CAN CEO POWER MODERATE THE INFLUENCE OF INTERNET FINANCIAL REPORTING ON MARKET REACTIONS?

This research aims at determining the effect of Internet financial reporting and CEO power on market reactions and how CEO power moderates the effect of Internet financial reporting on market reactions. The research method used a quantitative approach and the sampling used a purposive sampling method to obtain 126 manufacturing sector companies listed in Indonesia. In addition, the analytical method used Moderated Regression Analysis. The test results showed that Internet financial reporting has no significant effect on market reactions, CEO power significantly affects market reactions, but CEO power could not strengthen the influence of Internet financial reporting on market reactions. The practical implications of the research results showed that Internet financial reporting helps investors obtain valuable information for decision-making


INTRODUCTION
Transaction activities in the capital market experience change daily in the form of increases or decreases (Nugroho et al., 2020).Several factors influence changes, such as an event or information.Event factors can affect both economic and non-economic aspects, and the information used by investors consists of fundamental information, including information contained in the company's financial statements and macroeconomic information that can be obtained by analyzing the economic conditions of a stock (Iwanicz-Drozdowska, M. et al., 2021.Quoted from bareksa.com, one of the listed companies, namely PT Jasa Marga Persero, Tbk (JSMR), experienced an increase in share price by 3.8% after publishing the company's financial report, which recorded a net profit of IDR 1.89 trillion.The company's net profit reflected an increase of 28.57% compared to the previous year.The increase in the company's share price reflects that the information contains important information and has received a positive response from investors (Dwianto & Yulita, 2019).https://doi.org/10.23969/jrak.v15i2.9718https://doi.org/10.23969/jrak.v15i2.9718

METHODS
The population in this research were manufacturing sector companies listed on the Indonesia Stock Exchange, totalling 184 companies.The sampling technique used purposive sampling with the following criteria; 126 companies met the criteria and became the research sample.1) Manufacturing Sector Companies listed on the IDX; 2) The company had a website; 3) The company published annual reports on the website; 4) The date of publication on the company's website can be known; 5) the company had complete data related to research variables.Based on the specified criteria, 126 companies became the research sample.
The research independent variable is Internet Financial Reporting.The Internet Financial Reporting measurement uses the IFR index developed by (Handayani & Almilia, 2013).There are four IFR assessment categories: content, timeliness, technology, and user support.Each indicator that fulfils or is disclosed would be given a score of 1 and added per category.The amount per category then be multiplied by the respective weight with the following description, content (40%), timeliness (20%), technology (20%), and user support (20%).
This research also used a moderating variable in the form of CEO power.The CEO Power measurement uses the CEO power index developed by Haider & Fang (2018).There are five assessment indicators: professional certification, tenure, share ownership, institutional ownership, and education.Each indicator that meets the assessment criteria is given a score of 1.The CEO power index is the sum of scores worth 0 (least potent) to 5 (powerful).
Market reaction as the dependent variable measurement using abnormal returns.Abnormal return is the difference between expected return and realized return.To determine the expected return using the marketadjusted model, which is a model that estimates returns using market prices or IHSG data from the Indonesia Stock Exchange.
We used the Descriptive Statistical Test to test the sample data, then carried out the classic assumption test consisting of the Normality Test, Multicollinearity Test, Heteroscedasticity Test, and Autocorrelation Test.This research used Moderate Regression Analysis (MRA).The choice of this method was to determine the moderating role of CEO Power on the influence of Internet financial reporting on market reactions.

