“Ownership composition and intellectual capital disclosure: Indonesia as a case study”

This study explores whether ownership structure (comprising ownership concentration, foreign, managerial, and institutional ownership) affects intellectual capital disclosure (ICD) in Southeast Asia’s largest stock market and Indonesia’s emerging economy. The sample includes 323 public firms listed on the Indonesia Stock Exchange (IDX) from seven industries between 2008 and 2017, or 2,634 firm-year observations. Data were analyzed using the ordinary least squares (OLS) regression with robust standard errors. The results show that ICD is positively related to ownership concentration. A negative and substantial relationship was found for both foreign and managerial own-erships, while the institutional ownership variable had a negative and insignificant impact. Overall, the results show robust conclusions regarding the impact of the ownership structure on ICD. The findings of this investigation could be taken into account by capital market authorities such as the Indonesia Stock Exchange (IDX) to raise awareness of intellectual capital and improve ICD practices.


INTRODUCTION
The shift from physical capital to a knowledge economy has brought about significant changes in the nature, structure, and operations of companies. Most companies have started to focus on intangible assets or intellectual capital (IC) rather than tangible assets. IC is gradually replacing fixed assets as the most important matter for a company. IC is also considered important because competition does not only focus on tangible assets, but also on the company's innovation, its information systems, organizational management, and human resources. Therefore, the ability and knowledge become one of the focuses of a company at this time though the focus of increasing the company's intellectual capital must also be related to increasing disclosure of intellectual capital (ICD).
Disclosure of information by a company provides a signal that describes the quality of the company towards stakeholders. The information disclosed is in the form of mandatory disclosure and voluntary disclosure. Disclosures address costs and benefits, which are relatively difficult to measure, especially the measurement of benefits. How extensive the information is disclosed needs attention so that the information presented is not too much which can cause noise and not too little that can mislead users. Hence, it is important to carefully manage the information sufficient to influence stakeholders' judgments and decisions.
The objective of managing disclosure of information is not limited to what can be stated in a financial statement. Financial reporting also includes the provision of information that must be revealed in accordance with policies or laws by authorities, as well as information, which management considers beneficial for external parties to be disclosed voluntarily. Hence, the company does not only focus on increasing intellectual capital, but also provides the required intellectual capital information. This is an important factor in the company as a strategy in achieving corporate goals as a supplementary communication.
In Indonesia, officials have regulated the disclosure of information such as Act 14 of 2008 on Public Information (KIP), Financial Services Authority Regulation No. 60/POJK.04/2015 on the Transparency of Information of Particular Shareholders, Financial Authority Services Regulation Number 29/ POJK.04/2016 concerning the Annual Report of an Issuer or a Public Company, and most recently, the Financial Authority Services Regulation Number 43./POJK.04/2020 covering obligations of the information disclosure and corporate governance for public corporations or listed issuers falling into the issuer-class owning small or medium scale resources. However, an increase in intellectual capital of many companies does not match the level of intellectual capital disclosure (ICD). This can lead to increased information asymmetry, making it difficult for stakeholders to make decisions.

