family members involved in management) to “very family firm” (family members representing 100% of the management team). Moreover, as we have explained in the introduction, we believe that family involvement in management, and not mere family ownership, has strategic implications. This is a common measure of family involvement in management (e.g., Revilla et al. 2016).

Dependent Variables

Performance: although noneconomic goals are present in firms in general and also in family firms, in line with empirical research on family business performance, which has primarily dealt with economic goals (Chrisman et al., 2005), we measured performance through the 2015 Return on Sales (ROS), calculated as Earnings Before Interest and Taxes over total sales. We got this information from the SABI database, taking 1 year after the survey data was collected to guarantee the previous influence of HC, altruism, and FF on the firm’s performance.

We controlled for firm size using the firm’s total assets. We also controlled for the age of the company; the number of generations that the family had been involved in the FF, and the effect of having a family CEO, by including a dummy variable equal to 1 if the CEO was a member of the family and equal to 0 otherwise.

Results

Key Results

We carried out a CFA of the key latent factors, HC, and altruism of firm employees. We eliminated items with factor loadings below 0.6. Table 2 presents the final items included in the latent constructs. In the final model, all standardized coefficients were higher than 0.64 and highly significant (Table 2), on average 0.779 for the value of HC, 0.733 for that of uniqueness, and 0.776 for that of altruism. Even though the RMSA was slightly over the limit, other indicators suggest an acceptable fit. For instance, the normed chi-square was 2.18, which indicates a good fit. We also calculated the average variance extracted (AVE) from each of the three latent factors. The AVE was equal to 0.610 for value, 0.546 for uniqueness, and 0.610 for altruism. All of them were higher than 0.5, indicating that the proportion of variance explained by the factors is higher than the variance due to measurement error (Carr et al., 2011). The composite reliability (CR) for the two dimensions of HC was 0.945 (for value) and 0.877 (for uniqueness), while it was 0.886 for altruism. The three calculated CRs were far above the required threshold of 0.7. The correlation between both dimensions was 0.710 (p=0.000), which was relatively high and significantly different from zero, suggesting that both are different but are also dimensions of the same construct. Additionally, the confidence interval did not include the value of 1. Our results are in line with the literature regarding the validation of this scale, indicating that the HC scale has both convergent and discriminant validity. The correlation between HC and altruism was 0.32 (p-value=0.003), with a confidence interval that did not include zero.

Given that there were a few latent variables, we have preferred the use of Structural Equations Models rather than regression analyses as in these cases SEM is more reliable. We ran four structural equation models to test our two hypotheses (Figs. 1 and 2 and Table 3). We considered HC and altruism models separately because the correlation between them was high and it could have produced multicollinearity problems, particularly when we introduced the interaction effect.

Fig. 1
figure 1

Structural equation model. Includes the moderating effect of family involvement on the relationship between HC and firm performance (for simplicity, we have omitted in this figure the items corresponding to the latent variables)

Full size image
Fig. 2
figure 2

Structural equation model. Includes the moderating effect of family involvement on the relationship between altruism and firm performance (for simplicity, we have omitted in this figure the items corresponding to the latent variables)

Full size image
Table 3 Results of the structural equation models
Full size table

The coefficients associated with family involvement in management in the base models are nonsignificant, suggesting that family involvement does lead to higher levels of firm performance (Table 3 and Figs. 1 and 2). The results also showed that both HC and altruism positively affect firm performance (see models 1 and 3 in Table 3). Interestingly, we also found that older wineries seem to perform better than younger ones.

Our structural equation models 2 and 4 (Figs. 1 and 2) analyze the moderating effect of family involvement in management on the relationship between the two key variables, HC and altruism, and firm performance. Both models fit the data well. The root mean square error of approximation (RMSA) was in both cases slightly over the limit but still acceptable, particularly given that other fit indexes are well within the acceptable ranges (CFI of 0.816 and 0.888 and TLI of 0.795 and 0.851, for the HC and altruism models, respectively).

The moderating effect of family involvement in management on the relationship of both HC and altruism was found to be statistically significant in the two models considered (Table 3 and Figs. 1 and 2), supporting our hypotheses, although in the case of HC, it had to be accepted the 10% level. To analyze this effect in greater depth, we drew up graphs on both interactions for high and low levels of the moderator (Figs. 3 and 4), where high (low) levels of the moderator correspond to the mean plus (minus) one standard deviation. The results presented in Table 3 and Figs. 1 and 3 show that for low values of HC, firms with high family involvement in management outperform those with low family involvement. The same occurs for low values of altruism (Table 3 and Figs. 2 and 4), firms with high family involvement in management present higher performance than firms with low family involvement. As expected, when the HC and altruism among employees rise, the performance also increases, and the difference between the performance of FF and non-family firms decreases (Figs. 3 and 4). When HC is high, firms with a lower family involvement in management outperform those with a higher family involvement, as predicted in hypothesis 1. On the contrary, firms with a low family involvement present a similar performance to those with a high family involvement when altruism is high, suggesting that non-family firms resemble family firms when their members are highly altruistic.

