Voluntary Disclosure of Intellectual Capital: The Case Of Family And Non- Family Businesses In Jordan

The purpose of this paper is to examine the effect of firm’s characteristics (family vs. nonfamily firms) on disclosure behaviour regarding intellectual capital. The sample is composed of 201 firms from Amman Stock Exchange in 2018. First, this study differentiates three categories of intellectual capital: structural capital, relational capital, and human capital. Second, it examines the influence of the size, the profitability, the leverage, and the industry on the importance of intellectual capital disclosure. The study shows that family firms disclose more intellectual capital information than nonfamily firms. The results also show that the industry and the size have a significant and positive influence, that profitability and the leverage have a significant and negative influence on capital intellectual disclosure. Furthermore, Jordan family firms seem to disclose more about intellectual capital than Jordan non-family firms do.


1.
Introduction In the new knowledge economy, wealth is created through the development and management of intellectual capital (IC) (Ricceri and Guthrie, 2009;Hayes and Schaefer,1999;Stewart, 1997). Value creation by the firm is no longer only linked to financial and material resources, but a large part is generated by intangible resources such as research and development, innovation, corporate reputation, brands, employee skills and knowledge, processes, corporate culture, etc. (Ricceri and Guthrie, 2009; Hayes and Schaefer, 1999; Stewart, 1997). These intangible resources, also known as CIs, are essential to the creation of a company, and are essential to the creation of a new business. of value by the company (Aboody and Lev, 2000; Chaminade and Roberts, 2003;Habersam and Piber, 2003), and on the other hand in the construction of sustainable competitive advantages generating thus increasing market value for the shareholder (Bukh and Johanson, 2003; Holland, 2001). Edvinsson (1997) introduced the definition of IC in his study of the Skandia company. IC is then defined as the possession of knowledge, applied experience, a technology, customer relationships and key competencies that enable the company to be a leader in the field of competitive on the market. Mouristen et al. (2001) consider this term to include the assets intangible and allows the creation of value for the company. It is partly reflected in the difference between the market value of the company and its book value. However, Charimande and Roberts (2003) note that it is accepted that the term intangible (or incorporeal) is a technical word used in financial accounting whereas the term CI is used in the management research field. Several definitions have thus been proposed to define the notion of IC without reaching a consensus (Beattie and Thomson 2007). Lev and PSYCHOLOGY AND EDUCATION (2021) 58 (1): 2819-2837 ISSN: 00333077 2820 www.psychologyandeducation.net Zambon (2003) add that, on the other hand, a broad consensus exists on the categories of intellectual capital, namely human capital, structural capital and relational capital. Similarly, the OECD (2008) considers that the scope of intellectual capital has evolved towards a broader concept that includes human resources and capabilities, structural capital (databases, technology, habits and culture) and relational capital (organizational concepts and processes, networks of customers and suppliers, etc.). Our paper is based on the definition proposed by Edvinsson and Malone (1997), the most frequently used definition in the research work on intellectual capital reporting (Sharabati et al., 2010; Yang and Lin, 2009), a definition that also has the advantage of being convergent with the work cited above. The literature on family firms, which initially focused on succession problems, has evolved towards other issues such as performance and growth (Carney, 2005; Arrègle and Mari, 2010). Nonetheless, several researchers consider that the specificities of family-owned firms in relation to non-family firms have not been sufficiently studied (Castillo and Wakefield, 2007;Harris et al., 2004;Westhead and Howorth, 2006;. Salvato and Moores (2010), based on a metaanalysis of research (published between 1982 and 2010) on accounting practices in family firms, identified five articles on voluntary disclosure of information. In this study, no articles dealing with the voluntary disclosure of intellectual capital information by family businesses were identified. Our article is intended to fill this gap in previous work related to intellectual capital reporting by family businesses. It seeks to demonstrate that there are differences in the disclosure of IC information between family-owned and nonfamily businesses. To this end, we have chosen to analyze the case of companies listed on the Amman Stock Exchange in Jordan for the year 2018.
The article is divided into four sections. The first is devoted to the review of the literature. It includes on the one hand a summary of the work dealing with intellectual capital reporting and on the other hand the particularities of family businesses in terms of voluntary disclosure of information. The second presents the working methodology based firstly on a content analysis of annual reports and secondly on a regression model. The third is devoted to the presentation of the results. The last one presents the discussion of the results obtained.

