Resource dependence theory (Pfeffer and Salancik, 1978) provides a robust framework for understanding how board composition contributes to firm performance by facilitating access to critical external resources, with particular emphasis on the role of female directors on boards in this study. According to Hillman and Dalziel (2003), the board of directors serves not only as a monitor of management but also as a provider of key resources, such as advice, legitimacy, and networks. The effectiveness of this resource provision function depends on the board’s collective capital, encompassing both human capital, such as expertise and experience, and relational capital, including external networks and stakeholder ties.

In this context, female directors contribute unique professional experience, networks and stakeholder relationships, expanding the firm’s access to new business opportunities and enhancing its legitimacy, especially in environments where sustainable management and stakeholder engagement are highly valued (Kim, 2023). In the Korean setting, where traditional networks have historically limited women’s access to influential circles, the appointment of female directors signals a commitment to openness and modernization, thereby strengthening the firm’s legitimacy among investors, regulators and customers.

While upper echelons theory (Hambrick and Mason, 1984) explains how directors’ demographic and psychological traits shape internal decision-making, resource dependence theory emphasizes the board’s critical function in acquiring essential resources from the external environment. Female directors, as shown in prior research, tend to be more risk-averse and less overconfident compared to their male counterparts, leading to more prudent risk management and a reduction in overinvestment (Barber and Odean, 2001; Bernasek and Shwiff, 2001). Their presence broadens the diversity of perspectives within the board, enriches advisory capacity, and elevates the quality of deliberations (Gul et al., 2011; Kim and Starks, 2016). Female directors are also associated with higher attendance rates, more detailed information sharing, and a greater propensity to challenge management, thereby increasing accountability (Adams and Ferreira, 2009).

Moreover, female directors enhance the board’s advisory function by introducing perspectives absent in homogeneous groups, potentially leading to superior product development and marketing strategies (Harjoto et al., 2015). Overall, the inclusion of female directors not only improves governance and risk management but also enhances the board’s ability to innovate and respond to diverse stakeholder interests.

However, the inclusion of female directors on corporate boards does not always yield uniformly positive outcomes. Some research suggests that increasing gender diversity can introduce challenges to board dynamics, such as communication barriers, reduced cohesion, and a higher likelihood of conflict among members. Social identity theory (Williams and O’Reilly, 1998) and similarity-attraction theory (Jehn et al., 1999) share similar perspectives, positing that people categorize others into in-groups that share similar attributes and out-groups, those perceived as different. These theories suggest that such categorizations can reduce communication, weaken cooperation, and heighten conflict, ultimately undermining group processes and outcomes.

Second, gender tokenism theory suggests that increasing gender diversity on corporate boards does not necessarily translate into substantive influence in decision-making processes (Baik et al., 2024). Symbolic inclusion of female directors remains nominal rather than substantive, due to entrenched power structures favoring majority demographics. Westphal and Milton (2000) argue that without achieving critical mass, gender-diverse boards may exhibit no statistically significant improvement in investment decision quality compared to homogeneous boards. In other words, a critical mass of female directors is necessary for their voices to be heard and their perspectives incorporated into strategic decision-making.

Third, Confucian values offer a framework for explaining the role of female directors in South Korea. Yoon (2019) highlights how Confucian values can significantly impact the participation and influence of women in decision-making processes. The inherent emphasis on hierarchy and deference to authority within Confucianism can create a challenging environment for female directors, silencing their voices and hindering their upward mobility. At the same time, Horak and Suseno (2023) document how informal networks in Korean society systematically exclude women from decision-making, a practice rooted in Confucian values. Additionally, such informal networks are pervasive and often too entrenched for women to penetrate. This social exclusion is aggravated by traditional gender stereotypes, which undervalue their contributions and restrict their access to leadership roles.

The Modigliani and Miller (1958) theorem asserts that, under the assumption of a perfect capital market, a firm’s investment, financing, and dividend decisions are independent and do not affect firm value. However, real-world capital markets deviate significantly from this ideal. Market imperfections, primarily agency conflicts and information asymmetry between managers and external stakeholders, create frictions that can lead to suboptimal investment decisions, manifesting as both underinvestment and overinvestment (Dewi et al., 2020).

