Value Relevance Human Capital Information on the Annual Report of Indonesia Listed Companies

Human resources at the companies in Indonesia should be an important focus for investors, due to the general complaints of low labor productivity compared to other countries. This study aimed to investigate the value relevance of human capital disclosure and changes of human capital disclosure to share price and share return on listed companies at Indonesia Stock Exchange from all industrial sectors in 2016. Human capital information measured by word counting of information related to human capital in the firm annual report. The results showed that the disclosure information of human capital was value relevant with share price but only from qualification/ competencies category. For the disclosure changes in human capital, the information did not show any significant effect to return per share. Thus, generally Indonesia investors only focus on information on qualification/competency human resources in companies at that time, other information was ignored.JEL Classification: G11, G14, J24DOI: https://doi.org/10.26905/jkdp.v22i4.1945


| 615 |
In this 21 st era, the development of economics is more supported by knowledge. This knowledge is also known as a part of intellectual capital, which is recognized as a value creation of a company that is more important than the physical assets (Ellis & Seng, 2015). Research by Vergauwen & Van-Alem (2005) and Brggen, Vergauwen, & Dao (2009) the intellectual capital components are human capital, structural capital, and relational capital. One part that will be discussed is human capital in the organization, human capital considered a valuable resource and important things to the sustainability of the company's competitive advantage (Prahalad & Hamel, 1990). If we look from a strategic perspective, disclose a human capital as an intellectual asset of the company actually can be a good strategy for the effectiveness of the company and also can be the way to attract investors (Mariappanadar & Kairouz, 2017). Especially in the current economic conditions, companies should realize that some investors are not just looking for financial information but also nonfinancial information that can help them to evaluate every opportunity and risk of their investment (Alfraih, 2017).
Human resources at the companies in Indonesia should be an important focus for investors, due to the general complaints of low labor productivity compared to other countries, especially Japan, even though the increase in minimum wages is always guaranteed by the government for every year, as stated by the Indonesian Chamber of Commerce and Industry Chairman (Tribunnews, 2018). An issue of human resource in Indonesia is also reported by World Economic Forum in Global Competitiveness Report 2018. It's stated Indonesia's competitiveness is in ranked 45 th among 140 countries. The ranking of competitiveness was measured from 12 pillars, which was assessed and also ranked compared to other countries. The two pillars which rank is the lowest for Indonesia's competitiveness are the pillars of health (rank 95) and the labor market pillar (rank 82). Lev & Zarowin (1999), according to theory and empirical research, one way to improve the use of the financial statement for the investor is by paying attention to the recognition for the intangible investment (non-financial things). Also, face the changes in business and the economy, companies with non-traditional industry categories have a faster changes and have more complex intangible assets and intangible investments, so that this becomes something more considered by investors.
The value relevance and the disclosure of intangible assets: intellectual capital, has been the focus of several previous studies (Vafaei, Taylor, & Ahmed, 2011;Gamerschlag, 2013;Alfraih, 2017). But the research for the importance of human capital information still limited, one of them is the research by Gamerschlag (2013). Especially in Indonesia, which is a developing country with developing capital market conditions, it is interesting to learn whether investors in Indonesia have noticed the information about human capital owned by a company. Furthermore, human capital information is very broad, so that human capital information is important for investors. The condition of human capital in a company is dynamic for every year, then is information about the development of human capital also important for investors in Indonesia.
This research aims to obtain the evidence of the value relevance of the disclosure of human capital information and changes in human capital information on share prices and the returns for all companies listed on the Indonesia Stock Exchange (IDX).
| 616 | ence of each but is also the willingness of the individual to share this attribute with members in the organization to create value within the organization. Baron (2011) states, management of human capital speaks of the contribution of members to manageable company performance so that it becomes more positive and produces high value for the company. This is also supported by Stewart (1998) who states that knowledge will help in the process of increasing output and contribute to creating competitive advantage from a company.
Human capital according to Gamerschlag (2013), is divided into three categories; qualifications/competencies, motivation/commitment, and personnel. In each of these categories, there are keywords to describe: the categories of qualifications/ competencies include brain power, competence, competencies, education, expertise, intangible skills, intelligence, skills, know-how, learning, qualifications, specialists, and training. Motivation/commitment categories include absence, career, employee retention, employee turnover, employee satisfaction, entrepreneurial spirit, motivation, and staff turnover. The last category of personnel includes diversity, empowerment, human resources, personnel, recruiting, and recruitment.
One characteristic of financial information is relevant as in the Financial Reporting Conceptual Framework (IAI, 2017). Financial information is relevant if it is able to make a difference in decisions made by users. The research on value relevance was pioneered by Ball & Brown's (1968) research, which is how accounting information was responded to by the market through stock price movements and stock returns. So value relevance is how accounting information has value relevance to investors, and how investors react to the announcement of accounting information.
Ojo & Akkeren (2016) state that research on value relevance has an important role in helping to provide an understanding of the impact of infor-mation financial of a company. This is important because given the need for quality from the disclosure of accounting information needed for comparison and its consistency regarding accounting information.
Puspitaningtyas (2012) research shows that the investment value of a stock is influenced by investors' perceptions of the company's performance in the future. The value of a company's shares will increase if investors predict that the company's performance will increase. The opposite, the value of shares will decrease if investors predict that the company's performance will decline in the future. This research is also supported by research by Sumarni & Rahmawati (2007) that investor valuation of future earnings prospects can be obtained if investors have information relating to the company.
Usually, this value relevance approach refers more to accounting information in financial such as research by Sumarni & Rahmawati (2007), Puspitaningtyas (2012), and Ojo & Akkeren (2016). Lately, research has been developed on the value relevance of disclosure of nonfinancial information: human capital of a company as follows.
Research by Gamerschlag (2013), look for the relationship between human capital information to the stock market value relevance in Germany. Measurement human capital is carried out with use content-analysis that is by calculating the number of words from each human capital component from the annual report. While the market value is measured using share prices and share returns. The results obtained are the disclosure of human capital information is positively significant on the market price of the company, but change disclosure information human capital no to show existence a significant effect to share return.
Another study by Vafaei, Taylor, & Ahmed (2011), they had research on the value relevance of intellectual capital disclosure in Britain, Australia, Hong Kong, and Singapore. This study classifies intellectual capital into four broad categories: gen-

