Blood is thicker than water: an analysis of women’s presence on Pakistani boards

Abstract

Purpose

This study aims to provide insights into gender diversity in Pakistani boardrooms, particularly for the dominant family business type, which is strongly guided by (non-financial) family-related objectives when making business decisions, such as the appointment of board members. Pakistani companies operate within the framework of weak legal institutions and a traditionally highly patriarchal environment. This study examines how corporate decisions regarding the appointment of female board members play out in this socio-political and cultural environment.

Design/methodology/approach

Board composition and board characteristics were examined using hand-collected data from 213 listed family firms and non-family firms on the Pakistan Stock Exchange from 2003 to 2017. Univariate analyses, probit regressions and robustness tests were performed.

Findings

Pakistani family firms have a significantly higher proportion of women on their boards than do non-family firms. They are also significantly more likely to appoint women to top positions, such as CEO or chairs.

Practical implications

Evidently, women are allowed to enter boards through family affiliations. Gender quotas appear an ineffective instrument for breaking through the “glass ceiling” in this socio-cultural environment. Thus, gender parity must entail the comprehensive promotion of women and the enforcement of legal reforms for structural and cultural change.

Originality/value

The analysis focuses on a Muslim-majority emerging Asian market that has been scarcely researched, thus offering new perspectives and insights into board composition and corporate governance that go beyond the well-studied Western countries.

Citation

Wagner, E., Pernsteiner, H. and Riaz, A. (2024), "Blood is thicker than water: an analysis of women’s presence on Pakistani boards", Gender in Management, Vol. 39 No. 4, pp. 516-533. https://doi.org/10.1108/GM-04-2022-0139

1. Introduction

Research shows that “good corporate governance” pays off for companies, markets and countries (La Porta et al., 2002; Claessens and Yurtoglu, 2013). In this context, the board of directors serves as the central mechanism to ensure that management is controlled and monitored (Fama and Jensen, 1983; Carter et al., 2003).

Increasing the number of women in corporate boardrooms is a governance issue of high priority (Zukis, 2020). Several countries have introduced mandatory quotas for women on boards of publicly traded companies (for a current overview, see Catalyst, 2020). The Chartered Governance Institute, London, writes in its latest report:

It is also clear that board gender diversity is now recognised as not just an issue of equality between men and women but also as a contribution to more effective boards, corporate performance and to maintaining a social licence to operate (Hawarden and Greenwood, 2021, p. iv).

The report concludes, “Cultural factors, for example, traditional patriarchal attitudes, are a barrier in some countries but are not analysed in great detail” (Hawarden and Greenwood, 2021, p. 55). We attempt to fill this gap and discuss cultural and institutional factors, barriers and facilitating factors in Pakistan, an Asian Muslim-dominated country that has thus far received very little research attention.

Unlike other Asian countries [see Saeed et al. (2017) for India and China or Low et al. (2015) for several Asian countries excluding Pakistan], gender diversity in Pakistani listed companies has rarely been studied, despite Pakistan being the fifth most populous country in the world and an emerging market with a 10-year average annual GDP growth of 3.6% (World Economic Forum, 2019); it thus offers an interesting setting for several reasons. Firstly, Pakistan’s gender equality ranking remains one of the lowest in the world (UN Women, 2021). According to the Global Gender Gap Report 2022, Pakistan ranks second to last in terms of the gender parity index (World Economic Forum, 2022), and gender inequality plays a role in the appointment of women to boards (Terjesen et al., 2015; Grosvold et al., 2016; Iannotta et al., 2016).

Secondly, Pakistan is characterised by weak investor protections, judicial inefficiencies and weak enforcement mechanisms (Leuz et al., 2003). Pakistan’s legal system is based on English common law, in which companies have a one-tier board structure. The Securities and Exchange Commission of Pakistan (SECP) introduced its first corporate governance code in 2002. We chose the study period 2003–2017 because there were no mandatory quotas for women during this period, and thus a comparable regulatory framework. With the Companies Act 2017, the SECP introduced a mandatory gender quota requiring companies to appoint at least one woman to the board (Securities and Exchange Commission of Pakistan, SECP, 2017). The introduction of such regulations can be seen as an “exogenous political shock” that abruptly increases the representation of women [see Wang and Kelan (2013) for Norway or Sarkar and Selarka (2021) for India]. These two preconditions – high gender inequality and no corresponding mandatory corporate governance rules in a weak governance environment – led us to expect the proportion of women on boards in Pakistan to be low.

Thirdly, as is common in emerging markets, concentrated ownership is prevalent in Pakistan, where family firms (FFs) dominate (Javid and Iqbal, 2008), and FFs have distinct governance and board structures (Bartholomeusz and Tanewski, 2006; Rubino et al., 2017). The concept of socio-emotional wealth (SEW), which has recently gained popularity in family business research, predicts that exercising control and perpetuating family succession play a central role in FFs (Gómez-Mejia et al., 2007; Berrone et al., 2012). To protect SEW, FFs consider it essential to have family members on their boards, which presumably includes sending female family members to protect SEW and making strategic decisions on behalf of the family (Herdhayinta et al., 2021), suggesting higher female representation on the boards of Pakistani FFs. Alternatively, as Saeed et al. (2017) suggest, for China and India, there may be a low representation of women on FF boards, reflecting patriarchal family structures and discrimination against women (only male family members or relatives are appointed to boards).

Therefore, we examine the following research questions: (1) What is the representation of women on (family) business boards in Pakistan? (2) Do family-related objectives predominate; that is, is “family stronger” and are women comparatively more represented on boards in this type of business?

