Women’s leadership and SMEs’ CSR performance: Family versus nonfamily firms

Abstract

The purpose of this study is first to examine the impact of family control on SMEs’ corporate social responsibility (CSR) performance and, second, investigate how the proportion of female managers in the top management team (TMT) is related to CSR performance in family versus nonfamily SMEs. To answer these questions, we use panel data of manufacturing SMEs from the database of UNU-WIDER over a period from 2011 to 2015 and run fixed effect regressions. Our findings indicate that first, nonfamily SMEs outperform family ones in terms of CSR performance; second, similar to the previous idea, we find a significant and positive relationship between the increasing presence of female managers in the TMT and CSR practices. However, we also find that this positive relationship only occurs with nonfamily SMEs and not with family SMEs. This paper focuses specifically on Vietnamese SMEs and CSR performance. It tests the impact of family ownership on this relationship, as well as the moderating effect of this ownership on the role of female managers in promoting CSR performance.

Keywords:

• CSR performance
• family SMEs
• female manager
• nonfamily SMEs
• Vietnam

JEL:

• M14

1. Introduction

The participation of women in the labour force leads to their increasing presence in corporate leadership positions. Women account for 15% of corporate board members in the USA, UK, Canada, Australia (Terjesen & Singh, 2008) or at least 40% representation on boards of directors in Norway (Ahern & Dittmar, 2012). In Vietnam, based on the data from the World Bank, female leaders account for approximately 23% and 71% in household businesses and SMEs, respectively (Pham & Hoang, 2019). Previous studies have investigated whether female leaders can be different from their male counterparts and bring unique values to the firms. Compared with their male counterparts, female leaders are more emotional and social orientation (Groves, 2005; Taylor & Hood, 2011), more concerned about others (Van Emmerik et al., 2010), selfless (Deaux & Kite, 1993), and have better communication and listening skills (Eagly & Carli, 2003; Schubert, 2006). Due to these unique values and based on upper echelon theory, scholars claim that the presence of female executives in management team can have a favourable influence on nonfinancial performance. For example, corporations with women on their management boards have a higher orientation toward corporate social responsibility (CSR) and are more social and environmentally conscious than male CEOs (Biswas et al., 2021; Cook & Glass, 2018). Some studies, however, indicate that female leadership is neither significantly nor adversely connected to CSR performance (Amorelli & García-Sánchez, 2021).

Women make up 22% of family business top management teams, 55% of family firms have at least one female board member, and 70% of family businesses are currently considering hiring a woman as their next CEO (Chadwick & Dawson, 2018). However, compared to males, women are underrepresented in top leadership positions and have limited participation in the family company. Male heirs continue to be favoured over females as chief executive officers (Ahrens et al., 2015). Instead of significant and official business-related positions, such as CEO or CFO, women in family enterprises have responsibilities that are strongly tied to the family, such as spouse, mother, or mother-in-law (Bjuggren et al., 2018). Ruigrok et al. (2007) argue that female managers are chosen in family firms because of their family ties, not by their experience and knowledge. The objective is to protect the family’s control and ownership (Abdullah, 2014). Thus, female in family enterprises have been called invisible (Burke, 1997) due to the fact that their function is often restricted to giving emotional leadership Martinez Jimenez, 2009). Similarly, in Vietnamese context, Vietnamese society believes that females are responsible for housework and childcare and not appropriate for business-related work (Pham & Hoang, 2019). This also diminishes women’s leadership responsibilities in Vietnamese family-owned enterprises.

As mentioned above, female presence in the TMT may be a good sign of CSR-related activities. However, this linkage may be moderated in numerous ways by family control, since female leadership positions in family and non-family enterprises might vary. As mentioned above, this is because the role of female managers is not appreciated in family business and lose the right to make their own decisions. Moreover, according to Chadwick and Dawson (2018), because of constraints on management discretion in family firms, the upper echelon theory may not be applicable in the same degree in family businesses as in nonfamily enterprises. Therefore, it is necessary to have studies that link female leadership with CSR performance in family-controlled enterprises. Second, despite the fact that many earlier studies have been conducted, there is a shortage of research on the link between family SMEs and CSR performance in developing nations.

To fill this gap, this study has several objectives. First, this paper investigates the relationship between family SMEs and CSR performance. Second, we determine whether female managers have a favourable influence on the CSR performance of SMEs and if this effect is moderated by family control. To achieve these research goals, our study uses a strongly balanced data set from the UNU-WIDER database for 2011, 2013, and 2015. Theseis data consist of 1720 SMEs from 5160 firm-year observations, of which 3440 observations are family firms. Our empirical findings suggest that first, family control has a negative effect on CSR performance at both employees and environmental dimensions. Second, our research indicates that the increasing presence of women on the management team is positively related to employee dimensions but insignificant to environmental dimensions in nonfamily SMEs. However, this linkage is not seen across all CSR dimensions for family-owned SMEs.