RESULTS
The results of descriptive statistics showed that the average value of IFR was 37%, which was greater than the standard deviation value, indicating that the data had low deviations.CEO power has an average value of 1.33, and a smaller standard deviation value reflected low data variation.On the other hand, for market reaction variables shown to have abnormal returns, it indicated that the variable data has a high deviation value or the data varies.In the content category, 95% used the Indonesian language on websites; 94% of company websites also provided a foreign language to access company information.Other financial information is company stock summaries and charts.Of all the sample companies, only 51% presented stock summaries, and only 34% presented stock charts.Other IFR content included the presentation of company information.All companies in the research sample already provided company information content but still needed to improve the profile content and company CSR.In addition, financial highlights content, including ratios, growth, and graphs of 47%, 44%, and 17%, respectively, were found on the companies' websites that were the research samples.
Of the 85 companies that issued press releases, only 11 had updates in the past week from the observation time, while the other 74 only updated their press releases for a week.For the stock summary component, 33 of the 63 companies that provided stock summaries on the website were updated in the last week from the time of observation.Meanwhile, during observation, 30 other companies only updated stock summaries on their websites for a week.Apart from that, in the timeliness category, it also assessed company vision report items or future forecasts, which contained information on achievements during the company's operation for a certain period and a chart of future profit forecast, and none of the companies presented this information.From a technology standpoint, most companies have already provided plug-in downloads, and there was access to online feedback and support for user services.Whereas for multimedia components, only 49% of companies used multimedia such as images, videos, and presentation slides, and only 35% used information presentation technology to make information more engaging.For the analysis tool component, only 13% of companies provided tools for investors to facilitate information analysis, such as features in price analysis via stock price charts.As for XBRL technology, only some companies applied this technology in presenting information on their websites.
The company already had a link to the homepage menu for the user support category and a consistent web design.Whereas only 49% of companies provided a search menu (site search) 48% provided a link to the top menu, and 42% had a website map (sitemap) that could find the menu and the information location on the website.In addition, only 33% of companies required two clicks to access financial information.Finally, only 10% of company websites provided a help & FAQ menu to assist users and inform frequently asked questions.
The first indicator, namely professional certification, is 3%, indicating that out of 126 companies, only 3% of company CEOs had professional accreditation.The second indicator was tenure, 23% of company CEOs have a longer term than the average tenure.The third indicator is share ownership, showing that 18% of company CEOs had power over share ownership.CEO shares were larger than the average shareholding.The fourth indicator, namely institutional ownership with a percentage of 60%, indicated that institutional ownership in companies was more significant than individual ownership.Finally, the education level of the CEO as the fifth indicator was a minimum of a master's degree.The percentage of CEOs who gained power from education was 29%, indicating that there were 29% of all CEOs with a master's degree or more.
In the analysis unit, there were 13 companies with a CEO Power of 0.64, companies with a CEO Power of 1, 43 with a Power of 2, 6 with a Power of 3, and no company with a CEO Power of four and five.This result showed that most CEO Power in the company was still relatively weak or at least potent.A CEO who had strong power would better understand and consider the risks that may be faced from presenting financial information which can influence decisions taken by investors.
Abnormal returns indicated this market reaction.Abnormal return measurement by looking for the difference between expected and actual returns.The average abnormal return was negative and began to increase on H-1 and slightly on H-0.Then the average abnormal return was negative on H+1 and H+2 and increased on H+3 to a positive value.This empirical condition was likely to occur because investors are interested in published information and react from H+3.
This research used the One-sample Kolmogorov-Smirnov Normality Test.This study's sig (2-tailed) value is 0.200> 0.05, meaning the data is normally distributed, and the regression model met the normality assumption.The next test was the Multicollinearity Test, which tested whether multicollinearity existed.The data processing results showed that the tolerance value was> 0.10, and the VIF value was <10 for all variables, so the regression model equation did not contain multicollinearity problems.The next test was the Heteroscedasticity Test, which aims at determining whether there was heteroscedasticity based on its significance value.This research showed a significance value of > 0.05 for all variables, so the regression model equation did not contain heteroscedasticity.The final classic test was the Autocorrelation Test, which was to test whether there is autocorrelation.According to Santoso (2012:242), if the DW value was below -2, there was positive autocorrelation.If the DW value was above +2, then there was negative autocorrelation.If the DW value was between -2 to +2, then there was no autocorrelation.
The results of the classical assumption test explained that the data is typically distributed and free from multicollinearity symptoms, and there was no heteroscedasticity.Testing the hypothesis in this research used Moderated Regression Analysis (MRA).Following were the test results using SPSS 25 software: As presented in Table 2, IFR had a coefficient value of 0.007 and a significance level of 0.736 > 0.05, indicating that IFR had no significant effect on market reactions measured using abnormal returns.This research rejected the hypothesis that IFR has affected market reactions.Table 2 also showed that the value of the CEO Power coefficient was -0.009, and the significance level was 0.010 <0.05.These results indicated that CEO Power significantly has affected market reactions as reflected by abnormal returns.Meanwhile, the coefficient value of CEO Power as a moderating variable was 0.055, and the significance level was 0.073, more significant than 0.05.These results indicated that CEO Power could not strengthen the effect of IFR on market reactions as measured by abnormal returns.

