Corporate Social Responsibility in Canadian Family Businesses: A Socioemotional Wealth Perspective

Abstract

After having gained prominence in the late 20th century, corporate social responsibility (CSR) has emerged as a critical business aspect, adopted widely across the corporate landscape. Although family firms play a significant global role, research on their relationship with CSR performance remains sparse and inconclusive. This paper seeks to bridge this gap by employing the primary classification of family firms, the socioemotional wealth perspective, and its FIBER model to examine their influence on CSR performance. The focus is on Canadian public companies listed on the S&P/TSX Composite Index from 2014 to 2022. Utilizing the NBC Canadian Family Index, the findings suggest that family firms exhibit superior CSR performance compared to their non-family counterparts. Further analyses indicate that family firms with greater control and influence by family members, those named after the family, those with strong emotional ties, and first-generation family firms tend to have enhanced CSR performance. By developing a socioemotional wealth score through FIBER dimensions to classify family firms, this study underscores the association of family firms with higher CSR performance, validating the robustness of the results.

1. Introduction

Canada, among many countries around the globe (e.g., Italy), is characterized by the prevalence of not only ownership concentration but also family ownership (Bozec and Di Vito 2019). Family firms are a significant contributor to the Canadian economy, representing a large portion of businesses across the country. According to the most recent report by The Conference Board of Canada, family-owned businesses account for approximately 47% of all private sector employment in Canada and contribute over 63% of the Canadian economy annually (Conference Board of Canada 2019, p. 6). Furthermore, Canadian family-owned businesses have shown strong performance in the stock market, with a cumulative total return of 325.1%, outperforming the S&P TSX Composite, which only generated a total return of 221.9%. Besides, they demonstrated resilience during economic uncertainty, such as the 2008 financial crisis and the 2020–2021 COVID-19 market disruptions (NBC 2022). Additionally, in Canada, excluding Quebec, which operates under civil law, the legal system is based on the common law (Viviers and Steenkamp 2023). Research suggests that common law countries tend to be more stakeholder-oriented, which translates into low corporate social responsibility (CSR) engagement. In contrast, civil law countries tend to be more shareholder-oriented, translating into high CSR engagement (Liang and Renneboog 2017). This mixed legal framework has implications for CSR engagement and further adds to the Canadian context.

Despite this, research on the Canadian context’s relationship between family firms and CSR is scarce (Zeng 2021). To fill this gap, we focus on differences between family and non-family firms when it comes to their performance with respect to CSR in this study. We build on prior research on the relationship between CSR performance and family firms using agency and socioemotional wealth theories to explain the corporate social performance of family firms vis-à-vis non-family firms. In fact, contrary to the agency theory, which highlights the opportunistic behavior of the family, the socioemotional wealth theory suggests that family members are inclined to preserve their socioemotional wealth apart from economic considerations. The socioemotional wealth concept was introduced by Gómez-Mejía et al. (2007) and refers to the “non-financial aspects of the firm that meet the family’s affective needs, such as identity, the ability to exercise family influence, and the perpetuation of the family dynasty” (Gómez-Mejía et al. 2007). Afterward, while recognizing the heterogeneity of family firms, Berrone et al. (2012) explored the multidimensionality of socioemotional wealth in family firms. They identified five dimensions of socioemotional wealth, known as the FIBER model, which are particularly relevant to family businesses. Thus, to expand this research, we describe differences within family firms based on the socioemotional wealth perspective regarding their CSR performance. In this vein, we consider the socioemotional wealth perspective regarding family control and involvement, family ties, family identification, emotional attachment, and dynastic renewal, as developed by Berrone et al. (2012).

We create a dataset of 2097 firm-year observations, of which 396 observations relate to family firms and 1701 to non-family firms, obtained from listed Canadian companies within the S&P/TSX Composite for 2014–2022. Firstly, we use the Canadian Family Index to categorize family firms. Secondly, we employed a mixed research analysis, mainly content analysis and quantitative data, to assess the socioemotional wealth perspective and its FIBER model. The results highlighted family firms’ positive effect on CSR performance. Further, findings have shown that within family firms, family control and influence, family identification, and emotional attachment positively affect CSR performance. However, family-binding social ties within family firms harm CSR performance. In addition, the renewal of family bonds through dynastic succession is not associated with CSR performance.

This article significantly contributes to the literature on CSR within family firms in several ways. Firstly, by investigating the relationship between family firms and CSR, this study adds to the ongoing debate regarding conflicting results on differences between family and non-family firms when it comes to CSR (Cruz et al. 2014; Sharma et al. 2020), mainly performance (Yu et al. 2015; García-Sánchez et al. 2021). Secondly, this study aligns with recent research endeavors that highlight the growing need for a deeper understanding of the heterogeneity of family firms regarding CSR (Hsueh 2016; Hsueh et al. 2023). By recognizing the complexity of socioemotional wealth through its different dimensions and influence on the decision-making process in family firms, this research enriches our comprehension of factors influencing CSR performance within family firms. Thirdly, this research contributes to the emerging stream of research focused on quantifying the dimensions of socioemotional wealth (Hsueh et al. 2023; Jain et al. 2023). Unlike most prior studies that rely on interviews, our study, along with few others, employs content analysis and quantitative data to assess the FIBER model. This innovative approach provides valuable insights into the measurement and analysis of socioemotional wealth dimensions. Lastly, this study provides new evidence on the relationship between Canadian family firms and CSR. It takes a step further by testing the influence of socioemotional wealth level through a score assessment. By examining the socioemotional wealth dimensions and their impacts on CSR, this research expands our understanding of the Canadian context. Our paper complements the recent study of Zeng (2021) by employing the total of FIBER dimensions to investigate the influence of family firms on CSR performance in the Canadian context. Further, following the approach of André et al. (2018), we have developed a score using the totality of socioemotional wealth dimensions, mainly FIBER, to examine the influence of the level of family socioemotional wealth on CSR performance.

The rest of the article is organized as follows. Section 2 provides a literature review and hypotheses development, which firstly investigate the relationship between family firms and CSR, secondly explores how the socioemotional wealth in family firms and, in particular, its FIBER dimensions affect CSR outcomes. Section 3 presents the sample selection and the methodology adopted to collect data and perform the analyses. Section 4 shows the descriptive, univariate, and multivariate results. Section 5 discusses the results obtained. Finally, Section 6 outlines the conclusions.

2. Literature Review and Hypotheses Development

2.1. Family Firms and Corporate Social Responsibility

While CSR has gained traction over the past decades (Yang et al. 2024), research on CSR in family firms has developed considerably (Mariani et al. 2023; Stock et al. 2023). Nevertheless, a consensus regarding the relationship between CSR and family businesses has not yet been reached (Cruz et al. 2014; Fehre and Weber 2019; López-González et al. 2019). Existing studies report conflicting results on the influence of family firms on CSR, including positive (Madden et al. 2020; García-Sánchez et al. 2021), negative (Dekker and Hasso 2016; El Ghoul et al. 2016; Abeysekera and Fernando 2020), and indistinct relationships (Labelle et al. 2018; Cui et al. 2018; Terlaak et al. 2018). These mixed findings regarding the relationship between CSR and family firms could be related to the existence of different theories (Lv et al. 2020). In fact, the relationship between family business and CSR could be formulated based on multiple theoretical perspectives that include but are not limited to the socioemotional theory (El Ghoul et al. 2016; Labelle et al. 2018; Bhatnagar et al. 2020; Brahem et al. 2021).