This paper makes several implications to the knowledge of family SMEs and female leadership in family-controlled enterprises. From theoretical implication, based on socioemotional wealth theory, this study shows that the preservation of socioemotional wealth does not always mean that family ownership impact positively on nonfinancial performance. Second, drawing from Upper echelon and Socioemotional wealth theory, the research results demonstrate that family-controlled organizations can decrease the managerial discretion of managers and thus, decline the benefit that female managers can bring to family firms. From practical implication, this study shows that first family SMEs should pay more attention to CSR performance. Second, due to the limited role of female executives in family enterprises, it is necessary to promote the role of female executives in family business, which can improve CSR performance level in these firms.

2. Theory and hypothesis development

2.1. The family control—CSR performance relationship

Socioemotional wealth (SEW) theory posits that important strategic decisions and policy decisions of family businesses are primarily motivated by the maintenance of their SEW (Gómez-mejía et al., 2007). SEW involves nonfinancial characteristics or emotional tie between family members and their businesses (Berrone et al., 2012), which is stronger than the connection between entrepreneurs and nonfamily businesses (Yu et al., 2015). Previous research demonstrates that socio-emotional wealth preservation is a family business’s primary objective, even above financial performance, owing to fears about losing firm control (Berrone et al., 2012). Hence, Scholars believe that family businesses are more likely to survive and preserve noneconomic qualities including identity, pride, longevity and control, heritage, and tradition (Berrone et al., 2012; López-Pérez et al., 2018; Vandekerkhof et al., 2019). Sustainability practices are considered long-term goal that family firms prioritize than short‐term orientation (Antheaume et al., 2013). Gils et al. (2014) also argue that family-owned businesses are more aggressive in addressing social concerns and fostering relationships with their stakeholders than their nonfamily rivals. To preserve image and reputation, for instance, family companies operating in polluting industries tend to have a higher environmental performance, are more socially responsible in social concern dimensions, and are less likely to conduct irresponsibly than nonfamily organisations (Berrone et al., 2010; Dyer & Whetten, 2006).

However, most previous findings are from large family firms in developed countries. The adoption of the CSR management system, certifications, and yearly CSR reports may be expensive for SMEs due to the issue of insufficient resources, time, and experience (Lund-thomsen et al., 2014). In addition, Khan et al. (2007) suggest that SMEs in developing countries usually involve in labour rights abuses such as forced labour, child labour, human trafficking, and sexual harassment. These problems occur especially in developing countries because SMEs are not subject to the same level of inspection, including being monitored by non-governmental organizations, international auditors and nation-based trade unions as in developed countries in Europe and North America (Jamali et al., 2015; Lund-Thomsen & Lindgreen, 2013). In the context of Vietnam, in addition to the lack of financial resources, the deficiency of understanding of CSR concept may obstacle the adoption and implementation of CSR (M. M. Nguyen et al., 2018). In the study taken by Vo and Arato (2019), only a small fraction of Vietnamese SMEs are known to have made major efforts to integrate CSR into their company; the others see CSR as an expense and do not apply it voluntarily. The shortage of CSR awareness from SMEs’ managers is another problem. Many of them consider environmental, social and ethical problems are not important, costly and not compulsory by law (Vo & Arato, 2019). In addition, whereas earlier research has shown a beneficial association between family-controlled businesses and CSR performance based on the SEW theory, SEW preservation may also make family businesses less willing to participate in activities that diminish their financial performance (Chrisman & Patel, 2012; Cruz et al., 2014; Gómez-mejía et al., 2007). It is called the “dark side” of SEW when family firms are more concerned with their interests rather than social or environmental activities (Kellermanns et al., 2012). CSR investment may be seen inappropriate, expensive, and detrimental to a company’s financial performance by the owners of family-owned small and medium-sized enterprises (SMEs) whose primary problem is a lack of capital and whose primary priorities are firm survival and continuity. Based on arguments listed above, we posit the following hypothesis

Hypothesis 1. Family control decreases the level of SMEs’ CSR performance.

2.2. Female managers and CSR performance relationship

Upper echelon theory states that organizational outcomes, strategic choices and decisions can be impacted by decision makers’ background traits or characteristics (Hambrick & Mason, 1984). To be more specific, the personal characteristics of managers such as specialization, tenure, age and gender affect their decisions and performance, leading to the impact on corporates’ strategic choice and organizational outcomes (Nielsen, 2010). Based on this theory, numerous researches have examined the relationship between the gender of executives and organizational outcomes such as financial leverage (KiongTing et al., 2015); financial performance (Dezsö & Ross, 2012); corporate cash holdings (Doan & Iskandar-datta, 2019) and innovation (J. Chen et al., 2018).

Empirical research demonstrates that the presence of women in managerial roles has a beneficial effect on corporate social responsibility (CSR) due to their distinct qualities in comparison to their male colleagues (Lu et al., 2020). For instance, since women are more helpful, kind, pleasant, community-oriented, sensitive, giving, and thoughtful, they are more inclined to implement CSR techniques (Cook & Glass, 2018; Ibrahim & Hanefah, 2016; Lone et al., 2016; Lu et al., 2020). In addition, they are more inclined than males to value long-term non-financial achievement above short-term financial performance. In particular, they highlight creativity, fairness, teamwork, and equality, which are long-term commercial results that women may promote (Cook & Glass, 2018; Dijk et al., 2012). Moreover, they are influenced by gender role norms that require them to be affable, kind, and community-oriented while avoiding harshness and aggression (Cook & Glass, 2018). In the context of Vietnamese culture, gender stereotypes, and social conventions about women’s roles may be gleaned through social media, leading to a distinction between male and female leadership styles, such as friendlier, more thoughtful, more empathetic, and more adaptable (Vu et al., 2017). Because of these reasons, several articles have shown that female CEOs have a favourable correlation with CSR performance (Hyun et al., 2022; Velte, 2016; Zou et al., 2018).

