Diversity is a prerequisite for change and evolution, governed by a continuous interplay of system’s variety and selection mechanisms. Jacques Sapir claims that heterogeneity should not be treated as an accidental byproduct, but rather conceptualized as a norm and acknowledged as an alternative economic theory (Sandven et al., 2008). The idea that gender diversity on the board of directors adds value to companies is based on a few theories, e.g., agency theory, resource dependence theory, and stakeholder theory. Agency theory assumes that female directors bring different perspective on complex issues and therefore can help more efficiently solve complex corporate problems. Resource dependence theory suggests that women directors elected to boards effectively enable a better access to critical resources due to an enhanced repertoire of skills, cues, and attributes. Stakeholders’ theory indicates that heterogeneous boards may become more creative, even though they may take longer to negotiate and achieve consensus (Dang & Nguyen, 2016).

Brieger et al. (2019) state that structural barriers for female board candidates still exist in the practice, and several individual, firm, and industry-specific characteristics determine the status quo. Women often require higher academic achievements than men to be considered for a board nomination. Females also more frequently than men feel discouraged to aspire to such high-profile positions due to their modest self-image, discriminating stereotypes, or lack of networking opportunities in high-end corporate circles. Furthermore, firm size and masculine corporate culture may impede women’s election to the governance bodies. In contrast, public or non-profit character of an enterprise increases the chances for female directors, due to a more favorable business context. Lewellyn and Muller–Kahle (2020) find that a greater economic and political empowerment accompanied by shared cultural values, beliefs, and attitudes of a country successively helps women overcome the glass ceiling of corporate elites. Boardroom’s gender democratization happens however slowly, with the first wave of feminization being based on political background, the second one facilitated by academic achievements of the female board candidates, and the third one accomplished by exquisite and company grown female talent (Heemskerk & Fennema, 2014). From human capital perspective, young women have closed the gender gap and overtaken young men in many countries in enrollment and exam passes at school and university, but they fail to succeed in the corporate governance due to complex reasons (Walby, 2011).

Women are underrepresented in science and engineering professions as well as in academia on all academic levels (Adams & Kirchmaier, 2016). The global aggregate percentage of women in board positions reached ca. 20% in 2019, hence increased by ca. 10 percentage points since 2009 (Emelianova & Milhomem, 2019). In Europe, the female board ratios rose by 17 percentage points, up to 28% between 2009 and 2019 (EU Gender Statistics Database), which is still far away from either gender parity or the European Commission’s proposal of 40% by 2020 (Jourová, 2016). Sabatier (2015) confirms that quotas have prompted French companies to engage more female directors, which positively impacted companies’ performance, while Comi et al. (2020) post mixed results. They disclose only positive effect on firm labor productivity in Italy, but a negative one in France, including an insignificant impact on profitability of Spanish companies in the scope of their study.

The contemporary board statistics contrast with research findings that females’ characteristics may complement or ameliorate the input of male directors in multiple business management dimensions. Solakoglu and Demir (2016) argue that an increased portion of female directors alters board’s supervisory behavior. Nielsen and Huse (2010) propose “that it is not the gender per se, but the different values and professional experiences that women may possess that enable them to make a difference to actual board work and influence board decision-making” (p.16–17). They conclude that a female board member equivalent with similar (board relevant) professional experiences and different values can enhance board decision-making.

Our study analyzes the effects that increased female representation on a board of directors of business firms has on customer satisfaction (CS), described as an “overall evaluation based on the total purchase and consumption experience with a good or service over time” (Anderson et al., 1994, p.54).Footnote 1 Our paper contributes to the literature on the determinants of CS. The literature on this topic is rich but partly fragmented. Many studies have contributed by analyzing different variables, but none of the previous publications has tried to simultaneously test a larger set of variables identified to be significant in earlier studies. Churchill and Surprenant (1982) suggest that CS encompasses: expectations, performance, disconfirmation, and satisfaction. Spreng et al. (1996) specify that satisfaction arises when product fulfills both consumer’s desires and expectations. Cotiu (2013) reviewed further determinants of CS. Table 1 summarizes the major factors impacting customer satisfaction considered in our study.

