Agency theory emerged after the separation of ownership of company management and managers in large companies (Monks & Minow, 2004). Agency theory is the relationship between holders shares or the government as the principal with the company management as the agent. Jensen & Meckling (1976) define agency theory as an agreement in which the agent performs certain services delegated to him. The application of agency theory can be reflected in an agreement that regulates the rights and obligations of principals and agents while taking into account the interests of both parties. The agreement will be effective if there is a balance between the principal and the agent that shows the fulfillment of obligations by the agent and the provision of incentives by the principal.

FLID can reduce information asymmetry, which leads to a decrease in agency costs (Hassanein & Hussainey, 2015; (Kılıç & Kuzey, 2018)) and hence lower capital costs (Enache & Hussainey, 2020). To mitigate asymmetry, firms provide higher levels of forward-looking information disclosure so as to maintain a superior assessment of future firm performance (Kılıç & Kuzey, 2018). Reporting corporate financial information can help monitor managers and reduce agent costs. (Healy & Palepu, 2001). (Wang & Hussainey, 2013) state that managers may continue to invest in segments with low growth rates to maintain business complexity and increase compensation. Segment information may indicate resource transfers between segments, segment underperformance, or agency problems associated with ineffective diversification strategies.

Forward-looking information disclosure is the disclosure of future performance-oriented information disclosed by the company in the annual report (Agyei-mensah, 2017). FLID helps investors and users to make investment decisions because with this disclosure the information asymmetry between the company and related parties is reduced (Uyar & Kilic, 2012). One type of voluntary disclosure is the disclosure of forward-looking information, which is useful for external parties such as investors and shareholders in describing the company's future performance (Mardani & Kristanto, 2020). The disclosed aspects include financial aspects regarding projected revenue, profit, loss, profit per share, investment, capital expenditure, and other information. In addition, it also discloses non-financial aspects about plans related to product development that the company will carry out in the future (Oktavia & Imelda, 2022).

Companies that disclose forward-looking information have several advantages, such as reducing information asymmetry and providing information relevant to the company's future operations, plans, strategies, policies, and financial targets ((Celik et al., 2006); (Uyar & Kilic, 2012)). This helps investors to estimate expected cash flows and future firm value ((Bravo, 2016); (Menicucci & Paolucci, 2016)). Forward-looking information can also increase firm value by attracting investors who are motivated by the firm's commitment to creating value for its owners (Haque et al., 2023).

In contrast, there are still companies that hesitate to disclose risks and future prospects of the company for various reasons. First, companies are worried that the disclosed information will be utilized by competitors which will indirectly increase company costs. Second, companies are worried that the plans that have been disclosed in the financial statements cannot be achieved in the future.

Board diversity is an important topic in the business world, with increasing opportunities for women in senior management positions (Choi et al., 2022). Women's participation provides a broader perspective, crucial in decision-making and responding to stakeholder demands. Women's involvement is needed to gain a different perspective in dealing with business uncertainty (Dowling & Aribi, 2013). Previous studies note male and female differences in personality, communication style, decision-making, risk-taking, leadership, ethics, and business performance (Peterson & Philpot, 2007). (Alqatamin et al., 2017) and (Mardani & Kristanto, 2020) found that female directors tend to adopt conservative accounting policies in earnings management, indicating a tendency to be cautious in decision making. Gender differences in company directors affect the level of information disclosure in annual reports ((Aribi et al., 2018); (Mardani & Kristanto, 2020)).

H1: Gender diversity has a positive effect on forward-looking information disclosure

The board of directors functions to set company strategy and policy and plays a supervisory role that can influence managerial decisions to increase forward-looking information disclosure ((Dey et al., 2020); (Muliawati & Rohman, 2018)). A small number of boards of directors is considered more effective because it is easier to coordinate and communicate (Wang & Hussainey, 2013). Meanwhile, a large board of directors with a diversity of experience and financial expertise can expand the disclosure of forward-looking information (Elzahar & Hussainey, 2012). Nonetheless, research results regarding the relationship between board size and FLID are inconsistent. Some studies find no significant impact of board size on FLID ((Agyei-Mensah, 2018); (Buertey & Pae, 2021); (Kılıç & Kuzey, 2018); (Oktavia & Imelda, 2022); (El-Deeb et al., 2019)).