The socioemotional wealth theory was proposed as a homegrown theory for understanding family strategic decision-making (Berrone et al. 2010; Cruz and Arredondo 2016). In their seminal paper, Gómez-Mejía et al. (2007) introduced the concept of socioemotional wealth building on behavioral theory (Wiseman and Gomez-Mejia 1998). According to Gómez-Mejía et al. (2007), socioemotional wealth refers to the “non-financial aspects of the firm that meet the family’s affective needs, such as identity, the ability to exercise family influence, and the perpetuation of the family dynasty”. In other words, the decision-making process of family firms is driven not only by economic objectives but also by socioemotional goals such as family unity, identity, and continuity. The socioemotional wealth, viewed as an umbrella concept, is a prominent theoretical framework within the family business literature (Gómez-Mejía et al. 2011).

According to the socioemotional wealth theory, family firms aim to maintain, enhance, and prioritize the family’s socioemotional wealth over pursuing economic goals (Gómez-Mejía and Herrero 2022; Zona et al. 2023). Concerning CSR, family firms should invest more resources in CSR-related activities to preserve socioemotional wealth endowments. Consequently, family firms strive to avoid negative image and reputation that could affect their business and endanger their socioemotional wealth (Cennamo et al. 2012; Cruz et al. 2014). Regarding CSR performance, Yu et al. (2015) have found that Taiwanese family firms, proxied by family ownership, have better CSR performance than non-family firms. One potential explanation is that family firms have controlling owners prone to self-regulation by socioemotional wealth. Moreover, using a sample of US firms, Lamb and Butler (2018) show that family firms are more likely to value CSR, mainly by increasing CSR strength and reducing CSR concerns. These findings support the notion that family members may prioritize the socioemotional wealth facet of investing in CSR over a strict emphasis on maximizing short-term profits. Likewise, Cui et al. (2018) illuminates not just the favorable influence of family chief executive officers on CSR performance but also provide insights into how family firms can utilize long-term incentives to motivate non-family chief executive officers to prioritize the interests of the controlling families in preserving socioemotional wealth. Similarly, based on an international setting, López-González et al. (2019) and Chen and Liu (2022) provide evidence of higher CSR among family firms, as captured by ownership and management, compared to their non-family counterparts. Family firms’ commitment to CSR was attributed to the fact that the latter helps the firm achieve its socioemotional objectives. Taken together, these researches support the socioemotional wealth perspective, which predicts that family firms are beneficial for CSR. Thus, family firms are expected to provide additional value to CSR performance. Therefore:

H1. 

Family businesses are more likely to be associated with higher corporate social responsibility performance.

2.2. Socioemotional Wealth in Family Firms and Corporate Social Responsibility

Recently, Hsueh et al. (2023) recognized the heterogeneity of family patterns within and across societies. Thus, family firms are no longer seen as a homogenous group compared to their nonfamily counterparts. In particular, family firms’ heterogeneity has been explored using the socioemotional wealth perspective (Stanley et al. 2017).

The seminal work of Gómez-Mejía et al. (2007) is considered the reference point for the expansion of the SEW concept. Thus, we searched for the Socioemotional Wealth (SEW) in the titles of papers written in English indexed on the Scopus database from 2007. In order to limit these studies, following Reina et al. (2023), we have implemented the following criteria: First, the studies must come from journals that use peer-review systems and meet academic quality standards. Second, the family business context must be central to the manuscript. Third, the studies must explicitly follow the IMRAD structure (introduction, methods, results, and discussion).

A detailed analysis of the studies is presented in Appendix A. The publications are presented in chronological order to show not only the interest over time in studying the SEW but also the several attempts of exploring the multidimensionality of socioemotional wealth in family firms following the pioneer research of Berrone et al. (2012). In the latter study, they identified five dimensions of socioemotional wealth, known as the FIBER model, that are particularly relevant to family businesses: Family control and influence, which is related to the family members’ control and influence over the family firm’s strategic decisions, Family members’ identification with the firm, which is related to the close identity creation between family members and family firms; Binding social ties, which is related to the tight and trusting relationship between family members within family firms; Emotional attachment, which is related to the role of emotions within family firms; and Renewal of family bonds to the firm through dynastic succession, which is related to the time horizon of family firms.

These five FIBER dimensions serve as distinct reference points that may account for the diverse responses of family firms to different strategic outcomes (Gómez-Mejía et al. 2014). Regarding CSR, prior studies using socioemotional wealth theory proxied by FIBER dimensions to concretize the family firm’s heterogeneity yielded conflicting findings. In Family firms with chief executive officers serving as family members and owning high share ownership, the value of identification and commitment are evident, resulting in a substantial impact on the social pillar of CSR (Marques et al. 2014). However, drawing on medium-sized and unlisted Polish firms, Dick et al. (2021) provide evidence that founder-controlled family firms exhibit low levels of CSR engagement. Meanwhile, focusing on the Italian sample, Arena and Michelon (2018) show that with high family identity, middle-aged family firms provide more environmental disclosure. In contrast, old family firms provide less environmental disclosure. These studies overlook other aspects of heterogeneity among family businesses by neglecting the multidimensionality of socioemotional wealth. Hsueh et al. (2023) highlight the importance of considering the multidimensionality of socioemotional wealth, particularly all the possible socioemotional wealth dimensions (FIBER) identified by Berrone et al. (2012), to explain family firms’ heterogeneity concerning CSR. Thus, in the following, we explore Berrone et al.’s (2012) socioemotional model to understand how the FIBER dimensions affect CSR in family firms.

2.2.1. Family Control and Influence and Corporate Social Responsibility

The first dimension of socioemotional wealth applies to the family’s ability to control strategic decisions. The family’s drive to exert control and maintain authority often precedes financial prerogatives. The ability to exert control can be derived from a substantial ownership position, hereditary status, or personal charisma (Berrone et al. 2012). Thus, the control can be exerted directly by the founder for being a chief executive officer or chairman or indirectly through a dominant family coalition in the appointment of board and Top Management Team (TMT) members. Since control and influence are an integral part of socioemotional wealth and are highly desired by family members, these latter accept multiple roles in the family firms (Mustakallio et al. 2002; Sieger et al. 2011). In such circumstances, the family’s outsize authority in decision-making can significantly impact how family firms approach CSR.

Results on the impact of family control and influence on CSR present mixed evidence. On the one hand, firms with family control have more incentives to redirect firm resources, which can result in the expropriation of minority shareholders and ultimately lead to underperformance, including CSR performance, of family firms. Hsueh et al. (2023) suggest that without the motivation of biological, social, and material legacy transfer and opportunity perception through socioemotional wealth, family owners may choose an informal CSR strategy, limit CSR disclosure (Cabeza-García et al. 2017) and lower CSR performance (El Ghoul et al. 2016). On the other hand, the control and influence of family members can benefit CSR activities (Gómez-Mejía et al. 2011; Cui et al. 2018). To preserve or enhance their social and emotional wealth, family firms with concentrated ownership and controlled management are more likely to engage in CSR activities and positively impact CSR performance (Brahem et al. 2021). Therefore:

H2. 

Family firms in which the “Family control and influence” dimension of socioemotional wealth is the main frame of reference are associated with corporate social responsibility performance.