As these research questions can only be answered empirically, we draw on a large and unique (hand-collected) data set that includes 213 listed Pakistani companies from 2003 to 2017 (3,062 firm-year observations).

2. Literature review and hypotheses

2.1 The business case for diversity

Board diversity is heterogeneous among board members and involves various categories (see, for example, Brammer et al., 2007). However, this study focused specifically on gender diversity. Women on boards and mandatory women’s quotas are currently the subject of intense political and social debate, and many corporate governance codes around the world now include recommendations or mandatory women’s quotas (see, for example, Terjesen et al., 2015; Catalyst, 2020).

The business literature draws on various theories to demonstrate the benefits of gender-diverse boards on corporate financial performance. Agency theory is a particularly prominent theory in which the board of directors is a key governance mechanism, as directors are responsible for monitoring and controlling the CEO (Fama and Jensen, 1983) and should ultimately ensure value creation for their shareholders (Adams, 2017). The participation of women as directors on boards can result in improved monitoring and reduced agency problems (agency costs) for several reasons (e.g. greater independence, Adams and Ferreira, 2009), thereby increasing shareholder value (Adams and Ferreira, 2009; Carter et al., 2003; Adams and Ferreira, 2009) found evidence in line with this theory. In their study, gender-diverse boards were associated with fewer agency problems, particularly independence, improved monitoring activities, higher meeting attendance and higher stock-based compensation. Furthermore, it is argued – primarily with reference to the resource dependency theory (Pfeffer and Salancik, 1978) – that female directors bring unique skills and valuable resources to the board, cultivate a different management style, contribute to a greater diversity of perspectives and improve the quality of board decisions. Women are also considered more creative and innovative in terms of problem-solving (Carter et al., 2003; Hillman et al., 2007). Moreover, gender-diverse boards have a broader understanding of the marketplace and firms’ multiple stakeholders (Carter et al., 2003; Post and Byron, 2015). Overall, the resource dependence perspective (and related theories, see Post and Byron, 2015) demonstrates how directors can bring important resources to companies and that gender-diverse boards contribute positively to the company’s strategic positioning.

In addition to theoretical approaches, there is extensive empirical evidence [see the two meta-analyses by Post and Byron (2015) and Hoobler et al. (2018)] on the “business case”, that is, whether having women on boards pays off financially. Taken together, the results of the numerous studies, most of which were conducted for Western countries, are inconsistent and do not support the business case argument that putting women on boards significantly enhances performance (Carter et al., 2003; Post and Byron, 2015). Instead, they show that the relationship between board gender diversity and financial performance is complex (Adams and Ferreira, 2009; Post and Byron, 2015). Despite contradictory findings, the presence of women on corporate boards is increasingly viewed as good corporate governance practice and is associated with firm-value-enhancing attributes, such as a more beneficial board structure and higher board effectiveness (Adams and Ferreira, 2009; Sarkar and Selarka, 2021). Unsurprisingly, studies of emerging markets are inconclusive. While Kılıç and Kuzey (2016) find a positive correlation between board gender diversity and financial performance in Turkey, Darmadi (2011) shows a negative relationship for Indonesian companies and Abdullah et al. (2016) show no clear impact of gender diversity on firm value for Malaysia. Low et al. (2015) examine women on boards in Asian countries (excluding Pakistan) and argue that gender-diversified boards along Anglo-American/Western lines may not be promising for Asian companies as board composition should ultimately reflect socio-cultural values. Female nominations may be merely tokens that satisfy investors or meet regulatory requirements. It is evident from these studies that outcomes are highly dependent on country-specific circumstances and cultural contexts (Low et al., 2015; Post and Byron, 2015). Emerging markets differ considerably from developed countries in their institutional and legal environments and ownership structures (Saeed et al., 2016). The proportion of women on boards, and thus their potential impact on performance, varies widely by region and country.

2.2 Women’s presence on boards and institutional (socio-cultural) contexts

Internationally, women only appear in boardrooms to a small extent. Various reasons for this include the existence of “old boy networks” (Low et al., 2015), which makes accessing top positions more difficult for women, and the “double burden”, referring to the dual burden that arises when women work and also have to manage the household (Low et al., 2015). The presence of women on boards is less common in Asian countries than in Western countries (Low et al., 2015; Deloitte, 2019). A report by McKinsey and Company on women’s equality in the Asia Pacific region shows that gender inequality is high, particularly in Pakistan (Süssmuth-Dyckerhoff et al., 2012). Unequal educational levels between men and women are a prevalent issue in Pakistan. Moreover, women accounted for only 22% of the labour force and contributed 11% to the country’s GDP, placing Pakistan last in the Asia Pacific region in the study (Woetzel et al., 2018), and second last in the World Economic Forum’s gender parity index (World Economic Forum, 2022).

It is widely acknowledged that national cultures have an important role in shaping gender roles in society (Iannota et al., 2016). Traditional socio-cultural norms in which women are expected to take responsibility for the household and assume the “caretaker role” (Benson and Yukongdi, 2005; Chizema et al., 2015) are likely to be even more pronounced for women in Asia (Low et al., 2015). In addition to the distinctly patriarchal nature of many societies, also referred to as “secular patriarchy” (Low et al., 2015), another reason given for low female representation in Muslim countries such as Pakistan is religion.