Table 1 Literature review of the determinants of customer satisfaction in our study
Full size table

We also add to important strands in the economics and management literature on the potential gains and benefits that firms can achieve with gender diversity in leading positions in business enterprises. Despite a substantial body of literature having studied the gender boardroom diversity issue, no scientific consensus has emerged yet. The lack of unanimous research outcomes is mainly caused by different institutional, regulatory, and legislative systems for sampled companies, critical mass phenomena, different research periods, or individual approaches of scholars. We consider CS a socially responsible measure of corporate success and therefore would like to add one additional perspective on this matter. Our research results acknowledge a possibility of a significant association of female directors’ representation on CS, which we interpret in line with the studies that ascertain positive effects of women on boards in terms of better addressing consumer preferences and building better relationships with external institutions.

Boards of directors have been traditionally focused on controlling financial accountability, executives’ nomination and evaluation, and strategic oversight of a company. But changing consumers’ sentiment, the proliferation of corporate accounting and the rising complexity of the global economy have caused some boards to reassess their responsibilities and mission. In this process, female directors increasingly participate in the corporate governance process and have been witnessed to have specialized knowledge (Dunn, 2012) and pay a specific attention to decision-making factors that might also influence CS. Women directors are said to “humanize” personal interactions on boards and may have a positive impact on attendance at board meetings (Adams & Ferreira, 2008; Singh & Vinnicombe, 2004). Women on a board of directors tend to be more critical with the content of board work, which may improve the boards’ quality of advice (Freeman & Varey, 1998; Kim & Starks, 2016). Gender diversity leads to more transparency (Upadhyay & Zeng, 2014), higher detail levels of corporate sustainability reporting (Al-Shaer & Zaman, 2016), and a better adoption of sustainability reporting and external assurance (Girón et al., 2021). Female directors tend to be more diligent and may change not only the decision-making culture but also the tone of board discussions (Huse & Solberg, 2006). Oliver (1996) notes that, while men tend to measure their financial performance quantitatively, females do not hesitate to use either customer service ratings or diverse satisfaction scores. In accordance with the upper echelon theory, an increased female ratio may influence strategic choices made by the board (Nielsen & Huse, 2010).

Women may endorse a “power sharing” management style (Bradshaw et al., 1992), which may contribute to a more egalitarian decision-making process within an organization (Burgess & Tharenou, 2002). Women may prefer to engage as transformational and servant leaders (Duff, 2013; Eagly & Carli, 2003). Such a servant leadership style with behavioral dimensions such as empowerment and development, building trust, humility, altruism, authenticity, responsibility, and interpersonal acceptance and competence (Beck, 2010; Van Diernedonck, 2011) can add significant value to organizations (Amah, 2018; Erdurmazli, 2019).

Female advisors seem to be more sensitive to unethical judgments (Lunsford, 2000); simultaneously, women show more interest in corporate philanthropy (Selma et al., 2020) as well as utilitarian and altruistic endeavors (Simmons & Emanuele, 2007). Because psychological need satisfaction mediates the relationship between perceived justice and organizational identification and employee commitment (Malhotra et al., 2020), female leaders may be more likely to foster workforce motivation and loyalty (Burgess & Tharenou, 2002), in part because women often prefer a collaborative leadership style and take a more personal approach to employees and customers (Allen & Truman, 1993),Footnote 2 thus contributing a group-oriented organizational culture emphasizing commitment which may drive success of projects (Patanakul & Aronson, 2012). Women tend to pay more attention to the process of how an outcome is achieved and are therefore more likely to act as business defenders (Johansen, 2007). Female directors holding monitoring roles mitigate managerial opportunism focused on earnings. Saona et al. (2019) report that “equilibrated board tends to mitigate earnings management practices” (p.1). Zalata et al. (2019) and Yang et al. (2019) suggest that women directors generally limit a firm’s risk performanceFootnote 3 measured by a volatility of equity returns. A qualitative management survey by Adams and Funk (2012) finds female directors more risk-loving but also more benevolent and caring for universalism and stimulation. Women in the board room bother less about power, security, conformity, and tradition, which can help if any problematic status quo in customer relations is being questioned. Ringqvist (2014) found that women in Swedish businesses tend toward more conservative pricing. This is important because a moderate pricing strategy is an important level of customer satisfaction. The Swedish Women Resource Centers (WRC) have been classified as innovation systems that link actors from different spheres of society to develop new knowledge and transform it into innovations (Lindberg et al., 2012).