Yie Ke Feliana & Ester Novita
| 617 | eral terms, human capital, structural capital, and relational capital. The method for measuring disclosure is the same as Gamerschlag (2013), by calculating the frequency of disclosures made in annual reports. The results obtained by intellectual capital disclosure have a significant positive relationship to share prices for Britain and Hong Kong.
Research by Kehelwalatenna & Premaratne (2013) in New York that only use a banking company as a sample, who also want to find the value relevance of intellectual capital information (which includes human capital) to share prices. The measurement of human capital in this study is through the calculation of financial indicators, namely the value creation efficiency of human capital, which contains the burden of employees. He failed to find a significant relationship between human capital and share prices.
Based on the explanation of the relationship between variables and the results of previous studies, then the research hypothesis is as follows: H 1 : disclosure of human capital information for companies in Indonesia is value relevant to the share prices H 2 : changes in the disclosure of human capital information for companies in Indonesia is value relevant to the shares return Then the researchers also want to know the impact of human capital information when it viewed from each sub-category according to the research by Gamerschlag (2013), so that the hypothesis in above is translated into:

METHODS
The data collected for this study is annual reports from all industrial sectors. The industrial sectors include agriculture, mining, basic industries and chemicals, miscellaneous industries, consumer goods industries, property, infrastructure, finance, and trade and services. This study uses all types of industries because human capital is important for all industrial sectors and observations are made on investor reactions to human capital information in the entire company. However, to control the value relevant difference, the type of industry is still measured. The companies are listed on the IDX. Source of data is from the Fact Book and on www.idx.co.id (2016). The total population is 534 companies.
Criteria for being a sample in this research is the availability of the annual report of the company report, the financial statements ended on December 31 and measured in Indonesian Rupiah. After reducing the sample associated with sample criteria and sample reduction due to incomplete information for the required variables, the final sample used in the study was 402 companies because this sampling uses a nonprobability sampling method with type purposive sampling (taking a sample with use Jurnal Keuangan dan Perbankan | FINANCE Volume 22, Issue 4, October 2018: 614-624 | 618 | consideration certain corresponding with consideration researcher).
The method for measuring the disclosure of human capital information is the content analysis method, which uses the word count of the keyword human capital revealed in the company's annual report as was done in the Gamerschlag study (2013). This method is used for simple reasons, and many are done by previous research related to non-financial information (Brggen, Vergauwen, & Dao, 2009;Vafaei, Taylor, & Ahmed, 2011;Gamerschlag, 2013) as well as the most appropriate regarding the data varies (Gao, Li, & Clarke, 2008). In this study, human capital data is collected from annual reports that use varied terms for human capital information and varied location of information in annual reports. This method is developed based on the more words mentioned; the more information is expressed regarding these words.
The keywords for the disclosure of human capital information are divided into three categories according to Gamerschlag (2013), namely "qualifications/ competencies," "motivation/commitment," and "personnel." These keywords from the previous research are translated into Indonesian if the annual report does not use English, but if available in the English version the words are immediately used. Table 1 shows keywords from existing categories.
The market value is used as the dependent variable in this study was measured by using the share price (share price) and return on shares (share return) as in the research by Gamerschlag (2013 SPit is the share price by company i year t, BVEit/ Sit is the book value of equity per share outstanding by company i year t, EPSit is profit per share of company i year t, HCT is the total disclosure of human capital information (as other information from the model regression) company i year t, and IND is a type dummy industrial company i year t, and IND will be filled with numbers 1 agriculture, 2 mining, 3 basic industries and chemistry, 4 miscellaneous industry, 5 consumer goods, 6 properties, 7 infrastructure, 8 finance, 9 trades, and services.
Whereas for H 2 is measured by the equation: RETit is share return for company i year t, EPSit is profit per share of company i year t, PSEPSit