Religion plays an essential role in how society views and attitudes towards women in leadership positions (Chizema et al., 2015). The level of gender equality across countries is related to religiosity (Inglehart and Norris, 2003; Chizema et al., 2015; Chizema et al., 2015) examine the issue of women on boards in an international context and demonstrate that the level of religiosity has a negative effect. Islam is the state religion in Pakistan, and more than 96% of the country’s population is Muslim (Pakistan Bureau of Statistics, 2021). In this context, it should be emphasised that not only is Islam the central religion in Pakistan but also a patriarchal interpretation of Islamic guidelines prevails, determining women’s place in the community and their socio-cultural role, ultimately giving men control over their affairs, both in domestic and social life. For career advancement, women in Pakistan require the support and endorsement of their male family members (Sarwar and Imran, 2019). In this socio-cultural setting, especially owing to the very low participation of women in the labour market and pronounced gender inequity (Terjesen et al., 2015; World Economic Forum, 2022), as well as the religious setting (Chizema et al., 2015; Low et al., 2015), Pakistan can be assumed to have a very low representation of women on boards compared with the boards of other Asian or international firms. Consequently, we formulate the following hypothesis:

H1.

Gender diversity on boards (as measured by the proportion of women) in Pakistani companies is (comparatively) low.

2.3 Board composition in family firms

Institutional factors, including a country’s corporate governance system, play a significant role in women’s presence on boards (Terjesen et al., 2009). A country’s corporate governance system is largely determined by its overall development and institutional environment and, specifically, by its prevailing ownership structure (Claessens and Yurtoglu, 2013). In the context of Pakistani firms, it is important to consider the significant ownership concentration, particularly the dominance of family-controlled firms (La Porta et al., 1999; Javid and Iqbal, 2008; Claessens and Yurtoglu, 2013), as the governance and board structures of FFs differ significantly from those of non-family firms (NFFs; Anderson and Reeb, 2003; Bartholomeusz and Tanewski, 2006; Rubino et al., 2017). From an agency theory perspective, special features arise from family ownership and control. Jensen and Meckling (1976), Fama and Jensen (1983) and Anderson and Reeb (2003) argue that the agency costs of equity are rather low owing to less separation between ownership and control in FFs, and that monitoring improves because of family control. However, family control can create agency problems between the controlling and minority shareholders because the family can expropriate minority shareholders (or other stakeholders) to pursue private and family benefits (Anderson and Reeb, 2003). Therefore, agency problems in FFs tend to shift away from manager–shareholder conflicts and towards minority versus controlling shareholder conflicts (Villalonga and Amit, 2006; Claessens and Yurtoglu, 2013). Board monitoring is particularly important in mitigating agency problems between minority shareholders and corporate insiders (Fama and Jensen, 1983; Chen and Nowland, 2010). One assumption is that stricter monitoring is necessary to protect minority shareholders’ interests and prevent families from maximising their private benefits (Chen and Nowland, 2010). From the agency theory perspective, board diversity can contribute positively to a firm’s success because the presence of women on boards is associated with improved monitoring (Carter et al., 2003; Adams and Ferreira, 2009; Abdullah et al., 2016) argue that this positive effect is even more pronounced in emerging markets where weak corporate governance systems prevail. However, the effect of board monitoring by women may be less positive in family businesses because families have the ability and power to nominate directors, meaning that directors cannot be considered independent monitors (Chen and Nowland, 2010; Herdhayinta et al., 2021). Family-affiliated women may be present for reasons other than the monitoring or advising roles of professional directors, such as meeting family quotas or nepotism (González et al., 2020).

In this context, the SEW concept (Gómez-Mejia et al., 2007), which is well established in the family business literature, is useful for explaining the appointment of leadership positions. Gómez-Mejia et al. (2007) emphasised that family business owners do not exclusively pursue financial goals, as their decisions are also influenced by their SEW. SEW 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-Mejia et al., 2007, p. 106; Berrone et al., 2012) stress that SEW is a multidimensional concept that includes family control and influence, identification of family members with the firm, binding of social ties, emotional attachment among family members and renewal of family bonds to the firm through dynastic succession. Indeed, the desire to maintain family control over a business is a central factor influencing behaviour in family businesses (Gómez-Mejia et al., 2007; Chrisman et al., 2012; Dick et al., 2021); in some cases, the (intergenerational) control retention motive is even used as a criterion for defining a family business (see, for example, Dick et al., 2021). The important dimension of “family control and influence”, emphasised in the concept of SEW (Gómez-Mejia et al., 2007), is also presumably reflected in board member appointments. It can be assumed that female family members are sent to the board to act in the family’s interests and pursue the company’s family-related and non-financial goals (Miller and Le Breton-Miller, 2014; Herdhayinta et al., 2021). A board of non-family (outside) members may be pressured to follow their economic interests (Molly et al., 2019). By contrast, when the board is predominantly composed of family members, it is easier to assert family interests, enforce family-centred goals and maximise SEW (Gómez-Mejia et al., 2007; Molly et al., 2019).

Therefore, from the SEW perspective, family ties and control considerations are expected to dominate board nominations in FFs and play a greater role in decisions than reservations regarding women serving in leadership positions. In some circumstances, these SEW-driven considerations may lead to decisions in FFs that are not in line with shareholder value but serve important non-financial objectives, namely, maintaining family control (Chrisman et al., 2012; Herdhayinta et al., 2021). There is evidence for Western Europe (Campbell and Mínguez-Vera, 2008; Bianco et al., 2015) and emerging market firms (Saeed et al., 2016) that women are appointed to boards of FFs because of family ties. In contrast, Saeed et al. (2017) find that family ownership has a negative impact on the proportion of female directors in China and India, and claim that family businesses in these countries differ from FFs internationally because they generally follow traditional patriarchal and hierarchical social frameworks. Herdhayinta et al. (2021) and Sarkar and Selarka (2021) find evidence that non-family female directors are more effective than family female directors in monitoring management and protecting the interests of minority shareholders in Indonesia and India. Their analyses show that female family directors in Indonesia and India are likely to prioritise family and non-financial goals.