Turning from internal relationships and effects to external ones, females put more emphasis on communication that aims for a high level of customer loyalty and long-term satisfaction (Mukhtar, 2002). Gender diversity may make consumers feel that their preferences and profiles are better addressed (Burke, 1999; Bilimoria & Wheeler, 2000). Women directors are more concerned about supply chain and reputational issues (Groysberg et al., 2016), which benefit relationships with external institutions (Singh & Vinnicombe, 2004). Finally, companies with more diversified boards achieve a higher quantity and novel patents (Griffin et al., 2021). The innovativeness of complex enterprises can jump when the proportion of female leaders exceeds 20% (Lorenzo et al., 2017), as a feminine and servant leadership style can promote service-innovative behavior and intrinsic motivation among employees, especially in the case of individual identification with the leader (Su et al., 2020). Empowering leadership significantly impacts quality of people and methods such as technology endorsement, job design, or innovative culture, as well as creating employee engagement (Muafi et al., 2019). A larger female representation has a positive impact, both on financial aspects (Hsu et al., 2019; Lückerath-Rovers, 2013) and on stock pricesFootnote 4 when a favorable effect on a firm’s financial performance indicators (e.g., Tobin’s Q and Return on Assets) is larger in high-performing firms (Conyon & He, 2017). Turban et al. (2019) explicitly control for the reverse causality argument and confirm the positive effect of gender diversity on firms’ market valuation (e.g., Tobin’s Q or turnover ratio), but only in national contexts where gender diversity was normatively accepted (e.g., Western Europe vs. Japan). Rossi et al. (2018) determine that more women on board of directors increase company’s indebtedness level, but the invested capital is used more efficiently.

The positive impact of female directors, however, might be also contingent on external circumstances. The case studies conducted yield not unanimous results due to their individual data samples, geographic coverage, and different research periods. Dezsö and Ross (2012) find that women in executive management improve company performance exclusively if the female representation is moderated by a high innovation intensity (the ratio of the “research and development” (R&D) expenses to assets from the prior year). Farrell and Hersch (2005) find that the likelihood of electing a woman to a corporate board negatively depends on the number of females already appointed to it in the past. They suggest that firms may add female directors only as a defensive reaction to outside pressure and that scarce female top performers might proactively prefer the stronger- and better-positioned companies. There may even be negative impacts of female directors on a company’s share price (Ryan & Haslan, 2005), profitability (Shrader et al., 1997), and firm performance (Dang & Nguyen, 2016; Triana et al., 2013). Other studies report neutral findings and 0 effects on, for example, excess returns (Francoeur et al., 2008), Tobin’s Q (Rose, 2007), IPO underpricing (Mohan & Chen, 2004), or a dividend payout (Pucheta-Martinez & Bel-Oms, 2016). Ferreira (2015) criticizes the literature that seeks to prove the positive effects of female directors on firms’ profitability and suggests rather to focus the research on female directors’ potential benefits to society.

Since many sets of determinants of customer satisfaction have been tested independently from each other, we propose an integrated approach to accomplish a comprehensive analysis of parameters influencing CS overall and additionally investigate the effect of women on boards on CS. We apply a two-stage methodology. First, we run a regression model, including all the variables identified as significant in earlier studies. Second, we add gender-specific variables to test our hypothesis of an impact (favorable) by women directors on CS. We find a positive relationship between a higher female ratio on the board and customer satisfaction scores of companies, which we discuss along other results.