Disclosure of information on intellectual capital
In a knowledge-based economy, companies identify their core competencies as invisible rather than visible assets. In this context, intellectual capital has become a crucial long-term factor in company performance (Itami, 1989). The concept of intellectual capital, introduced by the economist John Kenneth Galbraith in 1969, refers to the difference between the market value and the book value of a company. Many researchers see it as a way for a company to create a competitive advantage and improve its performance. One of the first definitions of intellectual capital dates from 1997, according to which intellectual capital is the possession of knowledge, applied experience, technology, customer relations and distinctive skills that provide the company with a competitive advantage in the market (Edvinsson and Malone, 1997 (2000) state that the purpose of intellectual capital disclosure is to provide appropriate, reliable and timely information to those who need it to make decisions about their relationship with the company. Research on voluntary disclosure of intellectual capital-related information can be divided into two categories. The first category is descriptive in nature, while the second is explanatory in nature. This classification goes beyond the methodological aspect to include the nature of the results found. Indeed, studies of a descriptive nature aim either to examine the categories and sub-categories of intellectual capital most disclosed by companies, or to analyse the evolution over time of this reporting (Campbell and Abdul Rahman, 2010). Explanatory work, on the other hand, seeks to identify the factors influencing voluntary disclosure of intellectual capital information. Some of this work is briefly presented below.

Descriptive studies on intellectual capital
Descriptive studies have used manual content analysis or assisted by specialised analysis programmes (Bontis, 2003). Annual reports were the most widely used media (Beattie et al., 2002;Campbell, 2000). Other sources of information were also used (company websites, presentations during general meetings, press briefings, analysts' reports, etc.). This work was often carried out over a short period of time (less than 3 years). Because of the manual coding of data and the difficulty of coding a large number of media, few studies, apart from the study by Campbell and Abdul Rahman (2010), have followed longitudinal approaches to our knowledge. Other work has focused on the study of intellectual capital disclosure in a particular country ( It should be noted that there are two types of information disclosed on intellectual capital by companies. The first is information of a narrative and qualitative nature. The second is quantitative information (monetary or non-monetary). All the work focused on intellectual capital reporting suffers from several limitations. The main one is linked to the validity of the coding methods used in this research, which are not very stable and are based on sometimes subjective approaches. This does not make it easy to compare the different works in order to draw definitive conclusions. Several reasons for companies to disclose more information about intellectual capital have been identified in the literature. Among these reasons, we can cite the following:  reduce information asymmetry;  allow the reduction of the cost of capital of the company (Williams, 2001);  improve the liquidity of the stock market;  increase the demand for corporate securities (Diamond and Verrecchia, 1991);  have positive (external) effects on corporate reputation and trust in the management of the company. Based on the above theories, the influence of several factors was tested (size, industry, debt, financial performance, dual listing, age of the company, type of auditor, governance mechanisms). The results showed that the most important determinants of intellectual capital disclosure are size and industry (Whithing and Woodcock, 2011). But more and more research emphasise the role played by governance mechanisms. Indeed, they assume that oversight bodies, including the board of directors, are the main actors in financial communication (Holland, 2006 was not demonstrated. However, the results showed that the size of the board has a significant impact on intellectual capital disclosure. Table 1 presents a summary of the work analysed in this paper relating to the disclosure of information on the IC. They are listed in descending chronological order.