Thus, from the statements above, we suggest the hypothesis that:

Hypothesis 2. There is a positive relationship between the presence of female managers in the TMT and CSR performance.

2.3. Gender managers in the TMT and family firms’ CSR performance

Although the majority of prior study has relied on the Upper Echelon theory to determine the connection between female managers and CSR performance, researchers have shown the inadequacy of upper echelon theory (Hambrick, 2007). This is because focusing solely on the observable qualities of executives, such as gender and family-related demographics, may make it difficult to understand the socioemotional factors that also contribute to influencing executives’ cognition and behaviour (Hambrick, 2007). Since socioemotional wealth preservation is the priority of family firms, (Bauweraerts et al., 2022) claim that taking into account the moderating impact of socioemotional elements on demographic attributes is necessary to comprehend the impact of female executives on the behaviour of family-controlled enterprises. It is probable that female executives in family organisations are affected by socioemotional preferences, and their positive impacts on CSR performance are not as significant as in nonfamily enterprises.

To be more specific, because of SEW preservation priority, family control and influence can reduce female roles in the firm, and they have less power to influence the firm’s decisions. Carney (2005) suggest that the family tend to use their power and influence to adjust decision-making processes in the firm. In other words, the management discretion of managers is subdued because the family interfere with their decision (Chadwick & Dawson, 2018). Moreover, managers are expected to preserve and satisfy the family’s interests instead of making decisions independently. Based on the Upper Echelon theory, this situation can mitigate the positive, unique value female managers can bring to the firm as well as CSR performance (Chadwick & Dawson, 2018).

Second, the appointment of women in management positions maybe not because of their experience, skill and educational level but because of their close, blood, or material relationship with the owning family members (Bettinelli et al., 2019). In some situations, their appointment results from families’ control and influence preservation, called “family delegate” (Abdullah, 2014). Even if female members are better educated and experienced than their male counterparts, they are highly likely not to be chosen in succession or important positions in family businesses (Martin, 2001). The insignificance of female managers’ roles in family-owned enterprises might restrict their influence and diminish their beneficial influence on CSR performance.

Third, gender discrimination and social prejudices in families are other concerns. Traditional social prejudices argue that women are not suitable for leadership positions, and their presence in business links to a lower likelihood of success than men (Harveston et al., 1997; Martinez Jimenez, 2009; Salganicoff, 1990). Social role theory also suggests that because women are considered weak and vulnerable, they are usually advised to do “light work” in families such as taking care of their children and doing housework, instead of working in a competitive business environment (Eagly & Wood, 2016). Family firms do not assign leadership positions to women because they consider women “incompetent knowledge and skill” to manage the firm (Dumas, 1998). Similarly, Vietnam is where Confucian gender ideology impacts many social aspects. Due to this influence, the role of Vietnamese women in families is not respected and overshadowed by men (Pham & Hoang, 2019). Housework takes up most of their time and challenges women to achieve high positions in business (Pham & Hoang, 2019). In general, gender discrimination and social prejudices can mitigate the influence of female managers on family firms’ CSR performance.

To sum up, we suggest the third hypothesis:

Hypothesis 3. The positive relationship between female presence in the TMT and CSR performance is weaker in family SMEs than nonfamily ones.

3. Research method

The research data in this paper is drawn from the database of UNU-WIDER of Vietnamese SMEs for three years 2011, 2013, and 2015 in nine provinces of the country. These SMEs meet the definition of the Law on Enterprises of Vietnam in 2014 and are chosen randomly. This database has been collected through surveys, based on face-to-face interviews with SMEs’ owners, managers, and employees. The database includes corporate information (financial indicators, performance, employees, governance, and nonfinancial performance) and owner/managers’ characteristics (gender, education, ethnicity, age …). The data and sampling method can be checked more detail from the UNU-WIDER website. After deducting firms that do not exist in all years, we create a strongly balanced sample taken from this database. The final sample includes 5160 firm-year observations (the number of observations in each industry is in the table 7). In this sample, there are 3440 firm-year observations and we define them as family SMEs. This is because, according to the Law on Enterprises of Vietnam, household business is a business model owned by only one individual, a legal citizen of Vietnam, or a family. Based on the definition of Lussier and Sonfield (2015) study, these household businesses can be defined as family SMEs.

3.1. Variable measurement

Our primary goal in this paper is first, the relationship between family ownership and CSR performance; second, the relationship between the participant level of female managers in the TMT and CSR performance; third, whether this relationship is moderated by family control.