DISCUSSION
IFR is a form of information disclosure by companies using Internet media, a form of voluntary disclosure (Khlifi, 2022).IFR practices in companies through the website will provide investors access to make investment decisions.The theory of market efficiency explains that efficient markets react quickly and accurately to information with essential information content and will soon form a new equilibrium price.Based on the results of hypothesis testing showed that IFR has a coefficient value of 0.007 and a significance level of 0.736, which is greater than the significance level of 0.05.Hence, the conclusion is that IFR does not significantly affect market reactions as measured using abnormal returns.
The results are inconsistent with the research of Lai et al. (2009), which stated that IFR can affect abnormal returns, where IFR is an effective medium for communicating company information to external parties.However, this research's results aligned with the research of Immanuel & Purbandari (2016), which stated that IFR has no significant effect on abnormal returns.Investors can find company information using other media, such as the IDX website.The results of this research are also consistent with Muid & Hargyantoro's research (2016) which stated that IFR cannot affect market reactions because almost all companies have carried out IFR practices as a form of voluntary disclosure, so they were not a significant factor influencing market reactions.Furthermore, this research is in line with Nurlita's research (2017), which stated that there is no significant difference between abnormal returns after and before IFR because investors cannot interpret company signals correctly because of the quality of company disclosures which causes investors to hesitate in making decisions.
This also showed that IFR had no effect because of the low IFR score, where no company achieved a maximum score of 100%.Up until now, there have been no satisfactory provisions regarding the category of IFR scores.However, this showed that only companies can fulfil some disclosures that become an IFR assessment.
The signal theory explains that the quality of corporate disclosure determines stakeholder perceptions.Companies that disclose information widely will receive more attention from investors and reduce information asymmetry.Most IFRs carried out by companies on websites only provide accounting information, a mandatory disclosure such as an annual report.The company did this because of the Financial Services Authority Regulation (POJK) No.29/POJK.04/2016Article 7 stipulates that issuers or public companies must publish the company's annual report four months after the year ends.In implementing IFR, company website media should communicate other company information such as press releases, news, future forecasts, and additional information that the company has done.Companies with a high level of business complexity will be a driving force for implementing integrated reporting models that are useful in making internal company decisions and as relevant information for other stakeholders, especially investors, for decision-making (Komar, S., Ahmar, N., & Darminto, D. P., 2020).
Another factor that can cause IFR not to affect market reactions is that investors' decisions in investing do not only pay attention to the information that the company presents through the website but can use historical information, such as looking at stock price trends by observing the behaviour of other investors in the capital market.In addition, there is a time difference in presenting information where other media often present information, such as the IDX website, before introducing it on the company's website, causing information leakage and causing information to be irrelevant for decision making.Investors do not only look at the forward-looking information disclosure but also analyze the company's performance (Widiastuti, H., Utami, E. R., & Purnamasari, E., 2022).
In addition, the results of this research indirectly explained that the Indonesian capital market is a semistrong form of efficient market.The inefficient market theory is semi-strong and efficient if security prices fully reflect all published information in the company's financial reports.The prices of securities already reflected the available information, including company information through Internet financial reporting.In this market, abnormal returns are complicated to occur and only occur when an event and information circulating in the capital market has important information content.Then the signal theory also explains that in an efficient market, signals arising from the information, both from external and internal companies, will directly affect the price movements of related companies.