Voluntary disclosure of information by family businesses
The vast majority of research on IC disclosure has been conducted on the basis of listed and often large companies. Researchers believe that no studies have focused on analysing the disclosure of intellectual capital information by family businesses. The case of family businesses in Jordan could be interesting and would allow researchers to understand the practices of voluntary disclosure of information on intellectual capital and their determinants. Before dealing with this issue, it is necessary to briefly define the notion of family business.  (2007), in their empirical study of S&P 500 companies, consider as a family business any entity whose founder and/or his descendants become managers, board members or majority shareholders. This definition is used to distinguish between family and non-family firms in our sample. This will allow the researcher to compare reporting practices between family and non-family businesses. The theoretical framework is based on agency theory. Indeed, research, having applied agency theory to family businesses, has identified agency and rootedness as the two main characteristics that differentiate family from non-family businesses . Altruism and opportunism within a family business can manifest itself through the mobilisation and use of human capital. Indeed, in some cases, family businesses tend to recruit managers from within the family who do not have the required technical and managerial skills. This is not the case for non-family businesses. Chua et al (2003) consider that one of the most important questions that need to be addressed in a theory of the family firm is how and why this form of organization behaves and acts in a way that is distinctive of a non-family firm. These differences could arise from a disparity in vision or goals. They may also be the result of values, resources, capabilities, strategies, style, etc. Salvato and Moores (2010) find that accounting practices in family businesses have been the subject of very little research. As such, they show the value of conducting empirical research on the interface between family businesses and accounting practices. In the previous literature, research analysing the accounting practices of family businesses has focused on the quality of results, the quality of earnings forecasts, governance mechanisms (Ali et al., 2007), earnings forecasts, earnings warnings and the frequency of press briefings (Chen et al., 2008). The main theme of this work is results management and the specificities of family companies with regard to these practices. According to Salvato and Moores (2010), this could be explained by the fact that the literature on voluntary disclosure of information tends to consider shareholders as a homogeneous group. The latter all prefer companies to disclose more voluntary information in order to reduce the asymmetry of information between shareholders and managers. This hypothesis ignores the fact that some shareholders, especially families, who hold a large share of the company's capital, may have a different strategy in terms of extrafinancial communication.  (2007) argue in their study that S&P 500 family firms seem to communicate quarterly earnings forecasts to a greater extent than non-family firms. Chen et al (2008) find that family-controlled S&P 1500 companies prefer to disclose less information on a voluntary basis since they provide few earnings forecasts and press conferences. However, they communicate more earnings alerts to avoid disputes with shareholders and thus penalize their reputations if they do not disseminate bad news in time. This work shows that the results on voluntary disclosure of information by family businesses are contradictory. This can be explained by the diversity of voluntary disclosure practices and the multitude of such objects. The motivations for disclosing voluntary information may be different depending on its nature. Table 2 presents a summary of the work analysed in the literature on the disclosure of voluntary information by family businesses. These works are classified in descending chronological order. Agency theory has been used to explain the relationship between the extent of disclosure and the level of indebtedness of the firm (Abdulrahman and Abdul Hamid, 2012). According to Jensen and Meckling (1976), creditors demand more information about the firm if it is highly indebted. They want to know whether the entity is able to meet its liabilities. This is a way of assessing the risks of wealth transfer to shareholders. The demand for information on IC is therefore probably an increasing function of indebtedness (Escaffre, 2002). In this context, White et al (2007)  This non-exhaustive review of the work dealing on the one hand with IC reporting and on the other hand with the practices of family businesses in terms of voluntary disclosure of information, shows the need to conduct empirical research in order to examine the specificities of IC reporting by family businesses. Family businesses have certain particularities compared to non-family businesses. Shareholders generally take a long-term investment perspective and ensure the transmission of wealth from one generation to the next. Thus, the strategy of voluntary disclosure of intellectual capital information may be subject to counter-arguments. On the one hand, leaders who are often family members prefer to disclose less voluntary information (Chen et al., 2008). On the other hand, shareholders who are family members have easier access to information and have fewer problems of information asymmetry. On the other hand, a good voluntary disclosure policy can reduce the cost of capital, better inform the market and attract new investors. This would allow the company to develop better and not rely heavily on debt to finance its investments. In this context, the application of agency theory to family businesses has a certain peculiarity. Indeed, Ali et al (2007) distinguish between level 1 and level 2 agency problems. Compared to non-family businesses, family firm face fewer agency problems due to the separation between owner/shareholders and managers (type 1 agency ISSN: 00333077 2827 www.psychologyandeducation.net problems). However, they are subject to more agency problems between shareholders with the power to control the company and others (type 2 agency problems). These characteristics specific to family businesses seem to influence voluntary disclosure practices on intellectual capital. The researchers assume that there are significant differences between family and non-family companies listed on the Amman Stock Exchange with regard to the understanding of disclosure of intellectual capital information. The research hypothesis is that the level of disclosure of intellectual capital information in annual reports differs between family and non-family companies.

3.
Research methodology In this section, begins with an explain on how the sample was put together. Secondly, describe the method used to determine the intellectual capital disclosure score. Finally, present the regression model used.