3.1.1. Dependent variables

CSR performance is assessed by questions in the survey of UNU-WIDER database. Owners/managers answer these questions (yes/no question). We score “1” if the answer is “Yes” and “0” if the answer is “No” in each question. There are 21 questions, so the maximum score of CSR performance is 21. Two dimensions of CSR: ERP (employee responsibility performance) and ENV (environment responsibility performance) have the maximum score of 12 and 9, respectively. Detail about questions can be seen in the table 8. The CSR performance index is calculated by the sum of all scored in each question, divided by the maximum possible score (21). We follow the equation presenting for the calculation of CSR index (Dias et al., 2017):

X = ${\sum }_{j=1}^e\frac{\mathit{\text{ej}}}{e}$

Where:

X = (CSR, ERP and ENV)

CSR: CSR performance

ERP: Employee dimension of CSR performance

ENV: Environment dimension of CSR performance

ej = Number of scores that a firm achieve

e = The maximum score that a firm can achieve (maximum is 21 with CSR, 12 with ERP and 9 with ENV).

3.1.2. Independent variables

FAM (family firm or not). It equals one if the firm is a family firm and 0 otherwise. In this paper, we define that family SMEs are household businesses because they satisfy the family SMEs definition of (Lussier & Sonfield, 2015) study

FEMANAGER: The rate of female managers in the TMT. We calculate this rate by the ratio of female members divided into total members in the TMT.

3.1.3. Control variables

CSR performance is known to be influenced by multiple factors. Thus, we add some firm-level control variables in our regression model to isolate the impact of the independent variables on CSR performance. To be more specific, we control total labour force (ESIZE) and total asset because a firm with a larger scale can impact CSR performance (Udayasankar, 2008). We also control for the percentage of female employees in the total workforce because we suggest that this rate can impact the percentage of female managers and impact CSR performance. ROA indicator (return on assets) can impact the extent of CSR performance. Several studies prove that there is a relationship between financial indicators and CSR performance (Kuzey et al., 2021). Similarly, previous studies have found a significant relationship between revenue as well as leverage and CSR performance level (Z. Chen & Hamilton, 2020). Hence, we add REV (revenue) and LEV (leverage) variables in the estimations. Another financial indicator should be included in the cash holding ratio. The cash holding ratio level is associated with the extent of CSR performance (Yang & Susanto, 2020). Finally, according to the study of Jo and Harjoto (2011), firm age is highly associated with CSR engagement level, so we add this control variable in the model. All control variables are shown in Table 1

Table 1. Control variables description

Variable	Definition	Measurement
FAM	Family firm or not	Equals one if the firm is family firm and 0 otherwise
FEMANAGER	Female proportion in the TMT	The ratio of female members divided to the total member in the TMT
ESIZE	Total labour force	The natural logarithm of total number of employees (lagged one period).
WOSHARE	Female proportion in the total labour force	The ratio of female employees divided to total employees
FSIZE	Firm size (total asset)	The natural logarithm of the total asset (lagged one period)
ROA	Return on assets	Profit before tax, divided into the total asset (lagged one period)
REV	Revenue	The natural logarithm of revenue (lagged one period)
LEV	Leverage	The ratio of debt divided to the total asset (lagged one period)
CASH	Cash holding ratio	The natural logarithm of the ratio: cash divided to the total asset (lagged one period)
FAGE	Firms’ age	The age of firm (lagged one period)
Ɛ	Error term
k	Industry
t	Time dimension (year)
i	Cross-section dimension (firm)

3.2. Regression model

To examine the influence of family control on family SMEs’ CSR performance and the impact of female manager participant level on CSR performance, we use ordinary least squares regression models. In these models, first, we apply panel fixed effects (industry and year fixed effects). Second, all independent and control variables are lagged one period (two years). In this study, we apply two regression models for testing three hypotheses. The first model (for the sample of all firms) tests the first hypothesis. The second model (for the sample of family/nonfamily SMEs) tests the second and third hypotheses. We divide the sample to two sub-samples: nonfamily and family SMEs and compare the impacting direction of female manager rate on CSR performance in family and nonfamily SMEs. This method has been used in previous study of Bjuggren et al. (2018). Stata 14 software is used to manage and analyse data. Due to the dependent variables are fractional and vary between 0 and 1, we take the technique of Fractional Response Generalized Linear Models and use fracglm command in Stata (Williams & Dame, 2019).

3.2.1. The first model

CSR/ERP/ENV_it = Lag_FAM_it + Lag_FEMANAGER_it + Lag_ESIZE_it + Lag_WOSHARE_it + Lag_FSIZE_it + Lag_ROA_it + Lag_REV_it + Lag_LEV_it + Lag_CASH_it + Lag_FAGE_it + INDUSTRY_k + YEAR_t + Ɛ_it

3.2.2. The second model

CSR/ERP/ENV_it = Lag_FEMANAGER_it + Lag_ESIZE_it + Lag_WOSHARE_it + Lag_FSIZE_it + Lag_ROA + Lag_REV_it + Lag_LEV_it + Lag_CASH_it + Lag_FAGE_it + INDUSTRY_k + YEAR_t + Ɛ_it

Table 2 provides the descriptive statistics. In a range from 0 to 1, the average value of ERP and ENV are 0.2407 and 0.1248, respectively, implying that firms within the sample have low degrees of CSR performance (the mean value of CSR performance is 0.1909). The mean value for family firms (FAM) is 0.6667, which means that 66.67% of the SMEs in the sample is family firms. 38.09% of members in the TMT are female, which is coincidentally also the ratio of female employees to total employees (38.01%). The size of businesses in terms of the total labor force and total asset has an average value of 16.0793 and 5,827,566 (thousand dongs) separately. Next, the table represents the mean values for profitability and revenue, which are 0.2927 and 8,743,973 (thousand dongs) individually. On average, the leverage ratio is 0.0802, and the mean value for the ratio of cash in total assets is 0.0000973. The average age of firms is 16, ranging from 2 to 61 years.