Based on empirical evidence pointing predominantly in this direction and with reference to SEW considerations and specific goals, such as maintaining control, we expect to find greater gender diversity on FF boards of FFs. This should also be evident in a Muslim-majority country, which is characterised by a patriarchal culture and pronounced gender inequality. We hypothesise that the goal of protecting family interests in Pakistani companies dominates the decision to appoint board members by placing more women in leadership positions. Thus, we propose the following hypothesis:

H2.

Pakistani FFs are more likely to appoint women to boards and have greater gender diversity (as measured by the proportion of women on boards) than NFFs.

3. Empirical results

3.1 Sample and variables

We empirically investigated various board characteristics of listed Pakistani companies. In line with the existing literature (see the meta-analyses by Post and Byron, 2015; Hoobler et al., 2018), we extract important firm-specific data in addition to various board data. Our data set includes all companies listed on the Pakistan Stock Exchange (PSE) over the period of 2003 to 2017, except regulated companies in the financial and utility industries and state-owned firms. We chose the period up to 2017 because a mandatory women’s quota was introduced thereafter with the Companies Act, 2017, and such gender laws are found in the literature to represent an exogenous shock that strongly affects nomination decisions (Sarkar and Selarka, 2021; Bhattacharya et al., 2022). After excluding delisted, suspended and missing firms, we have the largest possible data window for the aforementioned period (unbalanced panel) with 213 listed companies and 3,062 firm-year observations.

Consistent with prior research (Anderson and Reeb, 2003; Villalonga and Amit, 2006), FFs are defined as firms in which 20% of the firm’s shares are held (either directly or indirectly) by a family and at least two members of the controlling family hold managerial positions, such as board members, CEO or chairmen. Based on this definition, approximately 71% of the firms in our data set are FFs (total firm-year observations: 3,062, with 2,182 firm-year observations for FFs and 880 firm-year observations for NFFs). The data on ownership and governance commonly used in this context (see, for example, Bianco et al., 2015; Saeed et al., 2016) are hand-collected, and various sources are used for this purpose. Ownership and boardroom data were manually collected from the shareholding patterns provided in the annual reports and websites of the PSE and firms. Financial data are obtained from annual reports and the DataStream WorldScope database.

To test our hypotheses, we use several control variables commonly used in the literature on board characteristics and the decision to appoint women to boards (Anderson and Reeb, 2004; Adams and Ferreira, 2009; Saeed et al., 2016). For board characteristics, we hand-collected information specific to women, namely, the proportion of women (Female Directors), the presence of at least one woman on the board (Female Dummy) and whether a woman holds the CEO position (Female CEO) or the chair position on the board (Female Chair). In addition to Board Size (number of directors), we extracted the percentages of independent and executive directors. For firm-specific variables, we use Firm Size (measured by the natural logarithm of total assets), which, according to the existing literature, is associated with gender diversity in boardrooms (Adams and Ferreira, 2009; Saeed et al., 2016). Firm Age is a common variable used in the analysis of boards in studies of FFs (Anderson and Reeb, 2004; Morikawa, 2016). Following Adams and Ferreira (2009), we use a market-based measure of performance, a proxy for Tobin’s q, and a financial statement-based measure, the return on assets (ROA). Our proxy for Tobin’s q is the ratio of a company’s market-to-book value (Market-to-book). Firm risk, which may correlate with the presence of women on boards (Hillman et al., 2007; Saeed et al., 2016), is measured by volatility (the standard deviation of daily stock returns over the past 12 months; Risk). For more details on the study variables, see Table 1 below.

3.2 Descriptive statistics and univariate tests

Table 2 presents the descriptive statistics for all board and firm characteristics, including the means, standard deviations and difference-of-means tests (t-tests) for FFs and NFFs.

We begin with a brief description and discussion of the board and firm characteristics of FFs and NFFs. In our sample, the FFs had smaller boards (Board Size). In NFFs, the average board size is approximately 8.4 members, whereas FFs have an average of only 7.7 members. Regarding the proportion of executive board members (Executive Directors), there was also a significant difference between the types of companies: 29% and 32% for NFFs and FFs, respectively.

Significant differences were also observed in the independence of the board members (Independent Directors). In NFFs, approximately 14% of board members are independent, whereas, in FFs, the proportion is considerably lower, amounting to only approximately 9%, which is consistent with several prior studies that reported lower percentages of board independence for FFs relative to NFFs (Anderson and Reeb, 2003; Scafarto et al., 2020). We find significant mean differences in the financial variables (Risk, ROA and Market-to-book), with FFs being less profitable, riskier and FFs being smaller.

We now turn to the gender diversity-related variables that are particularly relevant to our study. It turns out that female representation is limited in all types of firms: our sample statistics show that, on average, around 38% of our sample (1,157 out of a total 3,062 firm-year observations) exhibited boardroom diversity, with at least one female director on their board (Female Dummy), whereas 62% did not have a single female director. Additionally, female representation on boards is more frequent in FFs than in NFFs. Approximately 44% of FFs had at least one female director (Female Dummy), whereas only approximately 24% of NFFs had a female presence on the board. The proportion of women on boards (Female Directors) averaged around 9.2% for all firm-year observations during the period under review, putting the proportion of women on Pakistani boards above the Asian average, which was 6% according to the study by Süssmuth-Dyckerhoff et al. (2012) and 7.8% in 2016 (Deloitte, 2019). These results do not support H1, according to which we expected a (comparatively) low proportion of women on boards because of pronounced gender inequality (Pakistan ranks second to last in the World Economic Forum, 2022 gender parity index) and the absence of mandatory quotas for women. However, there were clear differences by company type. While NFFs had only around 4.1% women on their boards, FFs had around 11.3%. That is, the proportion of women on boards in FFs is around three times higher than that of NFFs, and that female presence is predominantly driven by appointment decisions made by family businesses, which supports H2. Finally, only around 2% of all CEO or chair positions are held by women (Female CEO or Chair Dummy) in NFFs. In FFs, women are appointed to these top positions significantly more often (approximately three times).