Description of the sample and data collection
The initial sample consisted of companies listed on the Amman Stock Exchange. The final sample is made up of 201 companies, after elimination: companies in the banking and insurance sectors, those whose 2018 reference documents were not available online and those whose financial data were missing. Family businesses were identified on the basis of the criteria of Astrachan and Kolenko (1994) and used by Ali et al. (2007), i.e., any business in which the same family owns more than 50% of the capital in the case of unlisted companies and 10% in the case of listed companies, or those in which one or more family members manage the company or have transferred management to other generations of the family. The classification of the companies in our sample involved several steps: -Firstly, the capital breakdown of each company in the sample from the reference document was retreived. -Then, the main shareholders were identified and added up their shares in the company's capital if they belonged to the same family. In some cases, it was necessary for more precise research on the company's main shareholders. The ASE database was used to identify the owners of financial holding companies who own shares in family businesses. An Internet search was conducted to identify family relationships between certain shareholders. The three co-authors cross-referenced and verified the data collected. -Finally, considering a company to be family owned (coded 1) when the cumulative share of family shareholders exceeds 10% of the company's share capital and some family members sit on the board of directors. Otherwise, the company is considered non-family (coded 0). This phase enabled us to identify 71 family firms and 130 non-family firms. As for the sectors of activity, Bozzolan et al (2006) and Li et al (2008) was considered the sectors related to the internet, biotechnology, entertainment and leisure, high technology, media, software design, integration systems, telecoms and web services, as highly knowledge-based sectors. They are coded with a value equal to 1. The other sectors are coded 0 and cover catering, automotive, chemicals, construction, industry, energy, oil, services, textiles, tourism, etc. The data collection and coding period for the reference documents lasted from January to July 2019. The other financial data (size, performance and indebtedness) were collected from June 2019. Table 3 summarises the composition of our sample  The hypothesis concerns the examination of the existence of a difference in the disclosure of information on IC between family and non-family businesses. This hypothesis was tested on the basis of the volume of information disclosed on IC in the reference documents of the companies in our sample. As such, researchers used the word as the unit of analysis and the number of words as an indicator of the volume of information on IC. For each reference document, researchers identified the words related to the IC. The words concerned are taken from the analysis grid. By adding the total number of words per IC category, we have the total number of words linked to the IC. The latter is divided by the total number of words contained in the reference document. The score (ICDSCORE) is calculated as follows: Human Capital Score + Structural Capital Score + Relational Capital Score Total number of words in the reference document Note that the different scores are calculated as follows: Human capital score = Sum of the words published on each of the human capital items. Structural capital score = Sum of words published on each of the structural capital items. Relational capital score = Sum of words published on each of the items of the relational capital.

4.
Results In this section, the results obtained are presented. These are firstly descriptive statistics and correlations and secondly the results of the model explaining the extent of IC disclosure.

Descriptive statistics and correlation
analysis Descriptive statistics for the variables in the total sample are presented in Table 4 (Panel A). For the subsets (family businesses and non-family businesses), the descriptive statistics of the variables are presented in panels B and C respectively in the same table. The variables are defined as follows: ICDSCORE: disclosure score on the IC, HUCSCORE: human capital disclosure score, STRUSCORE: disclosure score on structural capital, RECSCORE: disclosure score on relational capital, SIZE: logarithm of the total asset, PERF: performance measured by ROI, RDEBT: debt ratio SKBS: 1 if the company belongs to the knowledge-intensive sector and 0 if not.
The results show the extent of disclosure of information in the three IC categories. Familyowned businesses globally disclose more information about IC than non-family businesses. Table 4 also shows that the extent of disclosure differs between family and non-family businesses. The average disclosure scores for Relational Capital and Human Capital are higher for familyowned businesses. On the other hand, non-family businesses compared to family businesses disclose relatively more information on structural capital. This study shows that attributes of relational capital were more widely disclosed than those of structural and human capital. This finding is consistent with those of Bozzolan et al (2003) 2008), family businesses indexed to the Amman Stock Exchange seem to use debt more than non-family businesses. Appendix 2 presents the result of the correlation analysis between the independent variables of the model. No significant correlations were observed between these variables. In order to better demonstrate the existence of a difference in disclosure of IC information between family and non-family businesses, the mean difference test was performed. Table 5 presents the results.  Table 5 shows that there is a difference between the level of disclosure on IC at the global level and at the level of the relational capital category with a level of 5%. Disclosure on human capital appears to be different between family and nonfamily businesses at 10%. On the other hand, no significant difference is found for structural capital between the two types of companies. Table 6 summarizes the results of the regression of the level of IC disclosure and the independent variables. The latter reports an acceptable adjusted R 2 of 16.4%. 0.0000*** * significant at the 10% level, ** significant at the 5% level, *** significant at the 1% level R 2 = 0.184, adjusted R 2 = 0.164, F-value = 8.873, p-value = 0.000 ICDSCORE: total disclosure score on the IC, FAM: Binary variable = 1 if the company is a family business and 0 otherwise, SIZE: total asset log, PERF: performance measured by ROI, RDEBT: Debt ratio, SKBS: Binary variable = 1 if the company belongs to a strongly knowledge-based sector and 0 otherwise

The results of the regression model
The basic assumption is that family businesses have different IC disclosure practices than nonfamily businesses. Descriptive statistics have led to the conclusion that the overall score of familyowned businesses is higher. Difference-of-means tests showed that there is a difference in disclosure at the overall IC level as well as for relational and human capital. On the other hand, no difference was observed for structural capital. The multivariate regression model showed that the family character of the companies in our sample has an influence on the level of disclosure of information on IC. Our main hypothesis is confirmed.