Data and Empirical Strategy

We use the American Customer Satisfaction Index (ACSI), developed by the Michigan Ross School of Business and the American Society for Quality, as well as the National Customer Satisfaction Index (NCSI) UK measured by the CFI Group in the UK, as dependent variables. The ACSI is reported annually on a scale of 1 (min) to 100 (max) (http://www.theacsi.org) and is calculated consistently across service and manufacturing sectors from a telephone survey of indexed brands. The statistical pool of the ACSI represents 10 economic sectors, 47 industries, and more than 230 firms. The NCSI index scores are expressed as a number out of 100 (max), and results are based on survey data from 5000 customers, collected via an online panel.Footnote 5 The NCSI UK applies the technology of the CFI Group (https://cfigroup.com) and the methodology of the ACSI. The composition of the data sample regarding industry affiliation and geographic origin of the firms (including calculated average CS values) is presented in Table 2.

Table 2 Data sample: selected industries and average CS scores in 2009
Full size table

We review a group of 200 publicly listed enterprises profiled by the indexes’ creators, of which 158 had been scored by ACSI, 32 by NCSI, and 10 by both indexes in parallel (as they run multiple brands in American and European markets). However, in our research sample, we include a balanced longitudinal panel data set of 103 companies that were consistently scored by at least one of the indexes within a longer time period of 5 years (2005–2009) and had a full range of financial details available, so that our final sample has 515 multiyear observations in total. The observed firms reported over US $3.298 billion revenues in 2009 and employed more than 10.7 million people. The data sample includes 97 companies that provide consistent ACSI data and 6 companies that are continuously covered by both indexes.

Additionally, we include a range of independent determinants that have been found by different researchers to have an impact on CS. Table 2 provides a summary of the major determinants of customer satisfaction scrutinized by different studies on CS. Anderson et al. (1994) find a negative association between market share and customer satisfaction in the short term, arguing that companies might dilute their customers’ experience while instantly differentiating business offerings and overstretching firms’ responses to too many niche markets simultaneously. A proprietary Swedish survey and database also indicate that price and quality as well as quality expectations may have a positive impact on CS. Anderson et al. (1994) use a Pearson’s correlation and analyze the relationship between CS and market share (in 1989–90) and find a significantly negative relationship. Our market share (MARKET SHARE) data is calculated from individual company revenues and the total market value (on a 5-digit NAICS level) in 2007, released by the US census. Due to the lack of publicly and readily available information, we cannot test Anderson et al.’s variables (1994), which was tested with a proprietary dataset—such as the impact of quality, price, and customer expectations on consumer satisfaction.

Mittal et al. (2005) discover a significant positive relationship linking different corporate cost categories to customer satisfaction and attest a negative relationship for sales efficiency. They identify advertising expenses (ADDS); sales efficiency (SALES EFFICIENCY); sales, general, and administrative expenses (SG&A); and cost of goods sold (COGS) as valid input variables to their regression analysis on CS. They do not claim “to have identified the ‘correct’ set of satisfaction inputs, nor have (they) included the ‘exhaustive’ set of satisfaction inputs” (p. 548), since their selection has been based on data availability and some preliminary theoretical justifications. We extract the SG&A and COGS directly from 10 K reports. For firms that do not disclose advertising spending in their regular accounting publications, we used the data from Schonfeld and Associates (2010). We calculate the annual sales efficiency as a ratio of total company headcount divided by the total revenues per anno (in million US$). The cost ratios for SG&A, advertising expenses, and COGS are estimated by dividing the respective cost aggregates by the total revenues of a company in the corresponding years.

