There are a growing number of females who participate in business activities, and female entrepreneurship has significantly increased and has shown a characteristic advantage in business success (Moudrý and Thaichon, 2020). Females have the traditional gender advantage of being more careful and skilled at communicating and delivering information on time (Bravo and Alcaide-Ruiz, 2019; Pucheta-Martínez et al., 2016). Most female leaders are risk averse and avoid taking excessive risks (Faccio et al., 2016; Palvia et al., 2015; Hoang et al., 2019). Most of the time, risk is a negative factor in management and corporate finance, particularly if the risk is inappropriate for a firm’s development and strategies. However, risk is associated with return and growth. Without taking risks, returns are lower and the business becomes unsustainable if profits are too low. It is more important to estimate and manage a firm’s current risk in an overall and managerial sense rather than treating it individually or completely avoiding it (Nocco and Stulz, 2022). Without appropriate management, risk accumulates and becomes an even larger problem. Risk should be measured by accrual collection, earnings volatility, market competition and other potential events that could affect a firm’s stakeholders and interests (Bhuiyan et al., 2021; Settembre-Blundo et al., 2021). Many investors believe that female managers represent a good solution for alleviating these problems. Female managers tend to make more transparent disclosures, which reduces potential information asymmetry and increases the trust of investors (Abad et al., 2017; Post et al., 2019). Female managers could offer more than a purely monitoring role. Most firms with a female director leading the board place greater emphasis on stakeholder relationships. A female leader’s ability to communicate better helps to organize internal and external relationships and increases the firm’s efficiency as an organization (Hanousek et al., 2019). Additionally, female leaders care more about the environment and place greater emphasis on environmental protection and business sustainability (Birindelli et al., 2019; Glass et al., 2016). They tend to follow business ethics codes and better follow regulations and industrial policies. Female business leaders can promote changes in social roles and greater social inclusion, and in many developing countries, they can motivate more women to engage in more social activities rather than only focusing on family to increase gender equality.

To increase their competitiveness, manufacturing firms must have unique products that are well recognized by consumers and that are difficult for competitors to copy. The firm needs to invest in research and development to design and produce such products. Research and development investments are believed to be high risk since the outcome is highly uncertain (Blazenko and Yeung, 2015; Gu, 2016), but research and development could increase future profitability and motivate competition, which could significantly benefit technological improvements (Davcik and Sharma, 2015). Business leaders may need to balance between current profitability and investment in future opportunities. Excessive research and development could negatively impact current earnings and shareholder wealth, but too little research and development would cause the firm to lose future competition, and the possible negative future impact could also be reflected in the current share price and result in disappointing shareholders. Research and development require a stable social and political environment (Alam et al., 2019; Bhattacharya et al., 2017). Additionally, the firm needs to know the shortcomings of the current products and the functions and features of the products to meet the demand of the market and consumers (Yenipazarli, 2019). Innovation also needs to be balanced with product features and the consideration of environmental protection. Most industries are now seeking green innovation; for example, banking and financial services have developed online service platforms on which they distribute statements. These online services are paperless and provide a higher security level than sending actual mail to account beneficiaries does. The automobile industry promotes hybrid and electrical cars to provide a greater gas economy and more environmentally friendly products. Additionally, even with traditional cars, all makers have attempted to redesign engines to achieve higher gas efficiency. These innovations meet consumers’ requirements and use technologies to benefit the earth, receive consumers’ respect and trust and improve business sustainability both in terms of recognition and the environment (Li et al., 2019). Female leaders are more risk averse, which could reduce their risk-taking incentives, but such lower discovery incentives could increase the future failure rate (Rey-Martí et al., 2015). Risk-averse characteristics can negatively affect female leaders, limiting their capacity as innovators (Sundermeier, 2024), but the monitoring effect of female leaders is well known, and cooperation between female and male leaders can foster business innovation while maintaining appropriate risk-taking.

On the basis of the above logic, we propose the following hypotheses.

H1a. A firm led by a male chairperson and male general manager would have a higher incentive to invest in research and development.