H2: Board size has a positive effect on forward-looking information disclosure.

The frequency of board meetings has a complex influence on risk disclosure. Some studies show a positive relationship, where more frequent meetings lead to better disclosure and reduced information asymmetry ((Luthfiansyah & Wijayanti, 2022); (Muliawati & Rohman, 2018)). The Board of Directors needs to hold meetings to discuss strategy and implementation of management supervision (Domínguez et al., 2014). Alnabsha et al. (2018), the frequency of board of directors meetings has a significant positive effect on the disclosure of forward-looking information, board meetings also increase performance effectiveness and efficiency (Mauda & Serly, 2021) and reduce asymmetric information problems between managers and directors.

H3: Frequency of board meetings has a positive effect on forward-looking information disclosure.

Descriptive statistical results in Table 2. , it can be seen that the dependent variable, namely forward-looking information disclosure, has an average of 0.2418. This average shows that the company conducts 24.18% FLID. Where this value tends to be low because it is below 50%. Companies that have the lowest FLID values are Alkindo Naratama, Duta Pertiwi Nusantara, Fajar Surya Wisesa, Malindo Feedmill, Prima Alloy Steel Universal, and Trita Mahakam Resources with a value of 0.0667. On the other hand, Arwana Citra Mulia, Astra Otoparts, Multi Bintang Indonesia, and Mustika Ratu are the companies that disclose the most FLID aspects in their financial statements. Of the 30 measurement aspects in FLID, the four companies disclose 16 financial and non-financial measurements.

Board gender diversity has the lowest value of 0.0000. This means that there are companies that do not have a female board of directors or all boards of directors are male. The company that has the highest proportion of board gender diversity is Asia Pacific Investama with a value of 0.6667. The company has two female directors out of a total of three. The lowest number of boards of directors is 0.6931. There are 17 companies that have the least number of boards of directors. Of the 17 companies, there are 11 companies that have a consistent number of boards of directors from 2020-2022 with a total of 2 boards of directors. Meanwhile, the highest value on the board of directors is 2.3979. Asahimas Flat Glass, Indofood CBP Sukses Makmur, Indofood Sukses Makmur, Surya Toto Indonesia are the companies with the most directors with a total of 11 people.

Frequency of board meetings has the lowest value of 3,000. This means that in one year the company held a meeting 3 times, these companies are Mulia Boga Raya and Ekadharma International. For the number of board of directors meetings, it has the highest value at 60.0000. This means that in one year the company held 60 meetings, the company is Kimia Farma.

Profitability has an average of 0.0325 with a standard deviation of 0.1223. This means that. The lowest value of -1.0800 was held by the Trita Mahakam Resources company in 2020 with earnings after tax (EAT) of - Rp426,269,024,900. The highest value of 0.5990 is held by the Tiga Pilar Sejahtera Food company with an EAT of IDR 1,204,972,000,000 in 2020.

Leverage has an average value of 0.5080 with a standard deviation of 0.4855. The lowest leverage value was 0.0005 by Mulia Industrindo in 2020 with debt compared to assets of only 0.05%. Meanwhile, the highest value was 4.4467 for the company Bentoel International Investama in 2021 with a proportion of debt to assets of 444.67%.

The age of the company has an average of 2.5409 with a standard deviation of 1.0309. The age of the company with the lowest value is -0.1250, this value is negative because the actual age of the company at 31/12/2020 is only one year. Meanwhile, the highest value is 3.8153. The oldest company is Solusi Bangun Indonesia Tbk. The company has been established since 10/08/1977, so that in 2024 the company will be 45 years old.

Panel data regression models using Eviews have three methods to choose from: common effect model, fixed effect model, and random effect model. To choose the method used in the study, it is necessary to test first with the chow, hausman, and lagrange multiplier tests.

The chow test results are in Table 3. . If the probability is more than 0.05 then use CEM, otherwise use FEM. The chow test results in the study obtained a probability value of 0.0005. It is concluded that FEM is more suitable than CEM.

The results of the Hausman test are in Table 4. . If the probability is more than 0.05 then use REM, otherwise use FEM. The hausman test results in the study obtained a probability value of 0.2887. It is concluded that REM is more suitable than FEM.