Fig. 3
figure 3

Moderation effect of family involvement on the relationship between HC and firm performance

Full size image
Fig. 4
figure 4

Moderation effect of family involvement on the relationship between altruism and firm performance

Full size image

We tested multicollinearity in all models by analyzing the Condition Number for the Information Matrix. In all cases, this number was far from the threshold level.

Robustness Tests

As a robustness testFootnote 1, we ran a new model where we used the percentage of family ownership as a moderator of the relationship between HC and altruism and firm performance. Interestingly, we could not completely support our previous result. While we did this as a robustness test, the non-significance of the results supports the idea that it is not merely family ownership what makes a firm have a unique behavior but the actual involvement of the family in the management of the firm. This finding is consistent with previous research that noted conflicting results of family ownership in variables such as productivity and/or performance (Barbera & Moores, 2013).

In addition, we considered the Return on Equity (ROE) an alternative performance measure which is also a standard performance measure (Cruz et al., 2012; Mason-Jones, 2017; Maury, 2006; O’Boyle et al., 2012). The results were statistically more significant. In the case of the moderation model associated with HC (hypothesis 1) that had to be accepted at the 10% level in the previous model (in which the dependent variable was ROS) was now supported at the 5% level, which reinforces our previous conclusions. The moderation effect of family involvement on the relationship between altruism and firm performance was found to be highly significant and its effect was very similar to previous results, all of which provides our work with robustness.

Finally, and given that employees represent a key resource for wineries, of equal or—in some cases—even more relevance than assets, we also considered the number of employees as a measure of firm size. The results of this analysis also led to similar results.

Discussion

In this study, we have analyzed one of the three key forms of “family capital,” human capital, which has been recognized to be one of the most relevant assets of FF. While most of the FF literature only considers the effect of family members or the top management team in the concept of HC, we have gone further by including all the firm personnel. This way, we have answered to the call of some authors (Campos-García et al., 2021) that urge researchers to also consider the HC associated to non-family members as they can represent a key asset for the firm. In addition, we distinguish the talent (KSAs) of HR, which is captured in the term HC, from the attitudes of HR, which we measure in terms of altruism. By doing so, we can distinguish the effect of the one and the other on the firm performance. This is key, as they may have different effect because head and hand (HC) do not always coincide with heart (attitudes), as expressed in the terms of Dawson (2012).

Traditionally, the literature has assumed that family involvement in management benefits FF (Dyer et al., 2014). However, the arguments provided are conflicting, and many studies prove that family involvement in management is not free of disadvantages (e.g., Dyer 2006; Muttakin et al. 2014). Therefore, determining if HC and altruism could be reinforced or harmed by family involvement in management seems a relevant issue for the literature.

So, in our research, we have studied these relationships and analyzed FF’s impact on the relationships between HC and altruism and SMEs’ performance. The results obtained from a sample of Spanish small- and medium-sized wineries showed, consistent with previous literature, that both HC and altruism have a positive impact on SMEs’ performance (Anand & Dalmasso, 2020; Podsakoff et al., 2009; Zula & Chermack, 2007). What is more, it was found that a larger FF enhances the relationship between HC and altruism on performance when HC and altruism are low. However, for high values of HC, FF harms its relationship with performance, and for high values of altruism, non-family firms can perform similarly to FF.

Theoretical Implications and Contributions

Our findings make the following contributions to the literature. First, it has implications for the FF literature, showing that FF possess unique characteristics, some of which can be considered valuable assets, whereas others may be a liability. This is what Habbershon and Williams (1999) describe as capabilities leading to “distinctive familiness” and symptoms indicating “constrictive familiness.” Family involvement in management may be an asset that enhances performance when human resources (HC and altruism) are low by compensating for other mechanisms, such as reciprocity. In contrast, in firms rich in human resources, it may be a liability that curbs the effect of high HC through factors such as risk aversion or leniency (Herrero, 2017).

Second, our study contributes to the Human Resources Management literature. It has implications for this body of research since, following one stream of the previous literature’s proposals, it has independently analyzed HC and attitudes as altruism to determine their own role in SMEs’ performance (Brueller et al., 2016; Carmeli & Schaubroeck, 2005; Kidwell et al., 2019; Nyberg & Wright, 2015; P. M. Wright et al., 2014). That is, our results have found that the implications of HC, such as employees’ knowledge and skills (Brueller et al., 2016; Gutierrez-Gutierrez et al., 2018; Kloutsiniotis & Mihail, 2017; Lepak & Snell, 1999; Sung & Choi, 2018), on performance are different from the implications of attitudes. This is important since companies can face the situation of having highly skilled workers with not so good attitudes and the opposite. Understanding the independent role of these two dimensions is important to better comprehend the human resources impact on companies.