Table 2. Descriptive statistics

Variables	Mean	Standard Deviation	Median	Min	Max
CSR	0.1909	0.2216	0.0952	0	0.9524
ERP	0.2407	0.3229	0.0833	0	1
ENV	0.1248	0.1595	0.1111	0	1
FAM	0.6667	0.4715	1	0	1
FEMANAGER	0.3809	0.3824	0.5	0	1
ESIZE	16.0793	42.0798	5	1	1700
WOSHARE	0.3801	0.2594	0.375	0	1
FSIZE	5827566	2.46.10^7	1,464,550	4500	9.21.10^8
ROA	0.2927	0.7203	0.1263	−2.1896	15.7204
REV	8743973	1.38.10^8	900,725	5000	8.27.10^9
LEV	0.0802	0.1984	0	0	4.3770
CASH	0.0000973	0.0001178	0.000056	2.08.10^−7	0.0009569
FAGE	16.0689	9.7796	14	2	61

Table 3 illustrates the correlation between family control (FAM), female manager rate in the TMT (FEMANAGER) and some control variables. For instance, family firms (FAM) and total labour force (ESIZE), firm size by total asset (FSIZE), and revenue (REV) were negatively correlated, with coefficient = −0.6619; coefficient = −0.5523; and coefficient = −0.6379 (p-value < 0.01) respectively. There is also a high level of correlation between female proportion in TMT (FEMANAGER) and female proportion in the total labour force (WOSHARE), at coefficient = 0.4483 (p-value < 0.01). These correlations raise potential multicollinearity issues but variance inflation factor (VIF) tests indicate that the problem of multicollinearity can be excluded because VIF of independent variables (FAM and FEMANAGER) are 1.94 and 1.26, respectively (below 2).

Table 3. Correlation matrix

Variable	FAM	FEMANAGER	ESIZE	WOSHARE	FSIZE	ROA	REV	LEV	CASH	FAGE	VIF
FAM	1										1.94
FEMANAGER	−0.0615***	1									1.26
ESIZE	−0.6619***	0.0718***	1								4.00
WOSHARE	−0.0295**	0.4483***	0.0851***	1							1.31
FSIZE	−0.5523***	0.0364***	0.7092***	−0.0050	1						4.26
ROA	0.0552***	0.0285**	−0.0586***	0.0531***	−0.3019***	1					1.38
REV	−0.6379***	0.0403***	0.8370***	−0.0190	0.7535***	0.0434***	1				5.36
LEV	−0.2030***	0.0015	0.2193***	−0.0108	0.0879***	0.0493***	0.2312***	1			1.09
CASH	−0.0471***	−0.0151	0.0104	−0.0333**	−0.3056***	0.1994***	0.0261*	0.0950***	1		1.37
FAGE	0.2357***	0.0079	−0.1731***	0.0743***	−0.1551***	−0.0020	−0.2035***	−0.0944***	−0.0559***	1	1.08

*p < .1; **p < .05; ***p < .01. Standard error is showed in parentheses

Table 4 shows the empirical testing results for hypothesis 1. This table indicates that family firm (FAM) is significantly and negatively associated with CSR performance at the coefficient of −0.7708 (p < 0.01), which also means family ownership decreases the level of SMEs’ CSR performance. Thus, hypothesis 1 is supported.

Table 4. Family control and CSR performance

Variables	CSR	ERP	ENV
FAM	−0.7708*** (0.0494)	−1.0427*** (0.0664)	−0.2694*** (0.0532)
FEMANAGER	0.0627 (0.0553)	0.1328* (0.0764)	−0.0589 (0.0587)
ESIZE	0.1781*** (0.0296)	0.2802*** (0.0424)	0.0499 (0.0321)
WOSHARE	−0.0186 (0.0880)	−0.0553 (0.1223)	0.0289 (0.0986)
FSIZE	0.1617*** (0.0211)	0.1791*** (0.0299)	0.1511*** (0.0224)
ROA	0.0361* (0.0219)	0.0499 (0.0333)	0.0242 (0.0195)
REV	0.0093 (0.0255)	0.0063 (0.0357)	0.0227 (0.0288)
LEV	−0.0596 (0.0899)	−0.0780 (0.1188)	−0.0178 (0.0971)
CASH	0.0413** (0.0166)	0.0524** (0.0236)	0.0267 (0.0173)
FAGE	−0.0058*** (0.0023)	−0.0046 (0.0032)	−0.0080*** (0.0023)
Industry controlled	Yes
Year controlled	Yes
Constant	−3.6718*** (0.3621)	−3.8470 (0.4841)	−3.8582*** (0.4105)
Pseudo R2	0.1305	0.2134	0.0427