Yie Ke Feliana & Ester Novita
| 619 | is changing in earnings per share of company i year t, DHCTit is changing in disclosure of company human capital information for company i in year t, and IND is a type of dummy industrial company i year t.
The dependent variable used in this study is share price (SPt) and share return (RET). The share price is obtained from the company's shares closing price in December. RET is calculated by: Where SPt is the stock price of year t, SPt-1 is the stock price of the previous year, and DIV / S is the payment of dividends per share from the previous year.
The total variable disclosure of human capital information is the sum of the four categories of human capital information categories: Where HCT is the total disclosure of human capital, HCK is disclosure of human capital from the category of "qualification/ competence", HCM is the disclosure of human capital from the category of "motivation/ commitment", and HCP is disclosure of human capital from the category of "personnel." To test H 1a , H 1b , and H 1c use equation (5)

RESULTS
Having performed classical assumption test (normality, heteroscedasticity, autocorrelation, and multicollinearity), then the last sample will be used for the H 1 is 293 companies, H 1a,b,c is 287 companies, H 2 is 318 companies, and H 2a,b,c is 319 companies. The outlier elimination causes the difference in the number of samples during the normality test.
Descriptive statistics for the variables used in the four modes of regression are presented in Table  1. The average disclosure of information on total human capital (HCT) and human capital from the qualification/ competency category (HCK) has the highest yield compared to disclosure of human capital information from other categories or changes. That is, the average value of the disclosure of total human capital information is 78.7440, and the disclosure of human capital information from the qualification/ competency category is 48.9233. Table 2 illustrates the correlation between variables calculated using Pearson Correlations. Correlation between variables used in the first equation: regression used to test H 1 , variables that have a significant relationship to SP are BVE / S, EPS, and HCT variables, and each has a positive relationship. For the second equation, the regression used to test H 1a, b, c ; it turns out that HCK, HCM, and HCP also have a significant positive relationship to SP. This means that the more companies disclose things related to human capital, the higher the market price of the company's shares. In addition, HCK, HCM, and HCP are also positively related, possibly this means that if a company has decided to disclose information about human capital, it is likely that the disclosure is related to these three human capital categories.   While the correlation between variables used in equations 5 and 6, namely regression used to test H 2 and H 2a,b,c , variables that have a significant relationship (correlation) to RET are only IND variables (dummy industry), but for changes in the disclosure of information on human capital does not show a significant relationship with RET at all. Then between changes in the disclosure of human capital information that has a significant relationship between DHCK -DHCM and DHCK -DHCP, each has a positive relationship. Table 4 shows the results of the regression test for the four regression models.
From the results of testing multiple linear regression models for the four regression models, it is reported that simultaneously the regression models 1, 2, and 3 of all the independent variables are able to explain changes in the dependent variable. If seen from the ability to explain the independent variable to the dependent variable (adjusted R 2 value), then the regression models 1 and 2 are large enough, is when using stock prices as the dependent variable and human capital in a given year as the independent variable.
The results of the regression model 1 equation to test H 1 indicate a significant positive effect of BVE/ S and EPS on SP (stock price). But the variable disclosure of total human capital information (HCT) and type of industry does not indicate a significant effect on SP. While the equation of the regression model 2 to test H 1a, b, c, which shows a significant positive influence on SP is BVE / S, EPSt and disclosure of human capital information in the qualification/ competency category (HCK).
The results of the regression models 3 and 4 used to test H 2 and H 2a, b, c , indicate a negative significant effect on the IND (dummy industry) variable on RET (return per share), while for other variables namely EPS, "EPS, DHCT, DHCK, DHCM, and DHCP do not show a significant effect on RET.   Table 4. Analysis Regression Notes: SP (share price); RET (share return); BVE/S (book value equity per share); EPS (earning per share); "EPS (delta earning per share); HCT (total disclosure of human capital information); HCK (disclosure of human capital from the category of "qualification / competence"); HCM (disclosure of human capital from the category of "motivation / commitment"); HCP (disclosure of human capital from the category of "personnel"); DHCT (changes of HCT); DHCK (changes of HCK); DHCM (changes of HCM); DCHP (changes of HCP); IDN (dummy industry); *,** significant at the 5 percent and 1 percent level.