2.2.2. Family Members’ Identification with the Firm and Corporate Social Responsibility

The second socioemotional wealth dimension addresses the family’s close identification with the firm. Family firms possess a unique identity due to the blending of family and business (Dyer and Whetten 2006; Berrone et al. 2010). A family firm’s founder/owner’s identity is intertwined with the business that bears the family name (Berrone et al. 2012). In other words, the family’s societal standing is deeply tied to the organizational identity, as the business carries the family’s name (Beck and Prügl 2018). Consequently, the firm is perceived by stakeholders as an extension of the family. This perception will likely significantly impact internal attitudes (Teal et al. 2003; Carrigan and Buckley 2008). Therefore, to present not only a positive self-image but also an impression that appeals to a broader range of external stakeholders (Morsing and Spence 2019), The disproportionate power of the family in decision-making significantly impacts how family businesses approach CSR.

From the stakeholders’ point of view, family members may be motivated to seek CSR opportunities that mainly benefit family stakeholders (Spence 2016). Family members may be biased against “in-group members” or stakeholders who are connected to the family because of their shared past, as noted by Uhlaner et al. (2015) and Bettinelli et al. (2022). Furthermore, prior research suggests a strong connection families perceive between their image and reputation and their own firms (Dyer and Whetten 2006). Consequently, families hesitate to engage in irresponsible actions that could harm their reputations. With this respect, family firms are more likely to be associated with more responsible practices (Uhlaner et al. 2004). Indeed, family members often assign a higher priority to CSR as it serves to enhance their firm’s reputation, consequently strengthening and preserving the family’s socioemotional wealth (Dick et al. 2021).

Hsueh et al. (2023), while investigating the family firm’s commitment to formalizing the Italian firm’s CSR strategies, found that family identification may provide valuable signals. Although their study used fuzzy set qualitative comparative analysis (fsQCA), which considers that different configurations result in similar outcomes, their study suggests that family firms, especially those carrying the family’s name, are more inclined to engage in CSR practices. Similarly, based on a Canadian sample, Zeng (2021) provides evidence that family firms carrying the family’s names are more likely to be associated with higher CSR performance. Therefore:

H3. 

Family firms in which the “Family members’ identification with the firm” dimension of socioemotional wealth is the main frame of reference are more likely to be associated with higher corporate social responsibility performance.

2.2.3. Blinding Social Ties and Corporate Social Responsibility

The third dimension of socioemotional wealth refers to family firms’ social relationships. Family ties encompass the historical connections of genealogy, including relationships formed through kinship and marriage (Stewart 2003). These genealogical relationships between members influence their motivation to pursue strategic opportunities since their closeness often follows a path-dependent pattern (Arregle et al. 2007). Members with close ties in family firms commonly hold specialized knowledge and privileged information, leading to a collective comprehension of management practices (Martinez Jimenez 2009; Gedajlovic et al. 2012). In other words, the family’s history influences their strategic decision-making, including CSR.

First-generation family firms, where family members actively participate in the business, aim to establish a sustainable firm that can be passed down to their offspring. These family members, who share their everyday life and experiences, strive to create a business that can endure for future generations. Therefore, first-generation family firms are expected to pay high attention to socially responsible practices and engage in CSR activities (Campopiano and De Massis 2016). Contrarily, other researchers argue that first-generation family firms are more likely to be incentivized to survive in the market (Díaz-Díaz et al. 2022). Thus, limited resources may be left for CSR practices (Guo 2022).

Regarding later-generation family firms, where family members are usually less talented, committed, and entrepreneurial (James 2006; Miller and Le Breton-Miller 2011), they are more likely to minimize the efforts towards social activities and engage less in CSR (Campopiano and De Massis 2016). In contrast, later generations face a higher probability of having formal governance and control measures and established organizational routines that restrict family members’ discretion to extract personal benefits (Sciascia et al. 2014), which could preserve CSR activities. Further, later generations not only tend to have more formal educations and external work experience (Sonfield and Lussier 2004; Talke et al. 2010) but also are more aware of the family firms needs to response to external social and environmental pressure (Cruz et al. 2012), which could facilitate the expansion of CSR activities. To gain socioemotional wealth, later-generation families are more likely to consider their firms and family foundations as crucial in engaging in social activities (Klein et al. 2018). Therefore:

H4. 

Family firms in which the “Binding social ties” dimension of socioemotional wealth is the main frame of reference are more likely to be associated with higher corporate social responsibility performance.

2.2.4. Emotional Attachment and Corporate Social Responsibility

The fourth dimension of socioemotional wealth refers to the role of emotions in the family business context. Organizational work is inherently intertwined with emotions, as noted by Ashforth and Humphrey (1995). However, in settings where familial ties hold sway, a rich history of shared experiences and past events impacts present activities, events, and relationships. The combination of emotional factors stemming from family ties and business considerations is a key characteristic of family firms (Tagiuri and Davis 1996; Eddleston and Kellermanns 2007). Families themselves are marked by a broad spectrum of emotions, encompassing positive and negative feelings (Epstein et al. 2003). These emotions arise, evolve, and surface during pivotal events within each family business system (Gersick et al. 1997; Dunn 1999; Shepherd et al. 2009). Hence, family businesses are characterized by a transparent boundary between the family and the corporation (Berrone et al. 2010), leading emotions to seep into the organization and impacting the decision-making process of the family business (Baron 2008).

Emotional attachment within firms can significantly enhance CSR (Diéguez-Soto et al. 2021). While engaging in CSR activities, family members satisfy their need for belonging, affection, intimacy, and cohesion. The recognition received by family members following the implementation of CSR initiatives (Schulze et al. 2003) and the social support they receive from friends and acquaintances (Corbetta and Salvato 2004) fulfill these emotional needs. Consequently, family members who exhibit a stronger emotional attachment to the firm are likelier to demonstrate more significant social concern and prioritize CSR activities (Berrone et al. 2010). Furthermore, emotional attachment has been linked to the desire of family principals to protect their firm’s reputation (Martin et al. 2016), which is reflected in higher CSR performance. Drawing on this line of argumentation, the emotional structures of a family business are crucial for enhancing CSR performance. Therefore:

H5. 

Family firms in which the “Emotional attachment” dimension of socioemotional wealth is the main frame of reference are more likely to be associated with higher corporate social responsibility performance.

2.2.5. Renewal of Family Bonds to the Firm through Dynastic Succession and Corporate Social Responsibility

The fifth dimension of socioemotional wealth refers to the intention of handing the business down to future generations. This dynasty carries significant implications for the decision-making process, particularly regarding time horizons (Berrone et al. 2012). Rather than viewing the family firm as a disposable asset, family shareholders perceive it as representing their heritage and legacy (Casson 1999), thus considering it a long-term investment to be passed down to future generations (Berrone et al. 2010). Family firms prioritize preserving the business for posterity (Sieger et al. 2011), leading to extended planning horizons (Miller and Le Breton-Miller 2006; Miller et al. 2008; Nason et al. 2019).

To ensure a smooth transition of power to future generations and reduce dependence on predecessors’ influence, families driven to preserve their dynasty are motivated to establish a clear corporate vision and a system to support successors (Morsing and Spence 2019). As a result, family firms tend to engage in CSR strategies to secure the company’s future by being socially responsible (Aragón-Amonarriz et al. 2019; Hsueh et al. 2023). Furthermore, dynastic succession is seen as one aspect of high family involvement (Marques et al. 2014), leading to higher CSR activities (Stübner and Jarchow 2023). Given the above argument, we predict dynastic renewal may positively influence the firm’s CSR performance. Therefore:

H6. 