The results of the Lagrange multiplier test are in Table 5. . The results of the lagrange multiplier test are in Table 5. . If the probability is more than 0.05 then use CEM, otherwise use REM. The results of the Lagrange multiplier test in the study obtained a probability value of 0.2669. It is concluded that CEM is more suitable than REM. So in this study, the model used is CEM.

Hypothesis testing in research is carried out after the classic assumption test is fulfilled. The multicollinearity assumption test is carried out to show that there is no correlation between independent variables that exceeds 0.8 (Ghozali 2017). Table 6.

it can be seen that the highest value is only 0.1681. it is concluded that multicollinearity does not occur in the data.

The regression coefficient on the board gender diversity variable is 0.0806 with a positive direction. It can be interpreted that the greater the proportion of women on the board of directors, the more forward-looking information disclosure will increase. The regression coefficient on the board size variable is 0.0248 with a positive direction. It can be interpreted that the more the number of board of directors, the more forward-looking information disclosure will increase. The regression coefficient on the variable number of board meetings is 0.0006 with a positive direction. It can be interpreted that the more often the board of directors meets, the more forward-looking information disclosure will increase.

In the control variable, the regression coefficient on the profitability variable is 0.0463 with a positive direction. It can be interpreted that the greater the proportion of profit to assets in the company, the more forward-looking information disclosure will increase. The regression coefficient on the leverage variable is -0.0280 with a negative direction. Coefficient regression on the company age variable is 0.0030 with a positive direction. This can be interpreted that the longer the company stands, the more forward-looking information disclosure will increase.

Table 8. shows that the F-statistic is 3.8501 and the Prob (F-statistic) is 0.0010. This shows that the regression model is feasible to use or fit.

Table 9. shows the Adjusted R-square result of 0.0530 or 53.00%. This result means that all independent variables, namely board diversity, boars size, and the number of board meetings can explain the dependent variable, namely forward-looking information disclosure by 53.00%. While the remaining 47.00% is explained by other variables outside the study.

The first independent variable is board gender diversity, the results of Table 10. show that the probability value is less than 0.05, namely 0.0123. This means that the board diversity variable has an effect on FLID. The existence of a female board of commissioners provides a broader picture of decision making in the company. The presence of more women will provide a different point of view in dealing with the company's business risks. This is line with (Alqatamin et al., 2017), (Kılıç & Kuzey, 2018), (Aribi et al., 2018), (Mardani & Kristanto, 2020), (Effah et al., 2022), and (Choi et al., 2022).

The second independent variable is board size, the results of Table 10. show that the probability value is less than 0.05, namely 0.0411. It can be interpreted that the board size variable affects FLID. A large number of boards of directors with a diversity of experience and financial expertise can increase the disclosure of forward-looking information. This is line with (Akhtaruddin et al., 2009), (Elzahar & Hussainey, 2012), (Al-Najjar & Abed, 2014), (Dey et al., 2020), and (Muliawati & Rohman, 2018).

The third independent variable is the number of board of directors meetings, the results of Table 10. show that the probability value is more than 0.05, namely 0.4263. This can be interpreted that the variable number of board of directors meetings has no effect on FLID. This is because the disclosure of such information is more influenced by internal company factors such as management quality, corporate culture, and supervisory systems, as well as external factors such as regulations and markets.

The control variable in this study, namely profitability, shows a significant value of more than 0.05, namely 0.3313. This means that the control variable has no effect on FLID. It can be interpreted that the size of the profit earned by the company does not affect the application of forward-looking disclosure because companies with large profits as a signal to stakeholders, then companies with small profits also make disclosures to obtain funding sources. This is in line with the research of (Uyar & Kilic, 2012), (Kılıç & Kuzey, 2018), (Christian & Salim, 2022), (Oktavia & Imelda, 2022), and (Effah et al., 2022),

The control variable in this study, namely leverage, shows a significant value of less than 0.05, namely 0.0193 with a negative direction. This means that the control variable affects FLID. It can be interpreted that companies that have high liabilities tend to have high risks, so companies will tend to be careful in disclosing forward-looking information, to keep the company from public service supervision. This is in line with the research of (Kılıç & Kuzey, 2018), (Dey et al., 2020), and (Oktavia & Imelda, 2022).

The control variable in this study, namely company size, shows a significant value of more than 0.05, namely 0.5700. This means that the control variable has no effect on FLID. This is in line with the research of (Uyar & Kilic, 2012).