Simon and De Varo (2006) confirm that companies distinguishing themselves by an employee friendly human resource policy can achieve significantly higher customer satisfaction scores. Company profitability measured by return on assets impacts CS significantly and positively, but an opposite relation is stated for total assets. Simon and De Varo (2006) deploy “best company” status (BEST COMPANY)—a binary variable—as a positive driver of CS. They also control for financial and organizational factors, such as return on assets (ROA), logarithm of employees’ count (ORGSIZE), logarithm of total assets (TAS), unemployment rate, and industry fixed effects or multiple brands. We leverage similar key performance indicators on headcount and finance for all of the companies in our sample and for all 5 years, which we source from company financial reports. Annual corporate profitability is measured by return on assets and computed as a ratio of earnings before interest and taxes (EBIT) to total assets per company.

Information on a “best company” status for the years 2005 to 2009 was taken from an online database of The Great Place to Work Institute website (http://www.greatplacetowork.net/), published by Fortune magazine in a list of top employer brands. We included “best company” data per firm as a dummy variable with a value of 1 for companies awarded by the institute in the respective years. We also followed Simon and De Varo (2006) by controlling for industry affiliation of a company (consumer goods, utilities, technology, media, telecom, retail, automotive, and financial services). We did not include the unemployment rate in our model, due to its insignificance in the findings of Simon and De Varo (2006).

Singh et al. (2010), based on some 400 questionnaires from Indian companies, find a positive correlation between inventory management and CS, suggesting that a smooth supply chain and effective product delivery capability are crucial for customer satisfaction. Influenced by this thought, we proxy delivery capability by the indicator, “days of inventory outstanding” (DIO), which measures the number of days it will take a company to sell its entire inventory. Lower values are better from the cost perspective; however, minimized stock levels may conflict with steady sales’ delivery capacity and consumer satisfaction in the end. We calculate DIO as an annual ratio of total inventory value on the balance sheet divided through total sales per year. This financial data was based on individual company reports.

Chan et al. (2011) detect a significant impact of well-programmed product design attributes on CS. Inspired by these scholars, we use annual research and development (R&D) expenses in percent of company sales (2005–2009) as a proxy input variable. R&D data was captured from the annual reports of the respective companies in the mentioned years.

In addition to the exogenous variables used by the other scholars, we control for the female component on the board of directors. We build a proprietary data set and collect the company data in order to install several control parameters in this context, such as number of women on the board of directors (FEMALE BOARD COUNT), size of the board (BOARD SIZE), and annual share of female directors (in %) per firm (WBOARD). Alternatively, we include dummy variables for the companies (and years) with a female share of 0%, 1–9%, 10–19%, 20–24%, and 25–33%. There are only 12 observations out of our N = 515 data set with more > 33% female share; hence, we test this interval separately. Table 3 presents a full list of variables and their descriptive statistics. Figure 1 shows that female share in our sample slowly but surely increased over time by 13% and the female count by 10%. In 2005 exactly 13 boards had an exclusive male representation vs. 8 boards in 2009.

Table 3 Descriptive statistics
Full size table
Fig. 1
figure 1

Female directors’ statistics of our sample over time

Full size image

Table 4 shows the correlations of the independent variables. As the variables “female directors count”/ “women on board share”/ “zero women on board” are multicollinear, we use them separately in our estimations.

We use annual data from 2005 to 2009. Our methodological approach is twofold. In the first step, we simultaneously test all variables in an OLS regression as Mittal et al. (2005), implying the following model:

$$\mathrm{y_{\mathrm{iT}}=\alpha+\beta_1{{X}}_1,\;_\text{it}+\dots+\beta_{k}{{X}}_{k},\;_\text{it}+\gamma_{i}{S}_\text{it}+\delta_1{WBOARD}_1,\;_\text{it}+\dots\delta_{e}{WBOARD}_{j},_\text{it}+\varepsilon_\text{it},}$$

where yiT is the customer satisfaction score of firm i at period t. Following Simon and De Varo (2006), we use the logarithms of customer satisfaction values. X1 to Xk and Si are the set of firm level control variables as described above, and the terms WBOARD are the main corporate governance indicators related to gender.