5.
Discussion of the results Table 6 shows that the FAM variable is significant at the 5% level. Concluding, in line with our expectations, that there are differences between the level of disclosure of information at the  (2011) argue that this result is due to the fact that these firms have few physical resources. As a result, they tend to compensate for the lack of resources in the financial statements by disclosing detailed information about the IC. This communication would allow the company to find a kind of legitimacy in its sector. Researchers have also shown that larger firms tend to disclose more information about IC, consistent with the results found by Bozzolan et al (2003) and Hackston and Milne (1996). In contrast, Garcia-Meca et al (2005) found no significant link between IC disclosure and firm size. Based on the empirical results of our research, financial performance as measured by ROI is significantly associated with the extent of disclosure of IC in the registration documents of Amman Stock Exchange companies. The negative direction found means that the best performing companies do not pay attention to this type of publication. This result is not consistent with previous studies that failed to find a link between the extent of IC disclosure and performance As for the level of indebtedness, the results of this study show that the company's level of indebtedness is significantly and negatively correlated with the overall level of IC disclosure. This means that indebted companies disclose less information on IC. Concluding that companies listed on the Amman Stock Exchange seek to resolve agency problems through other means. Previous studies have had mixed results in this respect. Indeed, for the same Australian context, Whiting and Woodcock (2011) failed to find a significant link, while White et al (2007) found a positive and significant link between indebtedness and the level of disclosure on the IC. The result seem to be consistent with the results already obtained by Meek et al (1995) on the relationship between indebtedness and voluntary disclosure in general.

6.
Conclusion The objective of this research is to study the determinants of the level of voluntary disclosure of IC information in the reference documents of family and non-family companies listed on the Amman Stock Exchange in Jordan. Data was collected from the reference documents of 201 companies from different sectors. Firstly,a content analysis was conducted, which enabled researchers to calculate the disclosure score for the different categories of IC for the year 2018. The results showed that family businesses (compared to non-family businesses) disclose more on relational capital and human capital, but less on structural capital. Second, the regression results showed that family businesses appear to report more on IC in annual reports than non-family businesses. In addition, industry and size play a rather important role in communication in the reference documents of family and non-family companies. Indeed, larger companies and companies belonging to sectors based on the intensive use of technology and knowledge pay more attention to IC. Performance www.psychologyandeducation.net and indebtedness have a negative impact on the extent of disclosure on IC. This research is a continuation of the work carried out on the disclosure of information on IC. Understanding the link between the level of disclosure and its determinants should make it possible to improve knowledge of companies' behaviour in terms of voluntary disclosure of nonfinancial information. As with any research work, this article suffers from a number of limitations. The first concerns the size of the sample, which was limited to family businesses listed on the Amman Stock Exchange. However, the sample is comparable to the majority of work dealing with IC disclosure and using content analysis of annual reports as a basis for coding. The second limitation concerns the use of reference documents as the sole medium for IC disclosure. Other media are also relevant to analyse (web pages of company sites, financial analysts' reports, press articles, etc.). However, researchers believe that the reference documents, due to their regularity and the rules to be respected in the drafting of this type of support, seem to be the most adequate sources of information to assess the level of disclosure on the IC in Jordan. The third limitation concerns the measurement of the disclosure score on the IC. In this paper, researchers focused on the volume (quantity) of disclosure and not on its quality. It is recognised that the evaluation of the quality of disclosure is a complex subject that integrates other aspects of the company that go beyond the items identified in the IC analysis grid. To date, measuring the quality of voluntary disclosure remains an unresolved issue (Cerbioni and Parbonetti, 2007). The final limitation concerns the use of the word as a unit of analysis to assess the level of disclosure on the IC. This technique allowed researchers to quantitatively measure the importance of each item and therefore of each IC category in the annual report. However, this measure does not take into account the difference between qualitative (or narrative) and quantitative information. The calculation of a weighted score according to the nature of the information would be more relevant to assess the level of disclosure on the IC.