Table 3 shows the average percentage of women on boards over time, which barely changed during the study period (2003–2017), amounting to approximately 9% of all firms both at the beginning and end of the period under review. The proportion of women in non-family and family businesses, approximately 4% and 11% on average, respectively, has remained relatively stable, indicating that gender diversity recommendations included in corporate governance codes since 2012 have not had an impact on the proportion of women on boards, confirming that voluntary approaches (“soft law”) do not bring about changes in women’s representation (Terjesen et al., 2015). Female directors generally serve longer when they are from a sponsoring family in Pakistan (Shoaib and Saleem, 2022). In NFFs, female directors are also likely to serve for a long time (predominantly as the only woman on the board) and are typically well-educated women from wealthy families whose position results from close personal contacts and networks of the men [on sifarish, an Urdu term describing this type of favouritism, see Sarwar and Imran (2019)], who are also replaced by a woman upon leaving likely owing to the positive signal to capital markets and international business connections (Saeed et al., 2016).

3.3 Multivariate analysis

Next, we examine the reasons for nominating a female director to the board or a woman to a top position by firm type (family vs non-family) using regression models. For this purpose, we draw on a probit model (Adams and Ferreira, 2009; Bianco et al., 2015). We expected FFs to exhibit different behaviours with respect to women’s appointments (see H2). To determine whether this is true, we use a firm-clustered probit model to estimate the likelihood of women sitting on boards or holding top positions. Table 4 shows the results of the probit regressions and average marginal effects.

In Model 1 the dependent variable is a dummy that equals 1 if at least one female director is on the board (Female Dummy). In Model 2 the dependent variable is a dummy equal to 1 if at least one Female CEO or Chair is on the board (Female CEO or Chair). The main variable of interest is family ownership (Family Firm), a dummy variable equal to 1 if it is a family firm and 0 otherwise. The remaining independent variables are commonly used in related research on board composition and the decision to fill top company positions with women (see, for example, Anderson and Reeb, 2004; Adams and Ferreira, 2009; Bianco et al., 2015; Saeed et al., 2016). Specifically, we examine whether family ownership is related to female presence and control for the firm and board characteristics identified in previous research, such as size variables (Board and Firm Size), performance measures (ROA and Market-to-book Ratio), firm risk (Risk) and Firm Age (Firm Age). Year dummies are included, and standard errors are adjusted for potential heteroskedasticity and group correlations at the firm level. The results for the two ownership structures show that FFs are more likely to have female members on their boards and that the magnitude is economically significant (see Table 4, Model 1). Specifically, FFs are 20.7% more likely to have a female director on their board. Furthermore, smaller companies and firms with higher performance (Market-to-book Ratio) have a higher probability of female directors. Model (2) of Table 4 presents the probit estimation results for the determinants of a female CEO or chair. Because firms with female CEOs or chairs are rare, almost all the variables are insignificant (Morikawa, 2016). However, the coefficient of FFs was positive and significant at the 5% level. The results show that women in FFs are more likely to hold leadership positions such as CEOs or chairs.

In Models (3) and (4), we include two additional governance variables: the proportion of independent directors (Indep. Dir) and block ownership (a dummy variable that equals 1 if 5% or more of the outstanding shares are held by financial investors; Inst. Owner). According to Anderson and Reeb (2004), both variables can be relevant for efficient (external) monitoring and an effective board, and should favour decisions that are beneficial for the company. Adding these two variables does not change the results. The marginal effects of the dummy variable family firm are approximately the same as those in Model (1) and (2) and remain significant. Thus, even after adding this ownership and governance variable, FFs behave differently from NFFs with respect to the decision to send women to top management positions.

3.4 Robustness tests

Empirical research on board composition and gender diversity faces endogeneity problems, omitted variable bias and reverse causality issues (Anderson and Reeb, 2004; Adams and Ferreira, 2009). Consistent with the literature (Adams and Ferreira, 2009; Saeed et al., 2017), we draw on the proportion of female directors on the board (the “female directors” variable) as an alternative to the dependent dummy variable “female dummy” for further analysis. To this end, we present (analogous to Adams and Ferreira, 2009) the results of an ordinary least squares regression in Column (1) of Table 5.

The coefficient of the family firm variable is significant at the 1% level. As Table 5 shows, the coefficients of the other control variables are consistent with our previous results.