Family firms in which the “Renewal of family bonds to the firm through dynastic succession” dimension of socioemotional wealth is the main frame of reference are more likely to be associated with CSR performance.

3. Methodology

3.1. Data and Sample

This paper focuses on Canadian public companies listed on the S&P/TSX Composite Index from 2014 to 2022 (n = 2097). The sample starts in 2014 due to the non-availability of gender diversity data in earlier periods1, and ends in 2022. The categorization of family and non-family firms was carried out by reference to the NBC Canadian Family Index (NBC 2022). According to the National Bank of Canada, in the Family Advantage—Spring 2022 report, 44 Canadian family companies are controlled by either their founder(s) or by an individual or individuals who have significant influence over management because of their large ownership stake (NBC 2022).

Regarding the data sources, we supplement socioemotional wealth data from the NRG METRICS database with hand-collected chairman’s statements. In addition, CSR data was collected from Thomson Reuters’ Asset 4 database. Finally, accounting and financial data were gathered from the Mergent database.

Table 1. Sample distribution by industry.

Panel A: Sample Distribution by Industry
Two-Digit SIC Code	Industry Description	Non-Family Firms’ Frequency per Observations	Family Firms’ Frequency per Observations
10–14	Mining	612	36
15–17	Construction	18	0
20–39	Manufacturing	324	126
40–49	Transportation & Public Utilities	279	45
50–51	Wholesale Trade	50	0
52–59	Retail Trade	144	72
60–67	Finance, Insurance, & Real Estate	477	63
70–89	Services	198	45
91–98	Public Administration	9	9

presents the sample distribution by industry.

3.2. Variables Measurement

3.2.1. Dependent Variable

To evaluate CSR performance, prior studies have relied on ESG scores (e.g., Bae et al. 2019). The ESG score comprises three pillars: environmental, social, and corporate governance. The environmental score includes emission reduction, resource reduction and product innovation data. The social score includes data concerning employment quality, health and safety, training and development, diversity and opportunities, human rights, and community and product responsibility (Ertz et al. 2024). The governance score includes data about board structure and functions, compensation policy, shareholders’ rights, and vision and strategy. Each score reveals relative performance as a percentage ranging from 0 to 100 percent (Velte 2017).

3.2.2. Independent Variables

Family firm (FF). We identify family firms regarding the National Bank of Canada (NBC) Canadian Family Index. This latter considers a publicly traded company a family business if it is controlled by its founder(s) or individual (s) who hold at least 10% and 33.3% of the company’s voting rights, respectively. Thus, a family firm is a binary variable, coded as one if the company is included in the NBC Canadian family index and 0 otherwise.

Socioemotional wealth (SEW) dimensions. As aforementioned, we rely on the multidimensional scale (FIBER) developed by Berrone et al. (2012) to estimate our hypotheses to capture SEW scores. A mixed research analysis was employed, including content analysis and quantitative data.

Family control and influence (F). We identify the direct and indirect control and influence of the family member over family firm decisions (Berrone et al. 2012). More precisely, we have adopted three subdimensions to capture family control and influence suggested by Hsueh et al. (2023): Family Governance, family ownership, and Family Management. Firstly, Family Governance (FG) is measured by the percentage of family directors on the board. Secondly, Family Ownership (FO) is measured by the percentage of shares controlled by family members. Finally, Family Management (FM) is measured by the percentage of family members in the Top Management Team (TMT), including the chairman, vice president, and CEO. Therefore, in our main analysis the Family Control and Influence dimension is assessed via the sum of these three subdimensions.

Family members’ identification with the firm (I). We determine the degree to which family members associate their personality with that of the firm (Deephouse and Jaskiewicz 2013). Generally, the family members’ identity is tied to the family firm when it carries the family’s name (Cleary et al. 2019; Zeng 2021). Consequently, Family Identification is a binary variable, coded 1 when the family’s name is part of the firm’s name and 0 otherwise.

Binding social ties (B): We discern the genealogical relationship between members by capturing the family distance between family firms’ members (Hsueh et al. 2023). Therefore, binding social ties is a binary variable, coded 0 if the first generation manages the firm and 1 if the second and later generations manage the family firm.

Emotional attachment (E). Using NVivo 14 software, we assess emotional attachment as one of the SEW dimensions through a content analysis of the Chairman’s Statements retrieved manually from our sample companies’ annual reports. Firstly, the content analysis was suggested by Berrone et al. (2012) as a different and less obstructive approach to capturing SEW dimensions. Furthermore, this approach could be useful with data that are generally difficult to obtain by other means (Kabanoff et al. 1995). Additionally, despite some research in the family business literature using content analysis (e.g., Marett et al. 2018), this technique is underutilized (Cleary et al. 2019). Secondly, following Jain et al. (2023), the Chairman’s Statements are considered the most suitable to capture various FIBER dimensions of SEW in the case of family firms. In fact, in their message to stakeholders, chairmen mention their emotional ties. The software outputs provide frequencies for four scales of sentiments, mainly “very negative”, “moderately negative”, “moderately positive”, and “very positive”. Following Sender et al. (2021), Emotional attachment is measured as follows:

E = T₊ − T₋

where E is the emotional attachment, T₊ is the number of positive terms, and T₋ is the number of negative terms. As revealed in the Chairman’s Statements, a higher value of E corresponds to stronger emotional ties within family members. A value close to or equal to 0 indicates the absence of stronger sentiment.

Renewal of family bonds to the firm through dynastic succession (R). We identify the intrafamily succession in the top executives (Chief executive officer and chairman) to capture the dynastic renewal (Swab et al. 2020; Hsueh et al. 2023). Hence, the Renewal of family bonds to the firm through dynastic succession is a binary variable, coded 1 if the chairman or chief executive officer is the founder or descendant during our sample period 2014–2022 and 0 otherwise.

Table 2. FIBER dimensions description.

FIBER Dimension	Description
F	Measured by the sum of these three subdimensions: 1/ F.G: Measured by the percentage of family directors on the board. 2/ F.O: Measured by the percentage of shares controlled by family members. 3/ F.M: Measured by the percentage of family members in the Top Management Team (TMT), including the chairman, vice president, and CEO.
I	Measured by a binary variable coded 1 when the family’s name is part of the firm’s name and 0 otherwise.
B	Measured by a binary variable coded 0 if the first generation manages the firm and 1 if the second and later generations manage the family firm.
E	Calculated as follows after a content analysis of the Chairman’s Statements E = T+ − T− where E is the emotional attachment, T+ is the number of positive terms, and T− is the number of negative terms.
R	Measured by a binary variable coded 1 if the chairman or chief executive officer is the founder or descendant and 0 otherwise.

below summarizes the measurement of the FIBER dimensions used in our research.

3.2.3. Control Variables

We include an array of control variables in models testing CSR performance in family firms. Accordingly, we control for firm-specific characteristics, including firm size (SIZE), calculated as the log of total assets (Hwang et al. 2022), firm age (AGE), computed as the log of the difference between the current year and the founding year (Harjoto and Rossi 2019), financial leverage (LEV), quantified as the ratio of total liabilities to total assets (Kong et al. 2022), and financial performance, assessed as the return on asset (ROA) (Chen et al. 2019). Additionally, the legal system of Canada, excluding Quebec, which operates under civil law, is based on the common law (Viviers and Steenkamp 2023). Thus, we include origin law (OR_LAW) as a binary variable that takes 1 if the company is located in Quebec and 0 otherwise (Attig and Gadhoum 2003).