Agency theory, as developed by Jensen and Meckling (1976), provides a foundational framework for understanding these inefficiencies. Conflicts of interest arise because managers (agents) may prioritize their own utility over maximizing shareholder (principal) value. This divergence can lead to overinvestment, where managers utilize free cash flow to undertake projects with negative net present value (NPV) simply to expand the firm under their control, as highlighted by Jensen’s (1986) free cash flow hypothesis. Information asymmetry exacerbates this by enabling managers to misrepresent the quality of investment projects, resulting in poor projects being funded, while valuable ones are overlooked (Wang and Nishihara, 2025).

Conversely, agency problems and information asymmetry can also result in underinvestment. Myers (1977) first identified the debt overhang problem, in which highly leveraged firms may forgo positive NPV projects because a significant portion of the project’s returns would accrue to debtholders rather than equity holders, thus diminishing the incentive for shareholders to invest. Furthermore, Myers and Majluf (1984) demonstrated that information asymmetry regarding the firm’s true value can make external financing costly. If managers believe the firm’s equity is undervalued by the market, they may be reluctant to issue new shares, fearing dilution of existing shareholders’ value. This adverse selection problem increases the effective cost of external capital, potentially leading the firm to pass up valuable investment opportunities that could have been financed with internal funds. Empirical evidence supports the notion that higher information asymmetry generally leads to lower investment efficiency.

Research recognizes that market imperfections can lead to inefficiencies, highlighting the critical role of corporate governance mechanisms, especially board characteristics, in mitigating agency problems and enhancing investment efficiency. Several board traits have been examined for their influence on investment decisions. For instance, board size is considered a factor, with a prior study suggesting larger boards can improve investment efficiency through enhanced monitoring capabilities or broader expertise (Choi et al., 2013). Board independence, measured by the proportion of independent directors, is another crucial characteristic that strengthens oversight (Kamarudin et al., 2024). Chief Executive officer (CEO) duality, where the CEO also serves as the board chair, has also been studied for its impact on board effectiveness and subsequent investment decisions (Choi et al., 2024; Duru et al., 2016). Ownership structure, including managerial and institutional ownership, interacts with board traits to influence investment efficiency, affecting alignment of interests and monitoring intensity (Choi et al., 2024; Kim and Kim, 2025).

The influence of female directors on investment efficiency in Korean firms should be considered within a complex institutional and cultural context. Traditionally, Korean society has been shaped by Confucian values and a strong preference for sons, which have historically restricted women’s participation in economic and corporate spheres. These deeply embedded norms have resulted in persistently low representation of women on corporate boards, with many firms lacking female directors altogether.

In response to these structural barriers, a significant regulatory intervention was enacted in 2020 with an amendment to the Capital Markets Act, mandating the appointment of female directors in large publicly listed companies. This legislative change represents a pivotal step toward greater board diversity and inclusion in Korea, providing a unique setting to examine the effects of gender diversity on firm outcomes.

Resource dependence theory offers a compelling framework for understanding the potential impact of female directors in this context. It posits that boards are vital conduits for external resources as well as legitimacy—both essential for organizational survival and growth (Hillman and Dalziel, 2003). Female directors, by virtue of their diverse backgrounds and stakeholder relationships, can expand a firm’s access to new business opportunities and enhance its legitimacy, particularly in settings where sustainable management and stakeholder engagement are increasingly valued (Kim, 2023). In Korea, where traditional networks have often excluded women from influential circles, the presence of female directors signals a commitment to openness and modernization, thereby strengthening the firm’s reputation among investors, regulators, and stakeholders. Recent empirical research underscores both the advantages and challenges associated with board gender diversity. On the positive side, female directors have been found to reinforce the board’s monitoring and advisory functions, promote greater transparency and improve the quality of financial reporting (Barua et al., 2010; Francis et al., 2015; Gul et al., 2011). Their risk-averse and prudent decision-making styles can help curb managerial overinvestment and promote more disciplined resource allocation, thereby enhancing investment efficiency (Barber and Odean, 2001; Bernasek and Shwiff, 2001; Baik et al., 2024). Moreover, gender-diverse boards are associated with higher attendance rates, more rigorous information sharing, and a greater willingness to challenge management, all of which contribute to stronger governance and better firm performance (Adams and Ferreira, 2009; Kim and Starks, 2016).