As a robustness check, we analysed the relationship using the Tobit maximum likelihood estimator (see also Saeed et al., 2016; Gregorič et al., 2017), which specifically considers the distribution of our dependent variable [i.e. the values are limited to 0–100, and a large number of companies do not have female directors, see Gregorič et al. (2017)]. Columns (2) to (6) of Table 5 present the results of the Tobit estimates. All specifications include year dummies, and the t-statistics are reported in parentheses on robust standard errors clustered at the firm level. Model (2) confirms the previous main result that FFs are associated with significantly higher female representation on the board. In Model (3), we added two more variables: the proportion of independent directors (Indep. Dir.) and block holding (a dummy variable) by financial investors (Inst. Owner). Our variable of interest, FFs, continued to be highly significantly correlated with the proportion of women; we found that the coefficient of the proportion of independent directors was significantly negatively associated with the proportion of women and that of financial investors was also associated (but only at the 10% level) with a lower female presence. The estimated coefficients of the other explanatory variables were qualitatively similar to the earlier results. For further robustness, following Anderson and Reeb (2004) and Setia‐Atmaja et al. (2009), we vary the definition of “family firm” in Models (3) to (6) and create new dummy variables using the family’s fractional ownership of the firm’s outstanding equity (Anderson and Reeb, 2004). In Model (4), we code a firm as a family firm if the family holds 10% equity; in Model (5), we increase the share to 30%; and, finally, in Model (6), to 40% or more.

The results show that our main finding that FFs have a higher female presence (proportion of women) on boards is robust to the different family firm definitions in all models. The coefficient of the variable Indep. Dir remained significantly negatively correlated with female sex. Robustly across all models, smaller firms are associated with higher board gender diversity and market valuation (Market-to-book) is significantly and positively associated with female presence in all models. The results, with additional governance variables, independent directors and financial investors, provide new insights.

Independent directors and other blockholders perform particularly important monitoring functions and roles in good corporate governance in (family) firms (Anderson and Reeb, 2004). According to our analysis results, FFs appoint both more women and fewer independent directors to their boards and fill boards with family-affiliated members (as the saying goes, “Blood is thicker than water”). FFs are likely to perceive board appointments as important channels through which they can maintain and consolidate board control (Sarkar and Selarka, 2021). The results show that a higher proportion of independent directors is associated with significantly fewer family-affiliated female nominations and a lower proportion of females, implying that, similar to institutional (financial) investors, independent directors discourage the appointment of (family-affiliated) women.

Therefore, consistent with H2, FFs have a higher proportion of women on the board. Clearly, family status determines a woman’s board appointments. The fact that female presence and leadership positions are related to ownership structure suggests that this is primarily a matter of family policy decisions rather than value-enhancement considerations. Thus, female directors are selected by families primarily to serve their interests, which is consistent with the statement of a Pakistani CEO of a family business: “I do not like outsiders on my board. Our board only has family members. Some of them are women also. As we are a conservative family; our women do not attend the board meetings […]” (ACCA Pakistan, 2010, p. 12).

In summary, in terms of answering our research question – what drives the appointment of women to boards in FFs in the Pakistani context – we found evidence that Pakistani FFs follow self-serving SEW priorities. Family ties and control considerations dominate board nomination decisions in FFs and play a greater role than reservations regarding women occupying leadership positions. These results raise the question of whether women can positively influence firm value in this environment. It can be assumed that women cannot act independently but are instead forced to prioritise family goals over shareholder value (Low et al., 2015; Herdhayinta et al., 2021).

4. Discussion and conclusions

Pakistani FFs had a significantly higher female presence on their boards between 2003 and 2017. On average, FFs had approximately 11% of women on their boards, while NFFs had only approximately 4%. Women hold top positions, such as CEO or chair, significantly more often in family-controlled companies. The percentage of women on boards has been stable over time and has changed little over the 15-year period, confirming (in line with De Cabo et al., 2019) that mere recommendations regarding more diversity (“soft quota”), such as those in Pakistan’s 2012 Corporate Governance Code, do not trigger any changes in board diversity. The findings further imply that FFs continuously popularise boards with family-affiliated women, a practice primarily aimed at avoiding “outsiders”, thereby ensuring the protection of family interests and the maximisation of SEW (Gómez-Mejia et al., 2007; Berrone et al., 2012; Molly et al., 2019). According to the SECP, the mandatory women’s quota introduced by the Companies Act 2017, which requires at least one woman to be appointed to the board, has been very successful and currently, 87% of all listed companies have at least one woman on the board (Securities and Exchange Commission of Pakistan, SECP, 2023). This significant increase (in our pre-quota study period, only 38% of boards had at least one woman) corresponds with the results of the study by Bhattacharya et al. (2022) for India. There, it was shown that the new gender law (with the Companies Act, 2013, at least one female director must be hired) is implemented by a very high percentage of companies, i.e. “diversity mandates work” (Bhattacharya et al., 2022). However, most of the newly hired women belong to the personal network or family of the controlling shareholder (Vohra, 2020). To meet the women’s quota, Indian companies have relied on existing director networks. This limits the search for unique and independent women (Vohra, 2020) and thus does not optimally address diversity (Bhattacharya et al., 2022). In the spirit of “keeping intact the old boys’ network” (Bhattacharya et al., 2022), our research suggests that the new female directors on Pakistani corporate boards come mainly from family and friends.

This study responds to the call of the Chartered Governance Institute in a recent report (Hawarden and Greenwood, 2021) to address research gaps regarding board diversity in some countries, particularly those where barriers such as cultural factors (e.g. traditional patriarchal attitudes) play a role in preventing women from participating on boards. By empirically examining board characteristics and the female presence in Pakistani-listed FFs, we contribute to a better understanding of the factors that influence board diversity in emerging markets.