In the second step, we add gender-specific data to test our hypothesis of the positive impact that women on a board of directors have on customers’ satisfaction. The results of step 1 are shown in Table 5, column (1). The output of step 2 is shown in the other columns of Table 5 as well as in Table 6, which controls for potential nonlinear effects of women’s participation. Finally, as a test of robustness, we include the same data in a stepwise regression (Table 7). We are aware of the critics of the stepwise method—such as, “a bias in parameter estimation, inconsistencies among model selection algorithms, an inherent (but often overlooked) problem of multiple hypothesis testing, and an inappropriate focus or reliance on a single best model” (Whittingham et al., 2006, p.1)—and use this “hands-off” approach as a robustness test only.

Table 4 Correlation matrix. Notes: *total headcount/total sales in Mio. US$; **dummy variable; data from 2005 to 2009, N = 515
Full size table
Table 5 Effects of women on the board of directors on customer satisfactionL (OLS) (part I)
Full size table
Table 6 Effects of women on the board of directors on customer satisfactionL (OLS) (part II)
Full size table
Table 7 Effects of women on the board of directors on customer satisfactionL (stepwise LS)
Full size table

Results

Column (1) of Table 5 displays the result of a regression, which simultaneously controls for all variables that have been found to have a significant relationship to CS in earlier studies. The results confirm the findings of Mittal et al. (2005) of a positive and significant association of sales, general, and administrative expenses with CS and a negative significant impact of sales efficiency. We find a positive (but insignificant) effect of advertising expenses. Note that some promotional expenditures are included in the broader SG&A cost category. Marketing activities are uniquely important in the early-stage innovation, hence the phase between R&D and the beginning of the new product development (Schoonmaker et al., 2012). In this estimation model, the cost of goods sold does not have any significant association to CS. Mittal et al. (2005) conducted their regression analysis to ascertain, if the four mentioned inputs were statistically related to satisfaction and if the coefficients go in the assumed direction, which we were able to confirm to some extent.

Furthermore, we observe a positive but insignificant relationship of market share to customer satisfaction. Anderson et al. (1994) find a significantly negative relationship (slope) in the short term, when CS may drop, while company’s market share is increasing, as bigger firms must often serve diverse and heterogenous customers. However, the scholars also discuss a theoretical possibility of a positive influence of higher market share on buyers’ contentedness in the short-term, particularly when an enterprise is specialized in a niche or enjoys great economies of scale and scope. They also theoretically consider a short-term vs. long-term phenomenon, where customer satisfaction may grow in parallel with market share increase also in the long run.

Our estimation (1) also replicates the significant findings of Simon and De Varo (2006), who discover positive relationships of an employee friendly company policy, acknowledged by a “best company status.” In our equation, being a “best company” improves the CS by 3.2 to 4.2%, which is on average 2.4 to 3.1 CS points. Simon and De Varo (2006) estimated an improvement of 2.3 to 3 CS points in their models (excluding firm fixed effects). Similarly, we also find a positive relationship to control variables such as firm’s profitability (expressed by ROA), and organizational size to CS. Like these scholars, we also find a negative significant association to total assets. We also confirm existing industry effects regarding sector variation in customer satisfaction, specifically for the consumer goods industry, utilities, and retail as well as for the technology, media, and telecom industries.

Reflecting upon a positive association between our proxy variable inventory management and customer satisfaction as researched in a correlation analysis by Singh et al. (2010), we confirm a significant positive effect of days inventory outstanding on CS. Finally, in adherence to results from Chan et al. (2011), who emphasize a significant impact of well-programmed product design attributes such as “photo quality” and “take distant image” on CS, we detect a favorable impact of our proxy variable research and development expenditures on CS. Voutsinas et al. (2018) confirm a long-run relationship between R&D expenditure and innovation, too.