H1b. The involvement of a female as the chairperson or general manager reduces research and development investment.

Since the chairperson and manager may communicate before they make decisions, their wealth in the firm could affect decision-making. If the chairperson and general manager are not confident about the decision, their wealth in the firm could reduce their risk-taking. However, if they are confident about their judgment and believe that the decision is prudential, the wealth effect could encourage risk-taking.

H2a. The firm shares held by the chairperson and general manager could reduce risk-taking incentives when both the chairperson and general manager are males.

H2b. The firm shares held by the chairperson and general manager could increase risk-taking incentives when the general manager is female and the chairperson is male.

Finally, to be confident, female participation should improve research quality and outperform nonfemale participation by increasing the level of research success.

H3. Female participation increases the number of patents, resulting in higher risk management and research quality.

Agency costs and conflicts of interest are potential factors affecting a firm’s management and performance stability. At present, the common management instrument used to align the interests of general managers is pay performance contracts, which connect a firm’s performance to the manager’s income (Gao and Li, 2015). However, such pay performance sensitivity motivates managers to manage earnings, and it negatively contributes to earnings qualities, particularly in emerging markets where investor protection is weak (Almadi and Lazic, 2016). There is evidence that a female chairperson or manager could alleviate manipulation in instances of pay-performance general manager contracts (Usman et al., 2019). Another solution would be to reward general managers with shares of the business. When managers and shareholders hold a significant proportion of their personal wealth in the firm’s shares, their interests are highly aligned with the firm’s value, and any significant change in share prices affects their personal wealth, leading both parties to be more concerned with shareholder interest, particularly under systemic risk (Haider and Fang, 2018). Notably, since female leaders are risk averse, they may be less willing to hold firm equity and instead require higher cash wages (Carter et al., 2017); such characteristics could increase the level of cautiousness when they make decisions in uncertain environments.

H4a. Higher shares held by the chairperson and general manager reduce risk-taking under systemic risk if they are less confident about the decision.

H4b. Higher shares held by the chairperson and general manager increase risk-taking even under system shock if they are confident about the decision.

This study involves manufacturing firms listed on Chinese stock exchanges. The sample period is from 2013 to 2023. Detailed information about the chairperson and the general manager is collected from the China Stock Market and Accounting Research Database (CSMAR). Financial information at the firm level is collected from the choice database and merged with the chairperson and general manager information. The final imbalanced panel data include 20,352 firm-year observations. The variable definitions are listed in Table 1. The variable treatment column shows the estimation method or dummy variable definition. The general statistics are presented in Table 2. Table 2 shows that the number of female business leaders is much smaller than the number of male chairpersons and male general managers; hence, there is a great need to promote the participation of female business leaders in business activities and their contribution to social improvement.

The first set of tests focuses on the effects of gender on research and development investments. Eqn. 1 reflects such baseline models. The term “Comb” indicates the four different gender combinations of the chairperson and general manager. The key objective is to observe the coefficient of the male chairperson and male general manager. A positive coefficient indicates that when both the chairperson and the general manager are male, they are more motivated to make research and development investments, indicating greater risk-taking for future exploration. Additionally, if the coefficient in the case of a female chairperson and female general manager is negative, it indicates that females are risk averse.

(1)$${\mathrm{RD}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

Furthermore, in Eqns. 2,3, the combined the chairperson and general manager gender scenario is interacted with the value of the chairperson’s shareholding and the value of the general manager’s shareholding to determine whether the value of their wealth in the firm enhances or weakens the original gender effect observed in the baseline model.

(2)$${\mathrm{RD}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐶ℎ𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_8\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝐶ℎ𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}\right]+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}$$ $$+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(3)$${\mathrm{RD}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐺𝑚𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_8\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝐺𝑚𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}\right]+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}$$ $$+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

Since scientific research investment and research success can differ, some blind investments may not be efficient. If females are more risk averse, their prudential decision-making may improve research efficiency and increase the number of patents obtained by the firm, resulting in successful research results. Eqn. 4 tests such a relationship. Here, the interest variable “Gender” indicates the two same-gender combinations: both males or both females. It also includes the percentage of females on the management team and board to reflect the female prudential effect on the corporate decision-making process.