LITERATURE REVIEW
The separation of administration from ownership in a company creates a conflict of interest between shareholders and directors. Moreover, this is supported by the agency theory reiterating a clash caused by the division of control from ownership in modern corporations (Jensen & Meckling, 1976). Oliveira et al. (2006) posit there is a greater motivation in corporations with stronger ownership decentralization to reveal information freely and lessen expenses. Therefore, spread ownership influences the way news is disclosed (Eng & Mak, 2003). In fact, disclosure is likely larger in companies owned broadly, therefore owners of capital can effectively monitor the management, and their economic interests can be optimized (Hidalgo et al., 2011). Craswell and Taylor (1992) showed that the higher agency cost for non-disclosure and the cost of ownership for disclosure are the two factors that determine the manager's disclosure decision. Mckinnon and Dalimunthe (1993) state that ownership diffusion is a factor of a manager's disclosure decision in Australia.
Earlier investigations, however, discovered conflicting findings to show diffused ownership concentration causing a little extent of disclosure. This is because the average shareholder has a low percentage of ownership. Due to this low percentage of each shareholder, they cannot make decisions in a company (Barako et al., 2006). Ferreira et al. (2012) also stated that different interests in contracting parties caused high agency conflicts in companies with low ownership concentrations. This was because such companies had more indirectly involved shareholders, and dominant actors had access to management information (Prencipe, 2004). García-Meca and Sánchez-Ballesta (2010) revealed a relationship between deliberate disclosure and ownership concentration using meta-analysis. The results show that lower disclosures are supported by firms with a high degree of concentrated ownership. Agency theory explains that there is information asymmetry between the principal and the agent due to differences in interests. Therefore, high managerial ownership forces management to disclose little information as the company does not have an intense relationship with external parties. Jensen and Meckling (1976) stated that high share ownership by management over the company's capital could reduce agency problems. Meanwhile, managers who are company owners will be interested in disclosing information to increase the liquidity of shares and to adhere to the constraints imposed by insider trading regulations. Therefore, where there is strong administrative ownership in the capital organization, disclosure is encouraged and agency expenses are capable of being lessened. Moreover, with a quantum for ownership shares, agency expenses are also lessened, since shareholders' and directors' interests become unified (Jensen & Meckling, 1976). However, Fama and Jensen (1983) claimed there was a negative influence of large managerial ownership on capitalization of offered identity value by the managers and members in self-profit. According to Brown et al. (2004), access to finance, market knowledge, improved technology and management skills amongst foreign owners significantly affect productivity. Furthermore, foreign investors pay attention to management evaluation appraisals and keep high standard of information disclosure (Boubakri et al. 2005). According to Naser et al. (2002), due to more regional and international market experience, foreign investors demand high disclosure standards. Haniffa Cheng and Courtenay (2006) found no relationship in SGX listed companies. A study on intellectual capital disclosure was conducted by Muttakin et al. (2015). The results indicated that higher foreign ownership correlated with a larger quantum of ICD.
According to Jensen and Meckling (1976), institutional shareholders play a crucial role in reducing agency conflicts that can arise between shareholders and managers. The presence of these shareholders in a company is considered to be able to effectively control every strategic decision and action taken by company managers. Shleifer and Vishny (1986) confirmed the position and approved the experience of institutional investors and supervisory capability concerning corporate management costs that contribute to governance and ICD. Lakhal (2005) found that institutional ownership positively and significantly impacted voluntary disclosure of French firms. Barako

HYPOTHESES DEVELOPMENT
The research hypotheses were based on a combination of the theoretical background of voluntary disclosure and experimental investigations. There are various determining variables for intellectual capital disclosure, with a major element being ownership structure. The ICD theoretical outline indicates ownership composition affects whether intellectual capital is disclosed. In light of this discussion, the relationship between ownership structures and ICD was explored. Taken together, the following hypotheses are proposed: H1: Ownership concentration positively influences intellectual capital disclosure.
H3: Foreign ownership positively influences intellectual capital disclosure.
H4: Institutional ownership positively influences intellectual capital disclosure.

METHODOLOGY
The There are three types of variables -independent, dependent, and control. The dependent is ICD, and is classified into three groups (see Table 2), which include Internal Capital Category (ICC), External Capital Category (ECC) and Human Capital Category (HCC). The method for measuring the ICD was by using the disclosure index developed through a modified methodology by Muttakin et al. (2015) and Vergauven and Alem (2005).  2015), ICD measurements use content analysis. The analysis was conducted using an unweighted dichotomous procedure. Following the content analysis process, the score is 1 when the annual report contains the item disclosure. Conversely, the score is 0 when the yearly report displays absence of any item disclosure. The disclosure score indicator is structured as follows: , and Nadeem (2020), several control variables were used such as SIZE, LEV, ROA, AGE and MEET. The bigger the company (SIZE), the more the tendency to disclose information. In addition, those with a higher leverage ratio (LEV) will disclose more information, especially about intellectual capital due to a high level of financial risk. Moreover, there is higher possibility of corporations revealing more information when their financial statements show good performance (ROA). Those with an older age (AGE) disclose more information, and those that have a high frequency of meeting activity (MEET) like to share information with the public.
Regression analysis is employed in this study to assess whether ownership structure variables affect ICD levels. The equation for regression is as follows: , , where ICD = Intellectual Capital Disclosure, CON = Ownership of shares possessed by one or more individuals at 5%, MEN = Share percentage managers owned, FORG = Share percentage foreign financiers owned, INST = Share percentage institutional financiers owned, SIZE = The overall assets natural logarithm, LEV = Proportion of overall debt to overall equity, ROA = The proportion of net gains to overall asset, AGE = The company's age since the incorporation date, and MEET = Overall amount of yearly board meetings.
Furthermore, the study employed ordinary least squares (OLS). However, there were several assumptions in the regression analysis that needed to be the Best Linear Unbiased Estimator (BLUE) in estimating with OLS. Therefore, to deal with Heteroscedasticity and autocorrelation issues, HAC (heteroscedasticity and autocorrelation) robust standard errors were used in the panel data (Wooldridge, 2009).