3.3. Models

To examine our first testable hypothesis, which states that family firm significantly impacts CSR performance, we estimate the following regression Model (1).

CSR_it = β₀ + β₁ FF_it + β₂ ROA_it + β₃ LEV_it + β₄ AGE_it + β₅ SIZE_it + β₆ OR_LAW_it + Year + Industry + ε_it

(1)

To provide empirical evidence for our second set of testable hypotheses (H2–6), which states that, due to family firms’ heterogeneity, socioemotional wealth dimensions within family firms significantly impact CSR performance, we estimate the following regression Model (2).

CSR_it = β₀ + β₁ + (FIBER_it × FIBER_it) + β₂ ROA_it + β₃ LEV_it + β₄ AGE_it + β₅ SIZE_it + β₆ OR_LAW_it + Year + Industry + ε_it

(2)

For all Models (i.e., Equations (1) and (2)), ε_it is the error term, and the subscripts i and t stand for firm and time, respectively.

4. Results

4.1. Descriptive Statistics

Table 3. Descriptive statistics.

Panel A: Descriptive Statistics of Variables Included in Our Baseline Models
	Full Sample					Family Firms’ Sample
	Mean	Median	Std. Dev.	Min	Max	Mean	Median	Std. Dev.	Min	Max
CSR	0.464	0.460	0.195	0.013	0.902	0.450	0.430	0.187	0.168	0.859
ENV	0.385	0.363	0.271	0	0.801	0.423	0.396	0.272	0	0.832
SOC	0.463	0.458	0.202	0.127	0.777	0.473	0.418	0.206	0.167	0.805
GOV	0.550	0.581	0.211	0.104	0.984	0.452	0.438	0.190	0.104	0.946
ROA	0.019	0.024	0.063	−0.142	0.111	0.035	0.042	0.059	−0.115	0.114
LEV	0.541	0.545	0.239	0.089	0.911	0.600	0.614	0.245	0.092	0.966
AGE	1.336	1.361	0.375	0.477	1.857	1.538	1.612	0.341	0.778	1.959
SIZE	9.554	9.545	0.669	8.333	10.577	9.724	9.782	0.600	8.456	10.507
Panel B: Descriptive Statistics of Our FIBER Dimensions within Family Firms
		Mean		Median	Std. Dev.		Min		Max
F.G		0.174		0.158	0.158		0.082		0.333
F.O		0.538		52.25	0.522		0.145		0.868
F.M		0.086		0.090	0.090		0		0.2
E		13.433		12.5	12.5		3		27
									Proportion
I		0							0.782
		1							0.217
B		0							0.295
		1							0.704
R		0							0.203
		1							0.796

(Panel A) reports summary statistics of all variables used to test our hypothesis for the full sample (n = 2097) and Family Firms Sample (n = 396). On average, the firms in our full sample (Family Firms Sample) have a CSR score of 46.4 (45.0), slightly below the average score reported by Zeng (2021). This result could be attributed to the fact that the ESG score retrieved from Thomson Reuters’ Asset 4 database does not include the economic dimension. Further, standard deviation results, i.e., 19.5 and 18.7 for the full and family firms’ sample, respectively, suggest considerable variation in the CSR performance score among Canadian firms. To enrich results related to CSR scores, we provide descriptive statistics of the environmental and social dimensions. Table 3 exhibits an average of 38.5 (42.3), 46.3 (47.3), and 55 (45.2) for environmental, social and governance scores, respectively, for the full sample (Family Firms Sample).

Regarding our control variables, Table 3 indicates that the means for ROA are 1.9% for the full sample and 3.5% for the family firms’ sample. These results prove that family firms outperform non-family firms, as shown in the NBC Advantage Report (NBC 2022). The mean leverage value is 0.541 for the full sample and 0.600 for the family firms’ sample. Further, the mean age is 34 years old, measured as the log of the difference between the current year and the founding year (Pukthuanthong et al. 2013). Finally, the average firm size, as measured by the logarithm of total assets, is 9.55, consistent with the average value reported by Zeng (2021).

Regarding the descriptive statistics of the FIBER dimensions for the family firms’ sample (n = 396), Table 3 (Panel B) reveals that 17.4% of board directors and 8.6% of the Top Management Team’s members are family members. Furthermore, on average, family members control 53.8% of the firms’ shares. These reported results align with prior studies regarding Canadian family firms (Pukthuanthong et al. 2013; André et al. 2014). The mean value of the E variable shows a relatively low existence of emotional ties among family members, as revealed in the Chairman’s Statements. Table 3 (Panel B) shows that about one-quarter of the family firms carry the family’s name. For the Binding social ties captured by family generation management, most of our family firms’ sample (70.4%) is managed by the second and later generations. Finally, 79.6% of our family firm sample have a family or descendant CEO or Chairman. These results are consistent with the findings of previous studies that provide evidence from the Canadian context (Bozec and Di Vito 2019; Zeng 2021).

4.2. Univariate Analysis

Table 4. Univariate test.

Variable	Family Firms		Non-Family Firms		t-Test
	Mean	SD	Mean	SD
CSR	0.450	0.186	0.467	0.195	−1.392 **
ENV	0.420	0.267	0.377	0.271	−2.508 ***
SOC	0.469	0.203	0.461	0.202	−0.582
GOV	0.453	0.189	0.573	0.208	−9.432 ***
ROA	0.031	0.061	0.017	0.063	−3.930 ***
LEV	0.591	0.232	0.529	0.239	−4.460 ***
SIZE	9.728	0.618	9.512	0.674	−5.518 ***
AGE	1.511	0.349	1.294	0.369	−10.545 ***

Notes: ** and *** denote significance at the 5% and 1% levels. All continuous variables are Winsorized at the 1% and 99% percentiles to deal with extreme values.

reports the mean values of all variables used to test our first hypothesis for family (n = 396) and non-family firms (n = 1701). A comparison of mean values between the two groups is performed using the parametric t-test. Univariate tests of differences demonstrate that CSR is slightly higher for family firms, which is consistent with the preservation of the socioemotional wealth perspective (Zeng 2021). To complement this finding, we presented the test results of differences regarding environmental, social and governance scores. Table 4 shows that environmental and governance scores exhibit a significant difference between the two groups. Family firms tend to be high environmental performers and well-governed (Sharma and Sharma 2011). Furthermore, compared to non-family firms, family firms are more likely to be superior economic performers, larger, more experienced in terms of age, and have higher debt levels.

4.3. Pairwise Correlation Matrix

Table 5. Pairwise correlation matrix.

	CSR	FF	ROA	LEV	SIZE	AGE	OR_LAW	VIF
CSR	1.000
FF	0.033	1.000						2.49
ROA	−0.012	0.089 ***	1.000					1.43
LEV	0.204 ***	0.100 ***	−0.004	1.000				1.99
SIZE	0.501 ***	0.126 ***	0.145 ***	0.561 ***	1.000			1.79
AGE	0.342 ***	0.229 ***	0.166 ***	0.158 ***	0.403 ***	1.000		1.60
OR_LAW	0.146 ***	0.133 ***	0.119 ***	0.251 ***	0.153 ***	0.220	1.000	1.27

Notes: *** p < 0.001. All continuous variables are Winsorized at the 1% and 99% percentiles to deal with extreme values.

reports the outcomes of the correlation analyses performed to confirm the absence of multicollinearity between our explanatory variables. According to (Kervin 1992), there is no serious problem of multicollinearity for the Pearson test if the correlation coefficients do not exceed 0.7. For the VIF, according to (Neter et al. 1983), a variable is regarded as highly collinear when its VIF value is superior to 10. Concerning the variables included in the baseline regression models, all the pairwise correlation coefficients are below 0.7. The highest revealed coefficient is 0.561. Moreover, the highest VIF value is 2.49, largely inferior to 10.