Third, our paper contributes to the RBV (Barney, 1991; Barney et al., 2001) by showing how resources and capabilities, intangible assets in our case, really help to obtain competitive advantages. Fourth, we integrate the HC and FF literatures. By framing FF in terms of family involvement in management, we can draw sounder conclusions and clarify some contradictory or confusing findings in previous research. For example, most of the literature suggests that family HC leads to a higher FF performance. However, Danes et al. (2009) found that family involvement in management positively affected gross revenue but neither family years of experience in the business nor the gender of the business owner (both considered proxies for HC) led to improved performance. Finally, our contingent analysis of the influence of family involvement in management, HC, and altruism on firm performance helps to explain why, in some cases, FF have not taken advantage of very highly qualified HC and/or altruistic members and why other FF with less qualified HC or fewer altruistic members are able to compete successfully.

Finally, we contribute to the family firm literature by using a formal, validated scale of HC which is not the common practice among FF researchers, most likely due to these two disciplines having evolved separately. In the FF literature, HC has been often estimated using straightforward measures such as the number of graduate employees, years of business experience, the gender of the business owner, or the number of family employees (Cerrato & Piva, 2012; Danes et al., 2009).

Practical Implications

The proposed framework identifies some managerial factors that are expected to affect performance in SMEs. Firstly, it is important to manage human resources, both aptitudes (HC) and attitudes (altruism), to reinforce SMEs’ results. A company investing in HC and altruism within the firm will obtain a better performance than those not taking care of such important resources. Secondly, family involvement in management could be used to counter the lower performance obtained by those SMEs that do not have high HC or altruism. Thirdly, family involvement in management should be avoided in SMEs with high HC. However, this involvement does not harm SMEs with high altruism. Fourthly, it is important to adopt a holistic view in management to understand the big picture of SMEs’ direction and coordination.

Limitations and Future Lines of Research

Like other studies, ours has certain limitations providing opportunities for future research. First, our study focuses on a number of variables to explain the contingent role of FF in the HC, altruism, and performance relationship. While we believe that this was a good approach to reach our aims, other research could find it interesting to include different constructs or relationships. Our approach has permitted the understanding that attitudes and knowledge are different constructs with different implications for the company. This opens new research lines focusing on such variables independently and helps to understand the contradictions of previous research.

Second, we relied on cross-sectional data obtained in 2014, not analyzing family involvement changes over time. However, this is not a relevant problem since family involvement is a structural variable that is likely to be relatively stable over moderate periods (Revilla et al. 2016). Moreover, having included a performance measure from 2015 (1 year after obtaining the data for HC and altruism), our approach permits analyzing the effect of an independent variable on the following year’s performance. This opens avenues to design longitudinal studies in which the role of these variables could be analyzed in the long term. Third, our study is based on just one sector, the wine industry. Future research confirming our hypotheses in other settings would be desirable. Finally, since the Spanish wine industry has a limited number of wineries with HRs, the final sample was also limited. Future research on the global wine industry could overcome this problem.

Conclusions

Our findings could be summarized as follows. We have theorized that the value which can be extracted from SMEs’ human resources is contingent on the level of family involvement in management. Furthermore, our results show that when the level of HC or altruism is low, firms with a high family involvement are better performers than those with a low family involvement. We assume that this might be due to the fact that FF possess other mechanisms, such as family identity, reciprocity, or family social capital (Herrero, 2018; Pearson & Marler, 2010) which allow family SMEs to compensate for the lower level of human resources abilities and attitudes while enhancing their potential. Our theoretical argument, also maintain that non-family firms are better at drawing benefits from their situation than FF when HC is high, which is in line with our results. This finding could be due to the dark side of the same family characteristics that enhanced HC’s effect when it is low. The desire to preserve family identity or the conservatism that may plague family firms diminishes their potential when HC is high and the family firms’ basic needs are met by restraining entrepreneurial or high-risk activities. We have also proposed and obtained support for the idea that family involvement in management can enhance altruism’s positive effect when this is low. Interestingly, for high levels of altruism, when we expected family involvement to harm the positive relationship between altruism and performance, we have found that non-family firms achieve similar results to those with a high family involvement. This finding could be because high altruism cannot harm performance when the family involvement is high since it is inherent to the individual who feels the need to develop an organizational citizenship behavior (Mallén et al., 2020).