In model (2) in Table 5, we add the control variables board size and female count on the board of directors. We find a significantly negative impact of the board size, where one standard deviation increases of board members (three persons) decline the CS by 0.7 to 1% in CS rating (0,76 rating points on avg.). On the contrary (column 2) we find that the absolute number of women directors affects CS significantly positive. The effect is economically small, but statistically significant, where one female board member gives 1% increase in the CS score of a company. Consequently, doubling the average number of female directors on board would result in 4% growth of CS ratings. Accordingly, in column (3), we find a significant positive association of an annual female share on board of directors on CS. An all-female board would improve the CS rating by 12.4%. Or, e.g., doubling the average female share on boards of directors to 33% (4 women) could increase the CS by 4% (hence ca. 3 CS rating points). Simultaneously, we confirm the positive significant effect of SG&A, Best Company, ROA, organizational size, DIO, and R&D expenses on CS. We can also confirm that sales efficiency and total assets can affect it in a significantly negative way. Model (4) in Table 5 shows a disproportionately high impact of the female directors on CS, at least for our sample range of up to 33%. We will return to the issue of nonlinearities.

The models (5) and (6) in Table 5 scrutinize a similar set of variables under the additional consideration of time fixed effects to control for events, possibly affecting all companies (e.g., financial crisis) but not being specific to any particular company. In this case, female directors on board as well as their percentage share also indicate a positive impact on CS. We tested our models also with company fixed effects (results are available on request) and obtained a significant and positive relationship for the female count on the board of directors as well as for the women share on a board. The magnitude of impact of the female board count on CS is however smaller, which suggests that, while women’s presence on boards of directors can affect CS ratings positively, some part of this relationship is driven by unobserved firm characteristics. A similar reduction of magnitude is observable with the female share variable. However, other variables apart from the COGS turn out to be insignificant in this company fixed model specification, potentially indicating a shortcoming of fixed-effects models for panel data (Hill et al., 2020; Bell & Jones, 2015).

The models in Table 6 take up the idea of a nonlinear relationship between female participation and CS due to minimum critical masses found in other business outcomes (Kanter, 1993; Rosener, 1995; Konrad et al., 2008; Torchia et al., 2011; Elstad & Ladegard, 2012; Joecks et al., 2013; Schwartz-Ziv, 2017; Amorelli & Garcia-Sanchez, 2019; Saggese et al., 2020). Column (1) of Table 6 indicates that a complete lack of females on a board of directors does not significantly affect the CS. Column (2) and column (3) show that a positive ratio of women directors of 25–33% implies a 2.5 to 3,2% improvement on the CS score (ca. 2 rating points on average). Also, the lack of a critical mass (below 1/5 of directors) can negatively affect CS performance by 2.1%. Similar results are achieved if controlled for the time fixed effects in columns (4) and (5). Our outcome suggests that a minimum 1/4 of female directors’ ratio permits an effective action on customer satisfaction.

As a robustness test, Table 7 displays the results of stepwise regression analyses. We provide the best three models determined by the forwards stepwise algorithm in columns (1), (2), and (3). The results highlight a significant positive effect of a company’s effective inventory policy; advertising expenditures; “best company” awarded employee policy; firm’s profitability; research and development expenses; sales, general, and administrative expenditures; and organizational size. Significantly negative effects on CS have been indicated for sales efficiency (headcount/per earning) and total assets. Female directors’ shares of more than 1/4 significantly improve CS by 2.1 to 4.9%, while 0 or less than a 25% share may significantly negatively affect it by 2.1 to 4.7%. The downside risk for customer satisfaction of not having enough gender-diverse boards roughly equals the premium for establishing a set quota.

Conclusion and Outlook

We analyze the determinants of customer satisfaction (CS). We start by re-estimating earlier tested variables on CS with our data and add to the literature by testing the variables simultaneously. In a second step, we add variables that control for women´s participation on a board of directors. As a robustness test, we conduct a stepwise regression.