However, several studies caution that the benefits of female board representation are not automatic. In contexts where women remain a small minority, tokenism can limit their substantive influence, and entrenched male-dominated networks may exclude them from informal decision-making processes (Westphal and Milton, 2000; Joecks et al., 2013). Social identity theory suggests that minority directors may be perceived as outsiders, reducing communication and cooperation within the board (Williams and O’Reilly, 1998; Horak and Suseno, 2023). In Korea, these dynamics are exacerbated by Confucian norms prioritizing hierarchy and gendered loyalty, which can further marginalize female directors and limit their impact on strategic decisions (Yoon, 2019; Yang, 2012). Without achieving a critical mass, their contributions may be overlooked or marginalized, diminishing the potential positive effects on investment efficiency.

Taken together, the relationship between the proportion of female directors and investment efficiency in Korean firms is nuanced and context-dependent. When the advantages of board gender diversity are realized, investment efficiency is likely to improve. Conversely, when cultural and structural barriers persist, the impact may be neutral or even negative. Thus, the proportion of female directors on corporate boards is associated with the extent to which female directors can exercise substantive influence, access critical resources, and overcome cultural and structural barriers within the organization. This perspective, grounded in resource dependence theory and recent empirical research, provides a comprehensive foundation for examining the conditional impact of female directors on investment efficiency in the Korean corporate context. Considering these conflicting theoretical perspectives and the specific Korean context, this study proposes the following hypothesis:

Hypothesis 1: The presence of female directors on the board negatively impacts investment efficiency in Korean firms.

Despite significant regulatory reform, such as the 2020 amendment to the Capital Markets Act mandating female representation on boards, Korea’s corporate environment remains deeply influenced by Confucian values. These cultural legacies manifest in persistent gender hierarchies, male-dominated networks, and expectations that women prioritize domestic roles over professional advancement (The Korea Times, 2025). Such norms have historically limited women’s access to leadership positions and constrained their substantive influence within corporate governance.

During periods of economic strain, particularly when cost stickiness is present, the relationship between female directors on board and investment efficiency becomes especially pertinent. Cost stickiness describes a situation where costs do not decrease proportionally with declining revenue, often due to managerial reluctance to reduce resources, the presence of adjustment costs, or agency-related constraints (Homburg and Nasev, 2008). This asymmetry in cost behavior can give rise to inefficiencies, such as overinvestment during financial distress, as firms struggle to align expenses with declining revenue in a timely and efficient manner (Banker et al., 2016). Effectively addressing these inefficiencies demands rigorous resource management and cautious decision-making.

Behavioral and governance research suggests that female directors possess distinctive decision-making qualities, such as heightened risk aversion, conservatism, and a focus on long-term stability, which are particularly effective in mitigating inefficiencies exacerbated during economic crises. Empirical evidence demonstrates that firms with greater female board representation or female CEOs have outperformed their counterparts during periods of crisis. For instance, Thornton and Vasilakis (2024) found that firms led by female CEOs achieved higher profitability, stronger valuations, and lower risk-taking during the COVID-19 pandemic, reflecting a preference for stability and sustainable growth under uncertainty. Similarly, Sinha (2023) reported that German firms with female board representation delivered superior stock returns and risk profiles during the pandemic, although these advantages diminished after the crisis.

In environments characterized by cost stickiness, where rigid cost structures exacerbate financial pressure, the behavioral attributes of female directors, such as caution and disciplined capital allocation, can counteract managerial optimism or delayed cost adjustments, which often lead to inefficiencies. Evidence further indicates that female leaders are adept at reducing investment volatility and aligning expenditures with realistic revenue expectations, particularly under adverse economic conditions. Collectively, these findings support the proposition that female directors enhance investment efficiency by restraining overinvestment and ensuring that capital allocation remains disciplined during periods of economic stress. Based on these considerations, we propose the following hypothesis:

Hypothesis 2: Under conditions of cost stickiness, firms with greater female board representation exhibit improved investment efficiency.