*p < .1; **p < .05; ***p < .01. Standard error is showed in parentheses

To test CSR performance, we run separately regressions for nonfamily firms and family firms. From Table 5 with nonfamily SMEs, the result shows that the proportion of females in the TMT (FEMANAGER) is positively related to CSR performance (coefficient: 0.1589; p-value < 0.05). This result supports for the hypothesis 2 that the relationship between the presence of female managers in the TMT and CSR performance is positive. However, in term of family SMEs, the analysis in Table 6 shows no correlations between FEMANAGER variable and CSR performance. This means that the positive relationship between female involvement in the TMT and CSR performance is only occurred in nonfamily SMEs and not in family SMEs. Thus, the hypothesis 3 is supported: the positive relationship between female presence in the TMT and CSR performance is weaker in family SMEs than in nonfamily ones.

Table 5. Female manager and CSR performance (Nonfamily SMEs)

Variables	CSR	ERP	ENV
FEMANAGER	0.1589** (0.0773)	0.2412** (0.1169)	0.0616 (0.0918)
ESIZE	0.1601*** (0.0327)	0.2644*** (0.0535)	0.0624 (0.0396)
WOSHARE	−0.0033 (0.1145)	−0.0471 (0.1769)	0.0372 (0.1354)
FSIZE	0.0799*** (0.0262)	0.1114*** (0.0417)	0.0680** (0.0303)
ROA	0.0213 (0.0187)	0.0485 (0.0384)	−0.0013 (0.0215)
REV	0.0538* (0.0281)	0.0629 (0.0452)	0.0629* (0.0359)
LEV	−0.0504 (0.0688)	−0.088 (0.1014)	−0.0108 (0.0876)
CASH	0.0204 (0.0192)	0.0357 (0.0298)	0.0008 (0.0228)
FAGE	−0.0039 (0.0028)	−0.0063 (0.0044)	−0.0006 (0.0032)
Industry controlled	Yes
Year controlled	Yes
Constant	−3.1267*** (0.3527)	−3.6056*** (0.5739)	−3.3859*** (0.4494)
Pseudo R2	0.0307	0.0611	0.0205

*p < .1; **p < .05; ***p < .01. Standard error is showed in parentheses

Table 6. Female manager and CSR performance (Family SMEs)

Variables	CSR	ERP	ENV
FEMANAGER	−0.0157 (0.0625)	0.0456 (0.0860)	−0.0894 (0.0769)
ESIZE	0.1391*** (0.0450)	0.2243*** (0.0585)	0.0399 (0.0546)
WOSHARE	−0.2344** (0.0981)	−0.3514*** (0.1281)	−0.1187 (0.1326)
FSIZE	0.2193*** (0.0279)	0.2712*** (0.0381)	0.1644*** (0.0339)
ROA	0.0592** (0.0285)	0.0548 (0.0352)	0.0619* (0.035)
REV	0.0522 (0.0355)	0.0657 (0.0472)	0.0410 (0.0450)
LEV	−0.3266 (0.2498)	−0.1806 (0.2808)	−0.4823 (0.3389)
CASH	0.0455** (0.0219)	0.0543* (0.0304)	0.0355 (0.0263)
FAGE	−0.0054** (0.0027)	−0.0024 (0.0035)	−0.0087*** (0.0032)
Industry controlled	Yes
Year controlled	Yes
Constant	−5.9263*** (0.4135)	−7.1926*** (0.5380)	−4.5874*** (0.5556)
Pseudo R2	0.0409	0.0684	0.0249

*p < .1; **p < .05; ***p < .01. Standard error is showed in parentheses

The results support the first hypothesis that family control in SMEs is negatively associated with CSR performance. We also find that the increasing presence of female managers in the TMT positively impacts CSR engagement with nonfamily SMEs but not family SMEs. However, CSR includes multi-dimension and we should concern the multidimensionality of CSR instead of regarding CSR as a single aggregated measure. Thus, to increase the robustness of results, this research divides CSR performance into two dimensions: the employees and the environment dimension, relying on the information that CSR performance is direct.

First, we examine the relationship between family control and CSR dimensions: employee and environment. Similarly, the result in Table 5 shows that family control is associated negatively with both employees and the environment dimension of CSR. To be more specific, family control correlate negatively with employee dimension at coefficient: −1.0427 (p-value < 0.01) and environment dimension at coefficient: −0.2694 (p-value < 0.01). Hence, again, hypothesis 1 is supported, family ownership decreases the level of SMEs’ CSR performance.

Second, we examine the relationship between the presence of female managers in the TMT and CSR dimensions: employee and environment. The result in Table 6 shows that female manager participant is associated positively with the employee dimension of CSR (coefficient: 0.2412 and p-value < 0.01) but insignificantly with environment dimension in nonfamily SMEs. However, similar to the previous results, there is no significant relationship between female presence in the TMT and the dimensions of CSR performance in family SMEs as shown in table 9. To be more specific, the rate of female participant in the TMT correlate positively with employee dimension in nonfamily SMEs at coefficient: 0.2412 (p-value < 0.01).

