OWNERSHIP STRUCTURE AND RISK DISCLOSURE: A STUDY OF MALAYSIAN LISTED COMPANIES

This research focuses on the importance of ownership structure as a determinant of risk disclosure. It is expected to contribute to the literature particularly in the Malaysian context, where risk disclosure practice is in the infancy stage. This study uses multiple regressions in assessing the variability of the extent of risk disclosure. The overall results confirm that highly concentrated ownership would lead to high agency problem, which then leads to less disclosure. This implies that, to promote greater transparency in countries where many of the large listed companies are family-owned, more stringent laws that mandates adequate risk disclosure is clearly warranted. This would ensure that the needs of all stakeholders are properly met.


1.Introduction
Prompted in great measure by the exponential growth in information and communication technology and rapid change in global markets, coupled with shifting demographics and the homogenization of personal and organizational values, there has been an increased demand for information on risks by the stakeholders of firms.Firms today are pressured to report on risks to enable the stakeholders to quickly recognize, react and adapt to changing global catalysts.Information on how adept a company is in comprehending and managing threats and related risks, moreover, enables the stakeholders to assess the ability of top managers and other employees to manage risks successfully and pursue business opportunities (Korosec and Horvat, 2005).In fact, risk disclosure has become an integral part of good corporate governance and such information is expected to be increasingly sought by the firm's stakeholders and other users.
The collapse of many well known firms, such as Enron in the US, illustrates how shady disclosures can lead to disastrous results.With more similar cases appearing, researchers are motivated to study and examine risk reporting in a more global context (e.g., IFAC, 1999; Kajuter, 2001;Shrives & Linsley, 2003).
Risk disclosure undoubtedly benefits all stakeholders.For one, shareholders can use the information to assess future gains from their investments, the suitability of management, and the reasonableness of keeping on providing capital to the company.Creditors can benefit from the information by being able to assess the credit worthiness of the firm and its ability to settle financial liabilities in the future.On the flip side, risk reporting can expose the firm to scrutiny by its competitors in their bid to improve their own competitive positions.
Risk reporting is usually done through the prospectus and annual reports, the principal regulatory requirements.The prospectus normally covers a full range of relevant risk types and a consideration for prospective or potential investments, while the annual report provides both qualitative and quantitative risk information in the financial statements (mainly in the footnotes) or in the regulator's requirement sections, namely the Management Discussion and Analysis (MD&A) (Lajili and Zeghal, 2005).The MD&A, in particular, is intended to give the investors an opportunity to look at the company through the eyes of management (SEC, 1999, cited from Cole et al., 2005).It helps stakeholders by clarifying and validating the quantitative measures contained in the annual report, thereby helping them make wiser and better judgment.
While many countries demand firms to disclose risks in the MD&A, the Malaysian regulatory authorities have yet to explicitly do so.As such, MD&A in Malaysia is still very much unexplored.In lieu, many Malaysian public companies include descriptive information in their annual reports under headings such as -Chairman Statements‖ and many more.

2.Problem Statements and Objectives of Study
Unlike many developed countries such as the US and the UK (Linsmeir et al., 2002;Roulstone, 1999;Venkatachalam, 1996) where there are specific mandatory requirements requiring companies to report risks, Malaysia's experience in risk disclosure practice is at best in infancy stage.This is notwithstanding the fact that there are regulations imposed by the Bursa Malaysia on Malaysian companies to do so.Preliminary observation suggests that disclosure made by Malaysian companies relate to financial matters and future prospects.To date, however, there has not been any extensive study done on this.It would be interesting therefore to undertake an investigation into the current local risk disclosure practice to understand not only the existing state of affairs but also the factors that prompt local corporations to do so as well as the nature of risks reported.
Risk reporting is useful not only to the company's shareholders but other stakeholders as well.Nevertheless, the information passed on to the stakeholders is controlled by the owners.In this regard, the degree of ownership will have an impact on the extent of reporting since it affects the agency set-up to varying degrees.In the instance when ownership is diffused, as typical in the US and the UK, agency related problem would arise from the conflict of interest between outside shareholders and managers who own an insignificant amount of equity in the firm (Berle and Means, 1932; Jensen andMeckling, 1976 andRoe, 1994).If the ownership is concentrated to a level at which an owner obtains effective control of the firm, as in the case of East Asian and most other countries apart from the US and the UK, the agency problem manifests through the conflict between controlling owners and minority shareholders (Fan and Wong, 2002).
Different ownership structures therefore inevitably affect the extent of risk disclosure due to these conflicts.Thus this study aims to examine the influence of ownership structure toward risk management disclosure in the non-financial section or the narrative part of the annual report.This study includes size and level of risk (proxy by leverage) as the control variables.