EMPIRICAL RESULTS AND DISCUSSION
The descriptive data on the variables employed in analyzing the whole sample is displayed in Table 3.
Average ICD in the sample is 0.5196 and a standard deviation is 0.1641. These results show that the average ICD for the sample is more than half of the total actual disclosure of the total items (32 items).
Overall, the average of the variables is greater than the standard deviation. Therefore, it can be a good representation except for the managerial, foreign, and institutional ownership, as well as leverage variables.
In the multivariate regression analysis, the degree of correlation between the explanatory variables is shown in Table 4. The correlation matrix was not found to be highly correlated with the explanatory variables, justifying that multicollinearity is not an issue. According to Kennedy (2008), multicollinearity is not a problem in a data when the correlation is above 0.70. In this case, there is no issue.
The relationship between ICD and explanatory variables was estimated using OLS with heteroscedasticity and autocorrelation (HAC) robust standard errors. To specify the range of correlation, control variables were included in a hypothesized study determining the impact of ownership structures. Seemingly, distributed ownership concentration positively influenced disclosure through management behavior monitoring (see Table 5). Generally, a company improved ICD by achieving high ownership concentration thereby supporting H1. This is in line with Haniffa  High managerial ownership makes management tend to disclose low intellectual capital, because a company does not have an intensive relationship with external parties, and the majority shareholder of the company receives more information than is contained in the annual report. Therefore, this research outcome confirms H2 and indicates a negative statistical effect of managerial ownership on ICD. This is in support of earlier research by Eng and Mak (2003), Barros et al. (2013), and      Notes: * Levels of significance at 10%, ** levels of significance at 5%, and *** levels of significance at 1%. ICD = Intellectual capital disclosure, This study analyzed whether foreign ownership (FORG) affected the ICD level and found a negative substantial effect, meaning that a higher ratio resulted in lower ICD levels. Hence, H3 is not supported. These findings contradict Haniffa and Coke (2002), Barako  This section tested the strength of the central findings using two methods. First, following Nadeem Notes: * Levels of significance at 10%, ** levels of significance at 5%, and *** levels of significance at 1%. ICD = Intellectual Capital Disclosure, CON = Share ownership held by one person or more 5%, MEN = Percentage of shares owned by the managers, FORG = Percentage of shares owned by the foreign investors, INST = Percentage of shares owned by institutional investors, SIZE = The natural logarithm of total assets, LEV = The ratio of total debt to total equity, ROA = The ratio of net profit to total assets, AGE = The age of a firm from the date of establishment, and MEET = Total number of board meetings held per year.
(2020), the sample was divided into two groups, high ICD and low ICD firms, to check robustness of the main results. The findings from this analysis also suggested that the relationship between the ownership structure and the ICD is consistent with the main results, especially in high ICD firms (see Table  6). Secondly, according to Muttakin et al. (2015), this study also employed the extent of ICD for the following different categories of intellectual capital: Internal Capital Categories (ICC), External Capital Categories (ECC) and Human Capital Categories (HCC). Table 7 shows the estimated results by employing different categories of intellectual capital. As expected, the results of these robustness tests further validated the main findings to confirm that the ownership structure significantly affects ICD.

CONCLUSION AND RECOMMENDATIONS
Research related to voluntary disclosures is comparatively novel, with various explanations for why companies disclose information voluntarily, including intellectual capital disclosure (ICD). This study determines the influence of concentration, managerial, foreign, and institutional ownership on ICD. To determine the relationship, 323 public firms listed on Indonesia Stock Exchange (IDX) were analyzed. The results showed that ownership concentration positively influenced ICD. Furthermore, managerial and foreign ownership negatively impacted ICD. Finally, institutional ownership (INST) did not influence the extent of sample companies ICD. Additionally, the results passed a series of robustness checks, including alternative measures of ICD with different categories and alternative sub-samples.