(4)$${\mathrm{𝑃𝑎𝑡𝑒𝑛𝑡𝑠}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐺𝑒𝑛𝑑𝑒𝑟}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

The third set of tests, as shown by Eqns. 5,6, demonstrate whether females are better recognized as playing monitoring roles in the business. Eqn. 5 is a logit regression; it is expected that the term “Female” will have a negative coefficient, so female involvement would lower the likelihood of duality. In Eqn. 6, the interaction term is expected to have a positive coefficient, indicating that even if duality occurs, a female chairperson who is also the general manager would alleviate the negative impact of duality and attract institutional investors.

(5)$${\mathrm{𝐷𝑢𝑎𝑙𝑖𝑡𝑦}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐹𝑒𝑚𝑎𝑙𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(6)$${\mathrm{𝐼𝑛𝑠𝑡}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐹𝑒𝑚𝑎𝑙𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝐷𝑢𝑎𝑙𝑖𝑡𝑦}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_8\left[{\mathrm{𝐹𝑒𝑚𝑎𝑙𝑒}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝐷𝑢𝑎𝑙𝑖𝑡𝑦}}_{\mathrm{i},\mathrm{t}}\right]+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}$$ $$+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

Furthermore, we test the effect of female involvement as the chairperson on the alpha of stock returns in a volatile market. Eqn. 7 reflects such a relationship. The interaction term is expected to have a positive coefficient when the chairperson is female, indicating that a female chairperson is recognized by the market as playing a better monitoring role in highly volatile markets, which enhances share price performance.

(7)$${\mathrm{𝐴𝑙𝑝ℎ𝑎}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝑉𝑜𝑙𝑎𝑡𝑖𝑙𝑖𝑡𝑦}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_4{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_5{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_8\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝑉𝑜𝑙𝑎𝑡𝑖𝑙𝑖𝑡𝑦}}_{\mathrm{i},\mathrm{t}}\right]+{\displaystyle \sum \mathrm{𝐼𝑁𝐷}}$$ $$+{\displaystyle \sum \mathrm{𝑌𝑒𝑎𝑟}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

The last set of tests is used to test the fourth hypothesis by examining the effects of gender and its interaction with shareholdings during a systemic crisis. The key observation is whether shareholding reduces risky research and development investments during a systemic crisis. Eqns. 8,9 use the difference-in-differences method to test these relationships. The interesting variable is the triple interaction term, and a higher shareholding value is expected to reduce aggressive investment in research and development when both the market and economics are uncertain, particularly when both the chairperson and general manager positions are held by males.

(8)$${\mathrm{RD}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐶ℎ𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_8{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_9\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}\right]+{\beta }_{10}\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝐶ℎ𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}\right]$$ $$+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(9)$${\mathrm{RD}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{𝐺𝑚𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_4{\mathrm{𝐿𝑖𝑎𝑏}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_5{\mathrm{𝑆𝑖𝑧𝑒}}_{\mathrm{i},\mathrm{t}}+{\beta }_6{\mathrm{𝐶𝑢𝑟𝑟𝑒𝑛𝑡}}_{\mathrm{i},\mathrm{t}}+{\beta }_7{\mathrm{𝐵𝑃𝑆}}_{\mathrm{i},\mathrm{t}}+{\beta }_8{\mathrm{𝐶𝑎𝑡𝑢𝑟𝑛}}_{\mathrm{i},\mathrm{t}}$$ $$+{\beta }_9\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}\right]+{\beta }_{10}\left[{\mathrm{𝐶𝑜𝑚𝑏}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝑃𝑜𝑠𝑡}}_{\mathrm{i},\mathrm{t}}*{\mathrm{𝐺𝑚𝑣𝑎𝑙𝑢𝑒}}_{\mathrm{i},\mathrm{t}}\right]$$ $$+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

For robustness and to alleviate the endogeneity problem, the propensity score matching (PSM) method is then used to treat each gender combination as a treatment group, and the size of the firm, the top ten ownership and the return on assets (ROA) are used as the matching criteria to retest the baseline model. Furthermore, we remeasure the gender effect by considering a female chairperson and a female general manager as two separate cases to compare the results of the impact on research and development investments with those of the baseline model. We also use the instrumental variable approach to demonstrate that females are more risk averse; however, a female general manager and male chairperson increase risk taking when the general manager or chairperson hold large amounts of shares, indicating confidence in their research investment decisions.