4.4. Multivariate Regression

In our research, since the family firm variable is time-invariant and the variations of SEW dimension-related variables are very minimal, we employed ordinary least squares (OLS) regressions and random effect regression with heteroscedasticity-adjusted standard errors (Tenuta and Cambrea 2022).

Table 6. Regression results.

Regression Results of Hypotheses H1–6 Dependent Variable: CSR
	Random									OLS
	Model 1	Model 2								Model 1	Model 2
FF	0.159 *** (0.030)									0.130 *** (0.012)
F.G		0.008 (0.091)									0.044 (0.103)
F.O			0.108 *** (0.119)									0.138 *** (0.82)
F.M				0.529 ** (0.086)									0.408 *** (0.111)
F					0.007 ** (0.008)									0.047 *** (0.010)
I						0.133 ** (0.160)									0.213 *** (0.039)
B							−0.038 *** (0.134)									−0.011 ** (0.034)
E								0.002 ** (0.000)									−0.000 (0.000)
R									0.017 (0.016)									0.048 (0.021)
ROA	0.019 (0.029)	0.041 (0.037)	0.038 (0.037)	0.036 (0.0.37)	0.060 (0.041)	0.038 (0.037)	0.052 (0.038)	0.097 ** (0.046)	0.039 (0.037)	−0.077 (0.039)	0.163 *** (0.040)	0.141 *** (0.042)	0.141 *** (0.041)	0.160 *** (0.048)	0.135 *** (0.039)	0.160 *** (0.044)	0.138 ** (0.060)	0.156 *** (0.041)
LEV	−0.026 (0.027)	0.107 ** (0.040)	0.104 ** (0.041)	0.098 ** (0.042)	0.123 *** (0.041)	0.105 *** (0.040)	0.107 ** (0.043)	0.125 ** (0.055)	0.105 *** (0.040)	−0.048 (0.024)	0.034 (0.038)	0.017 (0.039)	0.005 (0.039)	0.014 (0.044)	−0.048 (0.037)	0.016 (0.042)	0.015 (0.060)	0.023 (0.039)
AGE	0.092 *** (0.010)	0.117 *** (0.039)	0.120 ** (0.049)	0.122 *** (0.041)	0.112 *** (0.039)	0.120 *** (0.041)	0.132 *** (0.049)	0.189 *** (0.067)	0.118 *** (0.040)	0.033 *** (0.005)	0.077 *** (0.019)	0.082 *** (0.020)	0.090 *** (0.019)	0.087 *** (0.058)	0.064 *** (0.019)	0.080 *** (0.026)	0.184 *** (0.053)	0.081 *** (0.020)
SIZE	0.050 *** (0.005)	0.042 *** (0.014)	0.041 *** (0.014)	0.041 *** (0.014)	0.046 *** (0.014)	0.043 *** (0.014)	0.047 *** (0.016)	0.039 (0.024)	0.042 *** (0.014)	0.069 *** (0.003)	0.010 (0.017)	0.014 (0.017)	0.018 (0.017)	0.001 (0.016)	0.050 *** (0.018)	0.009 (0.020)	0.051 ** (0.026)	0.001 (0.017)
OR_LAW	0.020 (0.163)	0.011 (0.180)	0015 (0.235)	0.002 (0.214)	0.039 (0.175)	0.073 (0.222)	0.003 (0.231)	0.083 *** (0.154)	0.047 (0.182)	0.068 (0.060)	0.156 ** (0.066)	0.044 (0.062)	0.009 (0.058)	0.055 ** (0.062)	0.078 (0.057)	0.001 (0.063)	0.074 ** (0.076)	0.096 (0.057)
Constant	1.259 *** (0.162)	0.769 *** (0.309)	0.753 *** (0.338)	0.757 *** (0.311)	0.818 *** (0.308)	0.721 ** (0.314)	0.920 *** (0.358)	1.055 ** (0.538)	0.778 *** (0.306)	2.455 *** (0.086)	0.583 *** (0.376)	0.529 *** (0.375)	0.100 *** (0.367)	0.327 *** (0.367)	0.627 *** (0.377)	0.074 *** (0.429)	0.805 *** (0.604)	0.295 *** (0.378)
Year dummies	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Industry dummies	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Observations	2097	396	396	396	396	396	396	396	396	2097	396	396	396	396	396	396	396	396
Overall R-squared/R-squared	62.87%	73.03%	73.07%	73.20%	74.09%	75.20%	70.56%	68.95%	73.18%	63.53%	72.71%	71.03%	72.18%	73.06%	73.74%	68.20%	67.12%	71.32%

Notes: ** and *** denote significance at the 5% and 1% levels. All continuous variables are Winsorized at the 1% and 99% percentiles to deal with extreme values.

(Panel A) reports the regression results for Model 1 with the full sample (n = 2097) and Model 2 with the family firms’ sample (n = 396). The estimated coefficient of FF is positive and significant at 1%, suggesting that family firms are more likely to be associated with higher CSR performance. These findings indicate that family firms are more likely to enhance their CSR performance to preserve their socioemotional wealth. This result is consistent with the findings reported by Zeng (2021), which proves that Canadian family firms are associated with higher levels of CSR performance. Further, these results support the socioemotional wealth hypothesis, stating that family firms are more likely to be associated with higher CSR performance since the latter helps the firm achieve its socioemotional objectives.

To examine the association between family control and influence and CSR, we constructed the F dimension as the sum of Family Governance (F.G), Family Ownership (F.O), and Family Management (FM). Indeed, FG, FO, and FM are transformed into binary variables using the median. The estimated coefficient of F is positive and significant at 5% and 1%, suggesting that within family firms, control and influence of family members are more likely to be associated with higher CSR performance. This finding supports the socioemotional wealth perspective and suggests that family firms with higher control and influence are more likely to increase their CSR performance. This result is shown to be attributed to Family Ownership and Management. In fact, contrary to FG, which exhibits an insignificant coefficient, the estimated coefficients of FO and FM are positive and significant, suggesting that within family firms, the control and influence of family members are presented through family ownership and management. This result is consistent with the findings reported by Brahem et al. (2021), which prove that family firms with concentrated ownership and controlled management are more likely to engage in CSR activities and positively impact CSR performance to preserve or enhance their social and emotional wealth.

Additionally, Table 6 reveals that the estimated coefficient of I is positive and significant at 5% and 1%, suggesting that family-named family firms are associated with higher CSR performance. This result corroborates the findings of Zeng (2021), suggesting that concerns with reputational damages could motivate family-named Canadian firms to engage in more CSR activities.

Moreover, the estimated coefficient of B is negative and significant at 1% and 5%, suggesting that family firms managed by the first generation (second and later generations) are associated with higher (lower) CSR performance. Indeed, first-generation family firms are high CSR performers as they aim to establish a sustainable firm that can be passed down to their offspring (Campopiano and De Massis 2016). In contrast, second and later generations have poor CSR performance since they are usually less talented, committed, and entrepreneurial (Miller and Le Breton-Miller 2011; Campopiano and De Massis 2016).