The decision of family businesses in a (Muslim-majority) country, which has strong patriarchal structures and ranks second last internationally in gender equality, to send more female family members to management positions seems to be in line with the saying “Blood is thicker than water”. Family and SEW-related objectives, such as exercising family control, seem to take precedence over certain socio-cultural values (e.g. traditional patriarchal attitudes). However, results from interviews in Pakistan imply that female family members are not empowered to contribute their unique skills to these positions (ACCA Pakistan, 2010; Chang et al., 2021); rather, they (must) follow the business policies set by male family members, which primarily serve to ensure family interests (such as continuity and control). Accordingly, women are “confidants”, taking a “passive role” (Chang et al., 2021, p. 4) and are, as one Pakistani male CEO of a family business puts it, “loyal” on boards (ACCA Pakistan, 2010, p. 13). These results have important implications for both policymakers and investors.

In countries with a weak institutional environment, a high concentration of (family) ownership and patriarchal social norms, gender quotas do not appear to be an effective instrument for breaking through the glass ceiling but rather lead to an increase in female family members being sent to the boards, who, however, do not have the power and legitimacy to make board decisions, but have to implement family policy interests (Low et al., 2015; Herdhayinta et al., 2021). This is also underscored by the statement of a female Pakistani CEO: “Quotas or positive discrimination or the Code of Corporate Governance requiring [female] representation will not result in gender-diverse boards in their true spirit” (ACCA Pakistan, 2010, p. 16).

For women to climb the corporate ladder successfully, quotas alone are not (equally) effective. A corresponding political and social environment is required to realise the economic and social advantages of diversity. Consequently, factors favouring women’s appointment and empowerment on boards need to be considered in national, institutional, social and cultural contexts, and the drivers may differ significantly between Western industrialised countries and emerging economies.

From an investor’s perspective, it is doubtful whether board participation in a patriarchal (family) environment in which women are strongly disadvantaged can positively affect company value. It is difficult for women on these boards to appropriately contribute their skills and experience, make decisions independently, and thus contribute to a more efficient board and positively influence corporate policy and shareholder value. This contrasts with the situation in Western countries, where studies show that women are largely independent directors [see Adams and Ferreira (2009) for the USA and Wang and Kelan (2013) for Norway]. Thus, policymakers might consider introducing the mandatory presence of independent female directors [for recent regulations in India, see Sarkar and Selarka (2021)]. However, given that the positive impact of women in leadership positions on firm performance cannot unfold until there is high gender parity and progressive attitudes towards women in a country (Post and Byron, 2015; Hoobler et al., 2018), quotas, in whatever form, will not be sufficient. Gender parity needs to be addressed at the root by authorities, requiring multi-level support and enforcement of regulatory and legislative reforms for structural and cultural transformation – not only in educational access and female workforce participation but also in social role structures (Sarwar and Imran, 2019).

This study has some limitations, as it is restricted to listed companies in a particular study period in Pakistan, a country with a specific institutional and socio-cultural environment; thus, the results have limited generalisability to other companies and countries or emerging markets. Furthermore, female directors in FFs were classified as belonging to the family, although their identities (i.e. family affiliation) could not be verified.

Overall, the current study sheds light on the factors that inhibit but also encourage the appointment of women on boards in an emerging market, which has been very little explored. According to our analysis, the appointment of women to boards in Pakistan is primarily a “family case”.

Table 1.

Definition of the variables

Variable	Definition
Board characteristics
Board size	Number of directors on the board
Executive Directors	Proportion of executive directors on the board
Independent Directors	Proportion of independent directors on the board
Female Directors	Proportion of female directors on the board
Female Dummy	Dummy variable equal to 1 if at least one female director holds a board seat
Female CEO	Dummy variable equal to 1 if the CEO is female and otherwise 0
Female Chair	Dummy variable equal to 1 if the Chair is female and otherwise 0
Female CEO or Chair	Dummy variable equal to 1 if the CEO or Chair is female and otherwise 0
Firm characteristics
Risk	The standard deviation of a firm’s daily stock return over the prior 12 months
ROA	Return on assets: net income divided by total assets
Market-to-book	Market value of common equity divided by book value of common equity
Firm age	Number of years since the firm’s inception
Firm size	Log of firm’s total assets
Family firm	Dummy variable equal to 1 if the firm is a family firm and otherwise 0
	Family must have at least 20% ownership and two managerial positions in the firm

Source: Authors’ own work

Table 2.

Descriptive statistics and difference-of-means test for family and non-family firms

		All firms (n = 3,062)		Non-family firms (n = 880)	Family firms (n = 2,182)
Variable		Mean	SD	Mean (I)	Mean (II)	(I)–(II)	t-statistics (III)
Board characteristics
Board size	No.	7.886	1.307	8.356	7.696	0.66	12.976***
Executive Directors	%	31.436	16.639	29.586	32.182	−2.596	−3.916***
Independent Directors	%	10.665	12.672	13.924	9.35	4.574	9.1608***
Female Directors	%	9.253	14.3032	4.14	11.316	−7.176	−12.870***
Female Dummy		0.378	0.485	0.235	0.435	−0.200	−10.519***
Female CEO (dummy)		0.022	0.145	0.011	0.026	−0.015	−2.468**
Female Chair (dummy)		0.026	0.16	0.008	0.033	−0.025	−4.013***
Female CEO or Chair (dummy)		0.047	0.213	0.019	0.059	−0.040	−4.693***
Firm characteristics
Risk	%	47.262	27.839	43.791	48.662	−4.870	−4.338***
ROA	%	5.034	10.017	6.915	4.276	2.639	6.635***
Market-to-book		1.32	1.277	1.78	1.135	0.646	12.995***
Firm age (years)		31.811	16.305	32.231	31.736	0.495	0.759
Firm size (log total assets)		15.227	1.436	15.39	15.162	0.228	3.989***

Notes:

Statistical significance at: ***p < 0.01; **p < 0.05

Source: Authors’ own work

Table 3.