2.1.Usefulness of Risk Management Disclosure
Studies conducted in the US to determine the usefulness of risk management disclosure are mostly market-based in nature.Examples of such studies are Linsmeir et al., (2002) and Venkatachalam (1996).These studies investigate the relationship between risk disclosure and the interest rates, foreign currency exchange rates and commodity prices.The study by Linsmier et al, (2002) found strong evidence of the usage of risk disclosure by investors.The researchers discovered that risk disclosure has the impact of reducing uncertainty and the diversity of opinions on the effect of market rate on the firm value.
In another study, Rajgopal (1999) examined the association between commodity price risk disclosure and market view of oil and gas price sensitivity.The study concluded that such disclosure proves to be reliable indicators of price sensitivity that can help investors make wise judgment.
There are also a number of researches that look specifically at the MD&A as to whether it provides information that is valuable to the stakeholders, especially investors.The study of the retail industry by Cole and Jones (2004), for example, found that disclosure in the MD&A is useful in forecasting future revenues and returns.In this regard, historical information on capital expenditure and store openings can help investors make informed decisions regarding the firm's future.Additionally, the researchers found that store sales growth information has the power to predict future incremental sales and stock returns.
In concluding this part of the discussion on the usefulness of risk management disclosure, it is worth noting the finding by Berreta and Bozzolan (2004) to the effect that institutional investors strongly demand increased corporate risk management disclosure to improve their investment decisions.

2.2.Ownership Structure and Risk Disclosure
Corporate ownership is highly concentrated in East Asia (Ghazali and Weetman, 2006).According to Claessens, Djankov, and Lang (2000), more than two thirds of companies in East Asia are owned by single shareholders with more than half of them family owned.As a point of fact, owner managed companies are rampant in Malaysia.According to Claessens et al. (2000), at the 20 percent cut off level, 67.2 percent of Malaysian public listed companies are in the hands of family members and out of this, 85 percent have owners as managers.This is also about the case in Indonesia and Thailand.When the controlling owner oversees the accounting reporting policies, they are perceived to have strong reasons and opportunities to hold up minority shareholders, in which case restricted accounting information may very well be expected (Fan and Wong, 2002).In such a scenario, the credibility of the reports is open to suspect.
In addition, Chen et al., (2006) in their study based on 4,415 firm-years from the S&P1500 firms in the period 1996-2000 found that family-owned firms tend to provide less voluntary disclosure of both good and bad forward-looking information as compared to non-family firms.They also reported that the potential entrenchment of the founding family, as proxied by extremely high family ownership or a dual-class share structure, leads to even less voluntary disclosure (which may include information on risk management).
One of the ways to reduce the controlling power is by dispersed ownership.Dispersed ownership corporations are characterized by a large number of owners but with no single dominant owner or owners that can significantly affect control of them.According to Schipper (1981), the problem of monitoring in an organization can be solved by increasing the number of owners.As the number of shareholders increases, disclosure also tends to increase if it can solve the monitoring problem.Cooke (1989) argued that when companies have large numbers of shareholder, the firm will be more likely to provide additional information to satisfy the needs of all its shareholders.
Most of Southeast Asian companies are controlled by the government.State control is significant in Indonesia, Korea, Malaysia, Singapore and Thailand.In Malaysia, privatization of government agencies was undertaken not only to meet the objectives of reducing the federal government financial burden but also to spur the socio-economic development of the country as a whole.Thus companies such as Sime Darby Berhad, Telekom Malaysia Berhad and Tenaga Nasional Berhad were established and controlled by the government.
The government invests in privatized companies through its investment arm, Khazanah Holdings or other related agencies such as Permodalan Nasional Berhad (PNB).In fact, the government appoints its representatives in some of the companies to mitigate company aspiration from becoming solely profit-oriented.Being government controlled, these companies need not attract potential investors because they can easily obtain funds from local banks.For these reasons, these companies may disclose less their risks in the reports.In their study of the Indonesian experience, Leuz and Oberholzer-Gee (2003) found that Indonesian companies that were politically linked to the former President Suharto had less publicly traded foreign securities.
One of the primary means of boosting the economy is to tap Foreign Direct Investments (FDI) into the country.According to Sumiani and Haslida, (2006), the Malaysian government's focus on FDI inflow has served the country well by enhancing the export sector and stabilizing the economy.To ensure a continuous flow of these funds, the government had gone to great lengths to create an attractive and conducive environment.Topmost among these initiatives are the creation of a stable political and social climate, the provision of various tax and pioneer incentives, the supply of skilled and semi-skilled labour and the placement of companies in strategic locations.Many large Multi National Companies (MNCs) like Intel Corporation, Motorola Berhad, Shell Refining Company and British American Tobacco (Malaysia) Berhad have since located their operations in Malaysia.
This reliance on FDI has a great implication on the way Malaysian companies conduct their business and do their reporting.Since most of the MNCs in Malaysia are subsidiaries of companies from countries like the USA and the UK where there is a heavy emphasis on good corporate governance (Che Haat, 2006), these companies are compelled to follow the legislation of their parent companies' home countries.What this means is that a high dependency on wide and foreign networks has the impact of making the local companies mimic the reporting practices of their foreign counterparts in the annual report.
Prior research by Ghazali and Weetman (2006) found director ownership (manager who owns a large number of shares in the firm) to be more significant than government ownership.The study was conducted to examine the level of awareness on disclosure as a tool of corporate governance in Malaysia after the economic crisis and whether an insider of the corporation influences voluntary disclosure.By using the annual reports of 87 companies for the year 2001, the researchers probed for disclosures based on an eleven criteria disclosure checklist.They found director ownership to be significant at 1 percent in explaining all types of information while government ownership is not significant in any category of disclosure, indicating that government ownership is not influencing disclosure level.Furthermore, they found a negative relationship (significant at 5 percent) for the proportion of family members on the board and the extent of disclosure.This implies that companies with a higher ratio of family members on the company's board and a higher proportion of shares held by executive directors disclose less voluntary information in their annual reports.In terms of disclosure categories, the amount explained varies from 16.6 percent in the case of corporate social responsibility to 21.5 percent for strategic information and 26.4 percent for financial information.
In the most recent study on factors that influence risk disclosure by Abraham and Cox (2007), the researchers examined the relationship of ownership, governance and US listing characteristics to the amount of risk disclosure.The study found a negative relationship between institutional ownership and extent of disclosure, suggesting that institutional investors react negatively to risk disclosure.In terms of governance, two variables namely, the number of executive directors and independent directors, were found to have an impact on the extent of risk disclosure.In addition, the study found that UK companies which are listed in the US listing exchange tend to disclose more.