Table 1 describes the data selection process. The sample for this study comprises Korean firms listed on the Korea Composite Stock Price Index (KOSPI) and the Korea Securities Dealers Automated Quotation (KOSDAQ) from 2014 to 2021, with financial data drawn from the Value Search database (https://www.nicevse.com/vse/main.html). The selection of the 2014–2021 period is grounded in both regulatory developments and data reliability. Specifically, 2014 marks the first year for which comprehensive data on female directors on boards became available. In addition, the 2020 amendments to the Capital Markets Act prohibit all-male boards in companies with total assets over USD 1.44 billion; this provision came into effect in 2022. This study spans both the period before the amendment and the transitional phase. To ensure data consistency, companies in the financial industry are excluded, since they have distinct regulatory and financial frameworks compared to non-financial companies. Additionally, we removed observations with missing financial data. After these adjustments, our final sample comprises 10,265 firm-year observations. Furthermore, all variables were winsorized at the 1% and 99% levels to mitigate the influence of extreme outliers and enhance the robustness of our results.

To verify the first hypothesis—that female directors serving on boards affect investment efficiency, we employed the ordinary least squares (OLS) regression model in Eqn. 3:

(3)$${\text{IE}}_{\mathrm{t}}={\beta }_0+{\beta }_1{\text{Female}}_{\mathrm{t}}+{\beta }_2{\text{Size}}_{\mathrm{t}}+{\beta }_3{\text{Lev}}_{\mathrm{t}}+{\beta }_4{\text{Roa}}_{\mathrm{t}}+{\beta }_5{\text{Growth}}_{\mathrm{t}}+{\beta }_6{\text{Loss}}_{\mathrm{t}}+{\beta }_7{\text{Ocf}}_{\mathrm{t}}+{\beta }_8{\text{Beta}}_{\mathrm{t}}+{\beta }_9{\text{Vol}}_{\mathrm{t}}+{\beta }_{10}{\text{Zscore}}_{\mathrm{t}}+{\beta }_{11}{\text{Age}}_{\mathrm{t}}+{\beta }_{12}{\text{For}}_{\mathrm{t}}+{\beta }_{13}{\text{Mkt}}_{\mathrm{t}}+{\beta }_{14}{\text{Da}}_{\mathrm{t}}+\text{YearD}+\text{IndustryD}+\epsilon$$

where, IE = investment efficiency, calculated as the absolute value of residuals from Eqn. 1; Female = ratio of female directors serving boards, calculated as the number of female directors divided by the total board members; Size = natural logarithm of total assets; Lev = total debt divided by total assets; Roa = net income divided by total assets; Growth = change in total assets for the current year relative to the previous year divided by total assets for the current year; Loss = 1 if a company incurs a loss, 0 otherwise; Ocf = cash flow from operations divided by total assets; Beta = systematic risk; Vol = stock return volatility; Zscore = Altman’s z-score, calculated as 1.2 (Working Capital/Total Assets) + 1.4 (Retained Earnings/Total Assets) + 3.3 (Earnings Before Interest and Taxes (EBIT)/Total Assets) + 0.6 (Market Value of Equity/Total Liabilities) + 1.0 (Sales/Total Assets); Age = the firm’s age from the year it was listed on the Korean stock market to the current year; For = ratio of foreign investors; Mkt = dummy variable equal to 1 if the firm is listed on KOSDAQ, 0 if listed on KOSPI; YearD = year dummies; industryD = industry dummies; Da = Discretionary accruals measured by the model of Kothari et al. (2005), described in Eqn. 4.

(4)$${\mathrm{Ta}}_{\mathrm{t}}/{\mathrm{A}}_{\mathrm{t}}={\alpha }_0+{\beta }_1\left(1/{\mathrm{A}}_{\mathrm{t}}\right)+{\beta }_2(\mathrm{△}\text{Sales}-\mathrm{△}\mathrm{Ar})/{\mathrm{A}}_{\mathrm{t}}+{\beta }_3{\mathrm{Ppe}}_{\mathrm{t}}/{\mathrm{A}}_{\mathrm{t}}+{\beta }_4{\mathrm{Roa}}_{\mathrm{t}}+{\epsilon }_{\mathrm{t}}$$

Where, Ta = Net income – cash flow from operations; A = Total assets; Sales = Sales revenue; Ar = Accounts receivable; Ppe = Plant, property, and equipment; Roa = Net income/total assets.