Furthermore, the association between emotional attachment and CSR is partially proven. The estimated coefficient of E is positive and significant at the level of 5% (for the random estimation only), suggesting that emotional attachment within family firms is associated with higher CSR performance. Family firms with strong emotional attachment to the firm provide higher CSR performance due to their enormous desire for their family and firm reputations (Martin et al. 2016).

Finally, Table 6 exhibits that the estimated coefficient of R is insignificant, suggesting that renewal of family bonds through dynastic succession is not associated with CSR performance.

To sum up, our results reveal a significant positive relationship between family firms and CSR performance, indicating that family firms tend to be high CSR performers compared to non-family firms. This finding aligns with the socioemotional wealth perspective, which suggests that family firms are more dedicated to engaging in CSR activities to preserve their socioemotional wealth. Further, our findings demonstrate that the prominence of family control and influence, family identity, and emotional attachment strengthens family firms’ inclination to protect their socioemotional wealth by providing high CSR performance. Contrarily, when binding social ties are more salient, family firms are more likely to be poor CSR performers. Additionally, our results fail to find an association between the renewal of family bonds through dynastic succession and CSR performance. The signs of the control variables are generally consistent with prior research.

4.5. Robustness Checks

To check the robustness of our results, we ran several checks to address potential methodological concerns, primarily endogeneity and causality. First, endogeneity may arise due to omitted variable bias (Wooldridge 2010). Thus, we included other control variables (Board Size, Board Age, Board Meeting, CEO Duality, CEO Age, CSR Committee existence) identified as relevant in prior literature regarding family firms and CSR (Lv et al. 2020).

Table 7. Control variables used for robustness checks.

Variables	Description
Board Size (BS)	Measured by the number of board of directors’ members
Board Age (BA)	Measured by average age of board of directors’ members
Board Meeting (BM)	Measured by the number of meetings held by the board of directors’ members.
CEO Duality (CEOD)	Measured by a binary variable. It takes 1 if the CEO serves as the chairman of the board and 0 if not
CEO Age (CEOA)	Measured as the age of the CEO
CSR Committee Existence (CSRCE)	Measured by a binary variable. It takes 1 if a CSR committee exists and 0 if not

below summarizes the measurement of additional control variables used in our robustness checks.

Table 8. Regression results regarding endogeneity checks.

Regression Results of Hypotheses H2–6 Dependent Variable: CSR
F.G	0.117 (0.127)
F.O		0.130 ** (0.093)
F.M			0.184 ** (0.111)
F				0.005 *** (0.012)
I					0.501 *** (0.053)
B						−0.015 ** (0.064)
E							0.003 ** (0.001)
R								0.004 (0.025)
BS	−0.033 *** (0.005)	−0.033 *** (0.005)	−0.030 *** (0.005)	−0.032 *** (0.005)	−0.002 (0.005)	−0.035 *** (0.005)	−0.043 *** (0.008)	−0.032 *** (0.005)
BA	0.000 (0.002)	0.000 (0.002)	0.000 (0.002)	0.000 (0.002)	0.004 ** (0.002)	0.001 (0.003)	0.000 (0.004)	0.000 (0.002)
BM	0.000 (0.002)	0.000 (0.002)	0.001 (0.002)	0.000 (0.002)	0.001 (0.002)	0.001 (0.002)	0.006 (0.004)	0.000 (0.002)
CEOD	0.056 ** (0.024)	0.052 ** (0.023)	0.060 *** (0.024)	0.051 ** (0.025)	0.006 ** (0.020)	0.055 ** (0.025)	0.011 (0.039)	0.054 ** (0.024)
CEOA	0.005 *** (0.001)	0.005 *** (0.001)	0.005 *** (0.001)	0.005 *** (0.001)	0.001 (0.001)	0.005 *** (0.001)	0.006 *** (0.001)	0.005 *** (0.001)
CSRCE	0.049 (0.074)	0.035 (0.072)	0.044 (0.072)	0.045 (0.076)	0.001 (0.059)	0.030 (0.077)	0.030 (0.105)	0.037 (0.073)
Constant	0.743 *** (0.568)	0.815 *** (0.568)	0.737 *** (0.565)	0.612 *** (0.588)	0.015 ** (0.469)	0.577 *** (0.680)	2.044 ** (1.321)	0.776 *** (0.579)
Initial control Variables	YES	YES	YES	YES	YES	YES	YES	YES
Year dummies	YES	YES	YES	YES	YES	YES	YES	YES
Industry dummies	YES	YES	YES	YES	YES	YES	YES	YES
Observations	396	396	396	396	396	396	396	396
Overall R-	86.71%	86.79%	86.85%	86.81%	91.21%	85.99%	87.75%	86.64%

Note: ** p < 0.01, *** p < 0.001.

presents the regression results of Model 2 including initial and additional control variables. The results regarding our main variable of interest, FIBER dimensions, are highly similar to those revealed in our initial regression (Table 6). In fact, these results demonstrate that the prominence of family control and influence, family identity, and emotional attachment show a significant and positive influence on CSR performance. However, binding social ties presents a significant and negative influence on CSR performance. While we are cautious to not claim endogeneity, these findings provide further confirmation regarding the relationship between FIBER dimensions and CSR performance.

Second, causality is regarded as another potential source of endogeneity (Wooldridge 2010). This issue is not unique to the family firm context; scholars have recognized the challenges of addressing reverse causality when examining CSR drivers (Chiu and Sharfman 2011). To mitigate causality concerns, we measured the dependent variable, which is CSR performance, at t + 1 in the main analyses reported.

Table 9. Regression results regarding causality checks.

Regression Results of Hypotheses H2–6 Dependent Variable: CSRt+1
F.G	0.435 *** (0.121)
F.O		0.164 *** (0.093)
F.M			0.566 *** (0.124)
F				0.062 *** (0.011)
I					0.215 *** (0.042)
B						−0.009 ** (0.037)
E							0.003 *** (0.008)
R								0.049 ** (0.024)
Constant	1.126 *** (0.467)	0.375 *** (0.468)	0.179 *** (0.444)	1.146 *** (0.447)	0.291 ** (0.464)	0.628 *** (0.515)	2.105 *** (0.749)	0.848 *** (0.469)
Control Variables	YES	YES	YES	YES	YES	YES	YES	YES
Year dummies	YES	YES	YES	YES	YES	YES	YES	YES
Industry dummies	YES	YES	YES	YES	YES	YES	YES	YES
Observations	396	396	396	396	396	396	396	396
Overall R-	76.51%	75.43%	77.33%	77.98%	77.77%	73.26%	77.70%	75.57%

Note: ** p < 0.01, *** p < 0.001.

presents the regression results of Model 2 using CSR performance at t + 1. The results regarding the hypothesized relationships are significant. In fact, these results not only prove the positive effect of the prominence of family control and influence, family identity, and emotional attachment and the negative influence of binding social ties on CSR performance, it show the positive effect of the renewal of family bonds through dynastic succession failed to be founded in the initial regression (Table 6). This result demonstrates that family firms tend to engage in CSR to secure the company’s future by being socially responsible.

While we are cautions no to claim causality, these results prove further the established relationship between FIBER dimensions and CSR performance.