3.3. Reverse causality concern

The reverse causality effect (one of the cause of endogeneity) can be from the impact of CSR performance on the likelihood of female presence on the TMT. To check this possibility, FEMANAGER variable is regressed on CSR performance dimensions (we do in both family and nonfamily SMEs). The results show that CSR performance in 2011 and 2013 does not correlate with FEMANAGER variable in 2013 and 2015, respectively, (p-value > 0.1). Therefore, there is no proof of reverse causality between CSR performance and the rate of female managers on the TMT, and reduce the risk of causality impact.

4. Discussion

This research has two main goals. The first objective is to investigate the influence of family control on the link between SME and CSR performance. The second is to investigate the nexus between female engagement in the TMT and SMEs’ CSR performance as well as the moderate role of family control in the circumstance of an emerging country like Vietnam. To test the proposed hypotheses, this paper uses the strongly balance sample from Viet Nam SMEs database, undertaken by UNU—WIDER, including 5160 firm-year observations for three years: 2011, 2013, and 2015. Our results are both supported and rejected with the hypothesis.

First, our findings demonstrate the negative relationship between family control and CSR performance. The reasons may include the fact that the family enterprises have an interest in preserving families’ SEW rather than a social or environmental focus (Morck & Yeung, 2004). This is because family altruism, coming from the “emotional attachment” dimension of SEW, can be the cause of unequal and unethical treatments between family and nonfamily members in family firms (Cruz et al., 2014). In family-owned businesses when inequity exists between family and nonfamily members, the employee responsibility pillar of CSR may therefore perform less well. Kellermanns et al. (2012) refer to this circumstance as the “dark side” of SEW, which SEW preservation does not correspond with CSR practices. Moreover, Cruz et al. (2014) argue that family firms are more inclined than nonfamily enterprises to cease social and environmental related activities when the firm’s financial situation is threatened. Compared with large family businesses, family SMEs face more resource shortage problems such as a lack of professional management skills, technology, and infrastructure capabilities, which might jeopardise their financial stability (Price et al., 2013). Thus, it is likely that family SMEs involve in unethical behavior relating to social and environmental issues to avoid expenses that increase the financial burden. Moreover, in the context of Vietnam, SMEs’ owners and managers lack knowledge and skills about how to apply CSR procedures appropriately (M. M. Nguyen et al., 2018; Vo & Arato, 2019).

Second, with the sample of nonfamily SMEs, the results find a positive relationship between the increasing participation of female managers in the TMT and the ERP dimension of CSR performance. This is due to the fact that the feminine leadership style is more communal, democratic, and participatory than the masculine leadership style (Boulouta, 2013; Eagly & Carli, 2003; Nielsen & Huse, 2010). This leadership style stems from their heightened sensitivity, empathy, helpfulness, and kindness compared to males. They are also more inclined to form relationships, react to the needs of others, and act properly toward society (Setó-Pamies, 2015). Therefore, firms, managed by female managers, are more likely to positively impact employees’ satisfaction because they are more democratic, interpersonally-oriented and experts in human resource management (Eagly & Carli, 2003). However, results also reveal that the rising representation of female managers in the TMT corresponds favourably with the employee component of CSR performance but not with the environmental dimension. It may be a trade-off between internal and external CSR performance since limited resources prevent SMEs from investing on both aspects (Hawn & Ioannou, 2016; Jin & Jiang, 2021). Thus, they tend to concentrate more on their internal stakeholders rather than external stakeholders (Aras-Beger & Taşkın, 2020; Baumann-Pauly et al., 2013). Furthermore, in the context of Vietnam, environmental protection procedures that fulfil national standard standards continue to be inadequate, particularly among SMEs. They are unable to afford manufacturing technology that improves the utilisation of resources and lower emissions (N. H. N. H. Nguyen et al., 2014). Thus, SMEs tend not to acquire cleaner manufacturing and more eco-friendly technology.

Third, the results show that in family-owned SMEs, the favourable impact of female managers on CSR performance is not seen, supporting the proposed hypothesis. As mentioned above, there are several reasons. Due to SEW and control preservation in family firms, managers’ influence on the business is diminished because of the dominant influence of the family in decision-making. As a consequence, in the context of family enterprises, the managerial independence of female top executives might be diminished. Since the influence of top managers in family enterprises is not as significant as in nonfamily firms, the beneficial effect of female leaders on the CSR performance of family firms may not be evident. This conclusion is further reinforced by the suggestion of (Chadwick & Dawson, 2018) that the application of the upper echelon theory is weakened in organisations where the managerial independence of executives is not ensured. Second, the unfair treatment toward women occurs in family business because female members do not receive rewards and recognition as deserved as their contribution (S. Chen et al., 2018). In reality, female successors are quite uncommon in family enterprises, while male family members are prioritised as prospective successors (S. Chen et al., 2018). Furthermore, culture should be considered, since many Asian organisations are unwilling to allow women to hold positions of leadership (Gupta & Levenburg, 2010). Vietnam is a nation where Confucian philosophy influences every element of society’s perspective, and Vietnamese culture generally perceives women’s duty in families to be mostly limited to housekeeping (Pham & Hoang, 2019). Similar circumstances exist in the workplace, where women face greater obstacles than males to attain executive positions in organisations. (Tu, 2017). Consequently, the family might underestimate the contribution of female managers in family businesses, which has a detrimental influence on the assistance of female managers to boosting CSR performance levels.