Table 3 shows the results of the baseline model. Our results show that when the chairperson or the general manager is female, the female’s risk aversion may negatively affect research and development investments. When only males hold the chairperson and general manager positions, they are more motivated to take risks. When both the chairperson and the general manager are female, the risk-taking incentive is significantly lower, demonstrating greater prudence. When the two positions are held by individuals of different genders, the research and development incentive remains moderate and is not significantly positive or negative.

If the chairperson and manager have significant wealth in the firm, the outcomes show notable changes. In Table 4, the interaction terms between the gender combination and the chairperson’s wealth are displayed. A major shift occurs when the chairperson is male but the general manager is female. The coefficient becomes positive when it is interacted with the chairperson’s stock-holdings, indicating that this combination could encourage research and development. Even in the previous case showing that females are unwilling to expand research and development and take risks, female managers are complementary when they work with a male chairperson, thereby leading to the promotion of research investments. These results indicate that when the general manager is female, the male chairperson and female general manager are both confident in their decision-making. However, the previous positive all male combination has a significant negative coefficient when it is interacted with the chairperson’s shareholding, indicating lower risk-taking incentive and greater hesitation when the chairperson has significant wealth in the firm. The results in Table 5 show that when the general manager holds larger shares of the firm, the combination of a male chairperson and a female general manager is still strategically complementary. More aggressive investment and risk-averse strategies only appear in the same-gender combinations. The combination of two males results in less enthusiasm for risk-taking, and the combination of a male chairperson with a female general manager results in a greater level of research investment.

Indeed, higher research investment interest may not indicate successful research outcomes to provide sufficient support in the manufacturing industry, as previously noted. Female participation in management provides prudential risk management and increases efficiency. The results are shown in Table 6 below. Column (1) shows that when both positions are held by males, the contribution to patents is significantly negative. Compared with other gender combinations, males display greater risk taking; however, such investments are also blind. In Column (2), when both positions are held by females, the contribution is significantly positive. Such high efficiency reflects female’s risk awareness and prudential decision-making. Column (3) uses the percentage of females on the management team and board and the coefficient remains positive and significant, confirming that females increase the research investment efficiency.

Tables 7,8 show that risk-averse females are suitable chairpersons and play a monitoring role in the firm. The data in Table 8 show that the likelihood of duality is much lower when the chairperson and general manager are both female than when they are both male. Furthermore, even in cases of female duality, firms still attract high levels of institutional ownership. Table 9 uses market (HS300) returns to estimate the expected returns and then calculates the alpha using the actual returns minus the expected returns. The results show that a female chairperson results in an increased alpha when the volatility of the firm’s returns is high, indicating that the monitoring role of females improves the performance of the firm.

The results show that the all male combination is strongly motivated by risk taking when exploring future changes and opportunities, but the all female combination is strongly risk averse and prefer to maintain their current state. Mixed-gender combinations demonstrate more interesting results. Female participation improves monitoring, whereas male participation leads to greater risk-taking incentives for research and development. The strategy should involve larger share vesting or its inclusion as part of the chairperson’s and general manager’s compensation since the wealth effect could alleviate the overinvestment when both positions are held by males and encourage risk-taking when the general manager is a female and the chairperson is a male. Considering the results in Tables 6,7,8 and based on the risk-averse characteristic of females, females’ decision-making leads to higher research and development success and efficiency, indicating that females have better research project-picking techniques, which explains why the wealth effect of the chairperson and general manager encourages firm risk taking, as shown in Tables 4,5, when the general manager is female and the chairperson is male.