DISCUSSION
From the results of processing research data, shows that for the first hypothesis only H 1a is accepted because the disclosure of human capital information from the qualification/competency category for companies in Indonesia is value relevant to the share prices.
This means that the overall disclosure of human capital information is not yet of value relevant in the stock market in Indonesia. This shows that investors in Indonesia still see human capital as something uncertain in providing future benefits (Kehelwalatenna & Premaratne, 2013). Moreover, information related to this human capital is spread throughout the annual report, making it more difficult for investors to link that information with the market value of the company. This is in line with the results of research Ferraro & Veltri (2011) re-garding intellectual capital that there are still investors who lack the ability to see the opportunities of information intellectual capital so that the investor consideration given the disclosure of intellectual capital is not a major consideration in investing.
However, if the human capital information is specific to qualifications/ competencies, it seems that investors in Indonesia are still able to link that information in determining the market value of the company. Things about something more abstract, the motivation and performance of employees and also how the new recruitment system seems not to be considered by Indonesian investors. It is also supported that most companies in Indonesia disclose human capital information related to these qualifications/ competencies, as reported in descriptive statistics attachment 1. As seen in the descriptive statistics section, the average information about human capital qualifications/competencies reported

Yie Ke Feliana & Ester Novita
| 623 | by company 5.4 times more than motivation information/human capital competency and 2.7 times more than personal information.
The second hypothesis of this study was rejected (both H 2 and H 2a, b, c) due to changes in the disclosure of human capital information is not value relevant to share returns. In the regression model shows that each industry group has different characteristics for investors, so that it affects the company's return.
The results of this second hypothesis test show that investors in Indonesia have not noticed changes in human capital and even more detailed changes related to qualifications, motivation, and personnel systems in a company.
This also seems to be in accordance with the amount of information presented by the company, most of which is also information related to the qualifications/ competencies of human resources owned by the company. The results of the study contributed to the need for education for investors in Indonesia to better look at the long-term prospects of the company in this case related to the company's human capital.

Conclusion
The disclosure of human capital information as a whole does not have a value relevant to the share price for the company in Indonesia. But for the disclosure of human capital information from the qualification/ competency category, it turns out that investors are still considered relevant. In other words, the disclosure of human capital information is not responded by external users as a consideration in making decisions to invest, but the information that is considered more important by exter-nal users is related to the competence of the company's human resources. Changes in the disclosure of human capital information also apparently did not have relevant value with share return. This means the market has not considered that information about the development of human capital. In other words, investors in the Indonesian capital market have not considered the development of human capital as one of the information that describes the prospect of a business entity. This can be caused by the low likelihood that the company can control the benefits of human capital in the future with the possibility of a substantial turnover. This reason is in line with the argument why human capital although it provides benefits to the company but cannot be recognized as an asset by the company in the financial accounting concept.

Suggestions
The results of this study should be interpreted under consideration some limitations. The period used is only on year controlling macro conditions between companies, but this causes not able to see the value relevance of human capital information in the long run. Samples in this study are limited to the companies that only publish full annual report information related to human capital, so that the results may be interpreted limited to the condition of the sample. Also, the calculation for the disclosure of human capital information is still limited to the use of certain words. This is all because this research is limited to using secondary data from annual reports. Future studies are recommended using longitudinal studies to see the long-term effects of human capital information on the capital market. In addition, subsequent research is recommended to use primary data so that the data obtained is deeper related to human capital.