Our study confirms that optimizing certain determinants of organizational performance is key to increasing customer satisfaction, and that gender diversity at the executive level can be an additional lever for improving customer satisfaction. We find a significant positive impact of sales, general, and administrative expenditures to CS, confirming the findings of Mittal et al. (2005). Our findings also confirm Simon and DeVaro (2006), who find that employee-friendly, generously staffed, and profitable companies attract more satisfied customers and who also acknowledge industry variation in consumer satisfaction. Our results reaffirm the importance of the capability to quickly respond to fluctuating demand and optimize inventory management levels (Singh et al., 2010) and that long-term corporate investments in research and innovation can pay off in terms of increased customer satisfaction (cf. Chan et al., 2011). We find that a larger female representation in corporate governance can significantly and positively affect consumer satisfaction. In our linear estimation model, as well as in the quadratic polynomial specification, a larger share of women directors has a significant and positive impact on customer satisfaction. Testing in more detail for further nonlinearities, we find that a critical mass of female board members—beyond one fourth of directors—is needed to favorably influence CS scores. Companies with less than 25% of female board members may experience decreased CS levels.

We acknowledge the limitations of our analysis. As with any empirical study, our results are based on a particular data sample that is specific to the time period chosen, geography, and data availability. Better access to proprietary and consistently measured firm characteristics as well as personnel-related data (e.g., biographical details of board members or detailed customer profiles), which are often protected by privacy policies, could presumably uncover further insights into the relationship between board member gender and customer satisfaction or other firm performance. In addition, employing other measures of gender diversity (e.g., female CEOs/CFOs/COOs) or additional types of diversity (e.g., ethnic) could be interesting for future research. Note, however, that with broader resource inequality data that include diverse data on demographic, leadership, and formal, personnel, and organizational factors, any analysis of gender causality may become (too) complex (Cruz-Castro & Sanz-Menendez, 2019; Mazur & Spierings, 2016).

Taken together, our findings suggest larger implications for the consumer goods or consumer services sector but could also stimulate further studies in other areas (e.g., public services). At first glance, studies like ours might be relevant also for firms with manufacturing operations that are largely absorbed by women. And, in sophisticated and transparent markets where the public is pushing for inclusion, the value of a brand may increasingly depend on a higher proportion of women directors. Further studies could examine how board dynamics and director emancipation affect customer satisfaction, taking into account, for example, specific cultural dimensions. In traditionally patriarchal countries, a shift in the selection process for women on boards from family ties to more performance-based quality criteria could be encouraged. Occasionally, it is argued that female leaders are less “driven by the rules” (Pace, 2009, p. 1), and (Pesonen et al., 2009, p. 337) state, that there is “a distinct possibility that you’d stir things up, disturb their [i.e., men’s] established ways [...].” Thus, an examination of the impact of women’s representation in the executive ranks of companies operating in markets with rapidly changing market trends or in markets with particularly creative and innovative products may be of particular interest.

Our study could not only be a starting point for studies on other sectors. Rather, it could also be a stimulus for other business dimensions, whether for studies or for implementation attempts in practice. For example, it is clear that the human factor is the primary driver of innovation (Martinidis, 2017)—our study could animate attention to the role of women in this regard. Overall, our findings could inspire companies to make further efforts to improve governance, with particular attention to the inclusion of women. It is not unlikely that a better balanced composition of management ranks would not only have a positive impact on customer satisfaction. Rather, it could improve a company’s reputation, its ability to expand into new markets, and its enterprise value.

If a sufficient number of companies in an economy were to achieve a better balanced composition of their management levels (through appropriate regulations or as a result of findings such as ours), this could even have positive consequences for national economic growth.Footnote 6 In this respect, our results are not only of managerial importance, but of macroeconomic importance, implying a potential role for policy makers and regulators. Our nuanced findings on the positive effects of increased female participation on boards, particularly a breaking point of 25% of board seats, should be able to help policymakers develop effective strategies to achieve a critical mass of female representation in the corporate world and beyond.

Another key question remains whether there are other ways of transmitting empirical evidence like ours on the positive effects of increased women’s participation in economic life outside of policy or regulatory requirements. In this context, the finding that nongovernmental organizations can play particular roles in bridging the gender gap may be interesting (Lindberg et al., 2014).

In sum, our study results may encourage more targeted gender research in business and economics, helping legislators around the world to enact balanced laws that promote work-life balance, free up infrastructure spending for women’s education, and reduce gender discrimination to eliminate the “glass ceiling” in leadership positions.