Da represents discretionary accruals, which is used as a proxy for earnings management. A higher value may indicate lower earnings quality, which can distort managerial decision-making and lead to inefficient investment (Linck et al., 2013). It is calculated in accordance with Kothari et al. (2005), estimating industry-year effects for the firms in the sample.

In Eqn. 3, Female is the independent variable and represents the ratio of female directors serving on boards. IE is the dependent variable, representing investment efficiency. For the control variables, Size, is included. According to Anderson et al. (2003), higher asset concentration leads to increased adjustment costs for management, as greater asset concentration amplifies the costs associated with asymmetry in resource reduction. Lev is included in this model, as it influences the spending behavior related to discretionary costs (Bhagat and Welch, 1995). Roa denotes return on assets, indicating a firm’s profitability and operational efficiency. Firms with higher Roa signal that they generate greater returns from their asset base, thereby enhancing their capacity to engage in proactive investment activities (Kwon, 2023). Growth is a proxy for growth opportunity. Firms with higher asset growth are generally considered to possess more investment opportunities and thus are more likely to allocate resource efficiently (Li and Chen, 2022). Loss is a dummy variable, coded as 1 for a loss-making firm, 0 otherwise. Firms experiencing losses may adopt a more conservative approach in their investment decisions, potentially leading to lower investment efficiency (Kim and Kim, 2025). Ocf, defined as the ratio of cash flow from operations to total assets, reflects the proportion of earnings generated from core business activities. Accordingly, Ocf is included as a control variable to account for its potential influence on cost behavior and investment efficiency (Chang and Paik, 2009). Beta indicates a firm’s exposure to systematic risk, reflecting each firm’s equity returns relative to the market. A higher value of beta implies greater responsiveness to market movements, which in turn affects both risk and expected returns on investment. Vol represents the volatility of a firm’s stock returns. Higher volatility means that the firm’s stock price is unpredictable, implying a firm’s uncertainty and level of risk. Higher stock return volatility is associated with lower investment efficiency, as uncertainty increases the difficulty of optimal capital allocation (Zhou et al., 2024). Altman’s Z-score is included in this model, as it measures a firm’s financial health and bankruptcy risk. It is computed as the weighted sum of working capital, retained earnings, earnings before interest and taxes, market value of equity, and sales. A lower value of Z-score indicates a higher probability of financial distress, whereas a higher value implies greater financial stability. For represents the proportion of shares held by foreign investors in a firm. Prior research demonstrates that foreign investors possess superior information acquisition and analytical capabilities, which can help constrain managerial tendencies toward inefficient decisions. Age refers to the number of years since the firm’s initial listing on the Korean Stock Exchange. As firms mature and accumulate operational experience, their propensity for inefficient investment is expected to decline. Mkt is a dummy variable indicating the segment of the Korean stock market on which a firm is listed. The Korean stock market comprises two primary segments: KOSPI and KOSDAQ, each characterized by distinct economic and regulatory environments. Firms listed on KOSPI are generally larger, exhibit greater managerial stability, and have higher access to foreign investors. In contrast, KOSDAQ-listed firms tend to be smaller in size but are typically more oriented toward growth and innovation. This market distinction captures important heterogeneity in firm characteristics that may influence investment behavior and efficiency.

To evaluate how female directors influence investment efficiency under conditions of cost stickiness, we extend the baseline model by incorporating an interaction term between female board representation (Female) and cost stickiness (CS) in Eqn. 5:

(5)$${\text{IE}}_{\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{Female}}_{\mathrm{t}}+{\beta }_2{\mathrm{CS}}_{\mathrm{t}}+{\beta }_3{\mathrm{FC}}_{\mathrm{t}}+{\beta }_4{\text{Size}}_{\mathrm{t}}+{\beta }_5{\text{Lev}}_{\mathrm{t}}+{\beta }_6{\text{Roa}}_{\mathrm{t}}+{\beta }_7{\text{Growth}}_{\mathrm{t}}+{\beta }_8{\text{Loss}}_{\mathrm{t}}+{\beta }_9{\text{Ocf}}_{\mathrm{t}}+{\beta }_{10}{\text{Beta}}_{\mathrm{t}}+{\beta }_{11}{\text{Vol}}_{\mathrm{t}}+{\beta }_{12}{\text{Zscore}}_{\mathrm{t}}+{\beta }_{13}{\text{Age}}_{\mathrm{t}}+{\beta }_{14}{\text{For}}_{\mathrm{t}}+{\beta }_{15}{\text{Mkt}}_{\mathrm{t}}+{\beta }_{16}{\text{Da}}_{\mathrm{t}}+\text{YearD}+\text{IndustryD}+\epsilon$$

where, CS = measure of cost stickiness proposed by Homburg and Nasev (2008), described in section 3.2.2; FC = the interaction term between cost stickiness and female directors serving on boards; see Eqn. 3 for definition of other variables.

Table 2 provides the descriptive statistics for the key variables. The first variable, investment efficiency (IE), exhibits a mean value of 0.374 and a standard deviation of 0.300. The median value of IE is 0.276, which is lower than the mean. The second variable, the ratio of female directors serving on boards, demonstrates a mean of 0.044 with a standard deviation (Std) of 0.090. Meanwhile, Cost Stickiness (CS) has a mean value of 4.492 and a standard deviation of 1.844, with Q1, the median, and the third quartile (Q3) all consistently recorded at 5.000.

Table 3 presents the correlation coefficients among the key variables. The correlation coefficient between investment efficiency and female representation is 0.039, indicating a positive association. The correlation coefficient between female representation and cost stickiness (CS) is 0.011, also showing a positive association, but this association is not statistically significant (p-value = 0.237). Lastly, the correlation coefficient between IE and CS is –0.003. It is worth noting that while a positive correlation is observed between female representation and IE, this result does not control for other explanatory variables. Thus, the observed association might change when additional control variables are taken into account, highlighting the need for further analysis.

Table 4 presents the results of testing the first hypothesis using Eqn. 3, which examines the relationship between the proportion of female directors on the board and investment efficiency, with the coefficient for the female variable being statistically significant at the 10% level. The positive coefficient implies that a higher proportion of female directors is associated with reduced investment efficiency.

This finding is consistent with recent literature highlighting the complex and sometimes limited impact of board gender diversity on firm outcomes. For example, Adams and Ferreira (2009) and Sila et al. (2016) suggest that while the presence of female directors can enhance board monitoring functions, their actual influence may be constrained by organizational culture and prevailing social norms. In the Korean context, characterized by hierarchical and male-dominated corporate governance structures, female directors often face barriers to meaningful participation in decision-making (Yoon, 2019). This can result in the appointment of female directors being more symbolic than substantive, limiting their ability to affect strategic outcomes such as investment decisions. Recent study provides evidence that the effectiveness of board gender diversity depends not only on the numerical representation of women but also on the existence of a supportive organizational culture and the attainment of a critical mass enabling female directors to exercise real influence (Joecks et al., 2023).

Table 5 presents results on how the presence of female directors on boards affects investment efficiency under cost stickiness, where costs do not decrease proportionally with declining sales, often reflecting economic stress or rigid cost structures. While the results indicate that the coefficient for the proportion of female directors is positive, the interaction term between female directors (Female) and cost stickiness (CS) is negative and significant at the 5% level. This finding reveals a more nuanced dynamic: while female board representation alone may not enhance investment efficiency, its effect becomes notably positive when firms face cost stickiness. In detail, under economic stress, a higher proportion of female directors improves investment efficiency.

This finding is consistent with recent empirical studies that underscore the distinctive strengths of female directors in crisis or highly stressful environments. Thornton and Vasilakis (2024) found that firms led by female CEOs outperformed their counterparts during the COVID-19 pandemic, achieving higher profitability and demonstrating more disciplined resource allocation. Similarly, Sinha (2023) found that female directors serving on boards is associated with superior risk management and financial performance during economically stressful periods. Taken together, the positive impact of female directors on investment efficiency becomes most apparent when firms experience economic adversity, highlighting the strategic importance of fostering meaningful gender diversity in corporate governance.