5. Conclusion

This paper contributes knowledge to the literature. First, this research expands our understanding of the “dark” side of SEW and demonstrates that preserving socioemotional wealth does not necessarily equate to positive CSR performance. Cruz et al. (2014) propose that SEW preservation is a cause of both socially ethical and immoral behaviour in enterprises, referring to SEW preservation as a “double-edged sword”. Family SMEs suffer a lack of resources, as well as a lack of technology and managerial expertise (Price et al., 2013). Consequently, with family-owned SMEs, CSR performance may not be a top priority. Second, the level of managerial independence of female managers can be reduced in these firms due to the family’s domination in management and control. This support for the study of Chadwick and Dawson (2018) which suggests that the upper echelon theory may not be applied in family firms, similar to the way applied in nonfamily firms.

Our results also generate some practical implications. First, the relationship between family-controlled firms and CSR is not uniform, whether positive, negative or insignificant. Therefore, we should not necessarily see family businesses as more socially responsible than other organisations. Other aspects, such as the size and financial status of the company, should also be examined. Second, although the role of female leaders in promoting social responsible behaviour is still proved, they should be supported in a workplace that respects their importance. Thus, female managers should be empowered and given greater autonomous authority in family enterprises; this would increase their advantages in promoting CSR participation.

Although the study contributes to the understanding of CSR performance in SMEs and the effect of female managers in the TMT and CSR performance in both family and nonfamily SMEs, it still has some limitations. First, this study bases on a single database of Vietnamese SMEs and lacks generalization from other countries. Thus, we suggest future research should take similar studies in other countries due to cultural differences and regulation. Second, the CSR concept is multidimensional and the survey questions may not include all aspects relating to CSR performance, which can impact negatively to the result. Thus, it should also be pointed out for future studies to take more detailed survey questions to cover the wide range of CSR activities.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors received no direct funding for this research.

Appendices

Table A1. Summary of industries

Sector	N	Percentage	Sector	N	Percentage
Food production	1551	30.06	Automobiles and other motor vehicles	30	0.58
Prefabricated made products	910	17.64	Drugs, pharmaceutical chemicals and medicinal herbs	25	0.48
Products from wood, bamboo	545	10.56	Coke and refined petroleum products	20	0.39
Manufacture of beds, cabinets, tables and chairs	372	7.21	Repair, maintenance and installation of machinery and equipment	16	0.31
Products from rubber and plastic	274	5.31	Other means of transport	12	0.23
Production of costumes	231	4.48	Wholesale	10	0.19
Other non-metallic mineral products	216	4.19	Electronic products, computers and optical products	10	0.19
Weaving	205	3.97	Retail	6	0.12
Print, copy records	122	2.36	Other personal service activities.	5	0.1
Paper and paper made products	119	2.31	Administrative activities	3	0.06
Beverage	112	2.17	Production and distribution of electricity, gas, hot water, steam and air conditioning	2	0.04
Leather and related products	100	1.94	Specialised construction activities	1	0.02
Electrical equipment	69	1.34	Sale and repair of cars, motorcycles, motorbikes and other motor vehicles	1	0.02
Metal production	59	1.14	Tobacco products	1	0.02
Chemicals and chemical products	53	1.03	Lodging	1	0.02
Machinery and equipment not elsewhere classified	40	0.78	Food Service	1	0.02
Other processing and manufacturing industries	37	0.72	Rental Property	1	0.02

Table A2. CSR questionnaires used in the survey

CSR performance dimensions	Questions
Employee dimension (ERP)	Is there any difference in wages between employees who work in hazardous conditions and those who do not (i.e. a premium paid for working in hazardous conditions)? (Yes/No)
	Are employees paid overtime for working outside standard hours? (Yes/No)
	Did you pay contribution to social insurance for your employees? (Yes/No)
	Did you pay contribution to health insurance for your employees? (Yes/No)
	Did you pay contribution to unemployment insurance for your employees? (Yes/No)
	Do you normally compensate your workforce directly for accidents or professional illness? (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Sick leave with pay) (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Right to paid maternity leave) (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Right to unpaid maternity leave) (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Annual leave with pay) (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Any payment (lump-sum) when worker retires) (Yes/No)
	Do the employees enjoy any of the following benefits (by regulation, directly or from the government)? (Survival Benefit (family)) (Yes/No)
Environment dimension (ENV)	Which of the following environmental factors does your firm treat (Air quality)? (Yes/No)
	Which of the following environmental factors does your firm treat (Fire)? (Yes/No)
	Which of the following environmental factors does your firm treat (Heat)? (Yes/No)
	Which of the following environmental factors does your firm treat (Lighting) (Yes/No)
	Which of the following environmental factors does your firm treat (Noise) (Yes/No)
	Which of the following environmental factors does your firm treat (Waste disposal) (Yes/No)
	Which of the following environmental factors does your firm treat (Water pollution)
	Which of the following environmental factors does your firm treat (oil degradation/pollution) (Yes/No)
	Which of the following environmental factors does your firm treat (Other) (Yes/No)