Average percentage of women on boards over the years

	All firms		Non-family firms		Family firms
Year	N	Mean	N	Mean	N	Mean
2003	181	9.11	59	4.17	122	11.51
2004	184	9.09	59	4.17	125	11.41
2005	190	9.00	61	3.59	129	11.55
2006	196	9.19	61	3.16	135	11.92
2007	198	9.27	64	2.87	134	12.33
2008	205	10.17	59	4.90	146	12.30
2009	209	9.98	59	3.84	150	12.39
2010	210	9.31	58	3.89	152	11.37
2011	212	9.08	61	4.15	151	11.07
2012	213	8.99	60	4.93	153	10.58
2013	213	9.27	59	5.04	154	10.88
2014	213	8.85	56	4.01	157	10.58
2015	213	9.28	55	4.10	158	11.08
2016	213	8.97	55	4.69	158	10.46
2017	212	9.24	54	4.82	158	10.75
Total	3,062	9.25	880	4.14	2,182	11.32

Source: Authors’ own work

Table 4.

Determinants of female directorship/leadership

	(1)	(2)	(3)	(4)
Variables	Female Dummy	Female CEO or Chair	Female Dummy	Female CEO or Chair
Family Firm	0.210*** (0.000)	0.062** (0.038)	0.196*** (0.000)	0.040** (0.012)
Board Size	0.011 (0.564)	−0.007 (0.318)	0.015 (0.455)	−0.005 (0.390)
Indep. Dir.			−0.004* (0.052)	−0.000 (0.408)
Inst. Owner			−0.110 (0.122)	−0.026* (0.097)
Firm Size	−0.050** (0.019)	−0.013 (0.231)	−0.054** (0.019)	−0.011 (0.188)
Firm Age	−0.001 (0.380)	0.000 (0.828)	−0.000 (0.601)	0.000 (0.708)
ROA	−0.002 (0.227)	0.000 (0.975)	−0.002 (0.247)	0.000 (0.934)
Market-to-book	0.034** (0.012)	0.010* (0.089)	0.038*** (0.008)	0.009 (0.106)
Risk	0.0003 (0.461)	−0.000 (0.487)	0.0004 (0.480)	−0.000 (0.504)
Year effect	Yes	Yes	Yes	Yes
N	2,965	2,965	2,965	2,965
Pseudo R2	0.051	0.0835	0.061	0.0920

Notes:

This table presents results from probit regressions. Standard errors are adjusted for potential heteroscedasticity and for group correlation at the firm level. Year dummies are included. The effect of the constant term is omitted. Models 1 and 3: the dependent variable is a dummy variable, which equals one if at least one female director is on the board. Models 2 and 4: The dependent variable is a dummy variable, which equals one if a female CEO or female Chair is on board. Average marginal effects are reported. P-values are reported in parentheses. ***, ** and * denote statistical significance at the 1, 5 and 10% levels, respectively

Source: Authors’ own work

Table 5.

Determinants of the proportion of women on the board

	(1)	(2)	(3)	(4)	(5)	(6)
Variables	Female directors (%)	Female directors (%)	Female directors (%)	Female directors (%)	Female directors (%)	Female directors (%)
Family Firm	6.432*** (4.68)	17.827*** (4.18)	16.375*** (3.90)	15.520*** (3.59)	18.363*** (4.71)	8.950** (2.35)
Board Size	−0.399 (−1.05)	−0.287 (−0.23)	−0.107 (−0.09)	−0.350 (−0.29)	0.244 (0.21)	−0.601 (−0.49)
Indep. Dir			−0.301** (−2.43)	−0.323** (−2.55)	−0.300** (−2.42)	−0.374*** (−2.86)
Inst. Owner			−10.185* (−1.94)	−9.824 (−1.86)	−7.653 (−1.55)	−8.736* (−1.69)
Firm Size	−1.571** (−2.49)	−4.011*** (−2.61)	−4.080*** (−2.71)	−3.965*** (−2.60)	−4.116*** (−2.77)	−3.945*** (−2.59)
Firm Age	−0.063 (−1.14)	−0.159 (−1.21)	−0.101 (−0.77)	−0.108 (−0.82)	−0.093 (−0.72)	−0.067 (−0.51)
ROA	−0.101 (−1.48)	−0.229 (−1.42)	−0.214 (−1.35)	−0.219 (−1.37)	−0.231 (−1.47)	−0.207 (−1.33)
Market-to-book	0.716** (2.13)	2.191*** (2.86)	2.222*** (3.09)	2.241*** (3.11)	2.211*** (3.08)	1.506* (1.94)
Risk	−0.005 (−0.37)	0.002 (0.07)	0.002 (0.06)	0.003 (0.10)	−0.001 (−0.05)	0.002 (0.05)
Year effect	Yes	Yes	Yes	Yes	Yes	Yes
N	2,965	2,965	2,965	2,965	2,965	2,965
R2/Pseudo R2	0.0891	0.0207	0.0251	0.0236	0.0294	0.0200

Notes:

This table presents results from OLS [Model (1)] and Tobit regressions [Models (2) to (6)]. Year dummies are included. The effect of the constant term is omitted. In all models, the dependent variable is the proportion of female directors on the board [female directors (%)]. In Columns 4 through 6, we use alternative family firm definitions (family ownership varies from 10% to 40%). The t-statistics are reported in parentheses on robust standard errors clustered at the firm level. ***, ** and * denotes statistical significance at the 1, 5 and 10% levels, respectively

Source: Authors’ own work