4.6. Supplementary Analysis

In order to further explore the negative association between family firms and CSR, shown in Table 6, we have developed a SEW score using FIBER dimensions, following the approach of André et al. (2018). Indeed, SEW is assessed using the following formula:

SEW = F + I + B + E + R

(3)

A higher value of SEW indicates that the family has absolute control and influence over the business, can identify with the firm through the name, possesses higher binding social ties, and is firmly emotionally attached, especially when dynastic renewal exists. Further, SEW takes 0 if the firm is a non-family business.

Table 10. Supplementary analysis.

	Dependent Variable: CSR Model 1
	Random	OLS
SEW	0.030 *** (0.005)	0.038 *** (0.003)
ROA	0.021 ** (0.029)	0.080 ** (0.039)
LEV	−0.019 (0.027)	0.039 ** (0.024)
AGE	0.090 *** (0.010)	0.036 *** (0.005)
SIZE	0.050 *** (0.006)	0.070 *** (0.003)
OR_LAW	0.042 (0.142)	0.040 (0.113)
Constant	1.255 *** (0.162)	1.494 *** (0.081)
Year dummies	YES	YES
Industry dummies	YES	YES
Observations	2097	2097
Overall R-squared/R-squared	63.20%	64.09%

Notes: ** p < 0.01, *** p < 0.001. All continuous variables are Winsorized at the 1% and 99% percentiles to deal with extreme values.

provides the results of Model 1 with SEW as our independent variables. The estimated coefficient of SEW is positive and significant at 1%, suggesting that family firms are more likely to be associated with higher CSR performance to preserve their socioemotional wealth. More precisely, family firms with family members that have absolute control and influence over the business can identify within the firm throughout the name, possess higher biding social ties, are strongly emotionally attached, and when there is a dynastic renewal, enhance their CSR performance to preserve their socioemotional wealth further. Overall, the results support the robustness of our findings.

5. Discussion

The prior research on CSR performance, although it has consistently provided evidence implying that family firms, based on ownership, influence CSR performance, it has not widely associated the FIBER dimensions, as developed by Berrone et al. (2012), to CSR performance. We therefore add to prior literature by exploring the heterogeneity of family firms under the socioemotional wealth perspective via the FIBER dimensions.

The testing a large sample of listed Canadian companies, we primarily find that family firms are more likely to enhance their CSR performance to preserve their socioemotional wealth. Testing for the heterogeneity of family firms through FIBER dimensions, we reveal that each specific dimension significantly influences CSR performance, highlighting the importance of socioemotional motivations in shaping corporate behavior.

We suggest that giving support for the positive effect of family firms on CSR performance could be explained by their aim to maintain, enhance, and prioritize the family’s socioemotional wealth over pursuing economic goals (Gómez-Mejía and Herrero 2022; Zona et al. 2023). In other words, family firms strive to avoid negative image and reputation by being high CSR performers, which could affect their business and endanger their socioemotional wealth (Cennamo et al. 2012; Cruz et al. 2014).

Furthermore, we suggest that giving support for the influence of socioemotional wealth dimensions, separately and in combination, emphasizes further the heterogeneity among family firms. In tune with the premise that family firms are not a monolithic group, we add to the literature by showing that the prominence of family control and influence, family identity, and emotional attachment strengthens family firms’ inclination to protect their socioemotional wealth by providing high CSR performance. Contrarily, when binding social ties are more salient, family firms are more likely to be poor CSR performers.

Collectively, we believe that our baseline research findings lend credence to the socioemotional wealth theory advocates who argue that the decision-making process of family firms is driven not only by economic objectives but also by socioemotional goals such as family unity, identity, and continuity. The reason behind this is that family firms, especially through the five dimensions of socioemotional wealth tend to value CSR by exhibiting better CSR performance.

6. Conclusions

Despite the increasing interest and growing research on family firms and CSR, we still know very little about how family firms cope with CSR performance and why they differ along this line concerning non-family firms. This paper explores whether family firms’ CSR performance differs from that of non-family firms. Family firms are more likely to be associated with higher CSR performance under the socioemotional wealth perspective since the latter helps the firm achieve its socioemotional objectives. Further, our paper addresses the heterogeneity of family firms by exploring the influence of FIBER dimensions developed by Berrone et al. (2012) on CSR performance.

Using a data set of 33,414 firm-year observations, of which 352 observations relate to family firms and 1304 observations to non-family firms, obtained from listed Canadian companies within the S&P/TSX Composite for the period 2014–2022, our results show that the CSR performance of family firms is significantly higher than that of non-family firms. This significant positive association between CSR performance and family firms corroborates with the socioemotional wealth perspective, which states that family firms are more committed to corporate social activities to preserve socioemotional wealth endowments. Also, this finding rejects the agency hypothesis, which states that family firms refrain from investing in socially responsible activities as it threatens their control and influence over the company. Further, drawing on the FIBER five-dimensional model developed by Berrone et al. (2012), we demonstrate the importance of recognizing the existence of different reference points that shape the influence of family firms on CSR performance. More specifically, when family control and influence, family identity, and emotional attachment are more prominent, family firms’ inclination to protect their socioemotional wealth will make them more willing to provide high CSR performance. By contrast, family firms will be associated with poor CSR performance when binding social ties are more salient. Additionally, the renewal of family bonds through dynastic succession is unrelated to CSR performance.

This article provides essential contributions to the literature at the intersection of the CSR and family business fields. This article is the first to investigate CSR performance in Canadian family firms by considering the socioemotional perspective and the FIBER dimensions. First, although abundant research addresses the relationship between family firms and CSR, we still lack understanding about how the active presence of the family in the business intervenes in CSR performance. This article contributes to the ongoing debate regarding conflicting results on CSR differences between family and non-family firms. It also enriches our comprehension of factors influencing CSR performance within family firms. This article goes beyond the dichotomization between family and non-family firms by exploring how family firms’ different frames of reference influence CSR by recognizing the complexity of socioemotional wealth through its different dimensions. By so doing, it aligns with recent research on the heterogeneity of family firms concerning CSR. Second, this article constitutes a first attempt to quantify all the FIBER dimensions of socioemotional wealth by using content analysis and quantitative data and to develop a SEW score that assesses socioemotional wealth level.

Thirdly, this research contributes to the emerging stream of research focused on quantifying the dimensions of socioemotional wealth (Hsueh et al. 2023; Jain et al. 2023). Unlike most prior studies that rely on interviews, our research and a few others employ content analysis and quantitative data to assess the FIBER model. This innovative approach provides valuable insights into the measurement and analysis of socioemotional wealth dimensions. Lastly, this study provides new evidence on the relationship between Canadian family firms and CSR. It takes a step further by testing the influence of socioemotional wealth level through a score assessment. This research expands our understanding of the Canadian context by examining the socioemotional wealth dimensions and their impacts on CSR. Our paper complements the recent study of Zeng (2021) by employing the total of FIBER dimensions to investigate the influence of family firms on CSR performance in the Canadian context. Further, following the approach of André et al. (2018), we have developed a score using the totality of socioemotional wealth dimensions, mainly FIBER, to examine the influence of the level of family socioemotional wealth on CSR performance.

This article also has important implications from a practical point of view. As mentioned, the socioemotional wealth dimensions can be negatively associated with CSR. Concerning our results, later generations managing the firm jeopardize CSR performance. Thus, family owners and managers must be conscious of the negative potential outcomes for the socioemotional wealth dimensions, and binding social ties, in our case, can work. However, it is important to exercise caution as these findings are specific to the context of Canada, and their generalizability should not be assumed.