We then check what happens during a systemic risk. In a similar manner as in the overall sample, the wealth of the chairperson and the general manager interact with the post-COVID-19 gender effect. Column (1) of Table 9 shows that the chairperson’s wealth significantly lowers the risk incentive when both the chairperson and the general manager are male. The original overall sample results in Table 3 show that the combination of a male chairperson and a male general manager has a significant positive coefficient, but a chairperson with greater wealth is more risk averse during the post-COVID-19 period. Column (1) of Table 10 shows that if the male general manager is aware of the impact of COVID-19 and is concerned about his own wealth in the firm, he will become risk averse and will also change the chairperson’s attitude toward risk. Such a change does not occur in the overall sample in Table 5.

According to the post-COVID-19 results, in addition to the scenario of a female chairperson or general manager, we may observe reduced risk when both the chairperson and the general manager are male. If shareholders believe that their current male general manager is an expert in the industry, one solution would be to encourage female participation as the general manager while increasing the share distribution as part of her compensation. Such a gender combination would increase research and development investment and its efficiency while controlling investments to a moderate level during crises. The other potential solution is to increase the firm’s shares as the proportion of the wealth of the chairperson and the general manager if both the chairperson and the general manager are males. When they have significant wealth as firm shares, the results show that they will reduce investment uncertainty.

Table 11 shows the propensity score matching (PSM) results. The chairperson and general manager are both male in the treatment group, and the size, total top ten ownership, and ROA are used as the matching factors; coarse exact matching (CEM) is used to assign different weights. The weighted data are used as the sample, and the results in Table 11 are similar to those of our baseline model. When females are the chairperson or general manager, research and development investment is reduced, which reflects females’ risk-averse nature. Note that the combination of different genders provides moderate research and development investment incentives compared with the same-gender combinations. When both positions are held by males, the investment incentive is the strongest, and when both positions are held by females, the investment incentive is the weakest. Furthermore, Table 12 shows similar PSM results when females are selected as the treatment group. The results confirm that female participation increases research outcomes and improves efficiency. When females hold at least one position (either the chairperson or general manager), research efficiency significantly outperforms the case of when both positions are held by males.

Table 13 employs different methods to demonstrate the robustness of the results. A small treatment is made for the sample. Any change of manager or chairperson becomes effective starting next year. The years overlapping due to changes in the manager and chairperson are removed. The first two columns add ROA and Tobin’s Q as additional control variables. The result of the interactive term remains similar to the original results. In columns (3) and (4), The male chairperson-female manager combination is interacted with, as in the baseline mode, where the chairperson and general manager shareholding is involved. Then, a similar PSM-CEM matching on firm size and ROA is performed, with the switch to involve females used as the control group. The results are controlled at the firm’s individual level. The interactive terms both have positive coefficients as in the baseline model.

Furthermore, in columns (5) and (6), the Heckman two-step selection model is used, and the selection model is based on female involvement. The ROA, the firm’s Sharpe ratios, and institutional ownership are used in the selection model to reflect that female involvement is affected by firm growth, performance, and ownership structure. The results of the outcome model indicate that the interactive terms remain positive and significant, consistent with the baseline model. Additionally, both columns show the insignificant t-value of the Inverse Mills ratio, indicating a lower risk of sample selection bias.

Table 14 shows the measurement method, where at least the chairperson or general manager is female. The results are similar to those of the baseline model. When the chairperson is female, the general manager is female, or there is at least one female among the chairperson or the general manager, research and development investments are significantly reduced. Table 15 shows the results when the instrumental variable method is used. The percentage of females on the management team and board is used as the instrumental variable for the female gender combinations. Columns (1) to (4) show that the percentage of females on the management team and board is a qualified instrumental variable. Since the percentage of females does not directly affect research and development investment in theory, we test whether females are more risk averse; thus, the percentage of females meets the condition of endogeneity. The results in Columns (5) to (7) are similar to those of the baseline model, indicating lower risk-taking incentives when females participate in the management team and board. However, the wealth of the firm increases risk-taking, indicating that the management team is confident in its research and development investment decisions.