3.1.Agency Theory
There are many theories that can be associated with corporate governance issues.The Legitimacy theory and Stakeholder theory are examples of widely used theories in this area.Nonetheless, from a financial economist's perspective, corporate governance deals with an agency problem that arises from the conflict of interest between managers and shareholders (Hart, 1995).This research therefore uses the ownership characteristics of Malaysian Public Listed Companies to capture empirically the extent of the agency problem.Theoretical and empirical findings usually consider concentrations of ownership and insider ownership as the main determinants of corporate governance (Ballesta and Garcia-Meca, 2005).As ownership concentration increases to a level where an owner obtains effective control of the firm, as is the case of East Asia, the agency problem shifts away from manager-shareholder conflict to conflict between the controlling owner and minority shareholders (Shleifer and Vishny, 1997).This situation is in stark contrast to US and UK where organizations there tend to be diffusely owned (Fan and Wong, 2001Wong, , 2005)).Agency problems associated with the separation of ownership and control have been the subject of considerable empirical research but none had yet looked into the disclosure of risk reporting.
While ownership concentration mitigates the agency conflict caused my managerial expropriation, it nonetheless creates a different kind of agency problem as described above-the potential expropriation of minority shareholders and other stakeholders by controlling shareholders (Guedhami and Pittman, 2006).These dominant shareholders usually exert full control over managers and frequently hold control power in excess of their cash flow rights, providing them with strong incentives to extract private benefits at the expense of minority shareholders (La Porta, Lopez-de-Silanes, and Shleifer, 1999).According to Shleifer and Vishny (1997, p. 758), -large investors represent their own interests, which need not coincide with the interests of other investors in the firm, or with the interests of employees and managers‖.This will directly or indirectly affect the dissemination of information to the minority shareholders.

3.2.Dispersed or Concentrated Ownership
Corporate ownership in East Asian countries, including Malaysia is concentrated in the hands of large owners or controlling owners (Ishak and Napier, 2006).Because controlling owners are in a position to influence the management on firm strategy and goals INCLUDING THE FIRM'S FINANCIAL REPORTING, THEY ARE perceived to have strong opportunistic incentives to hold up minority shareholders and drive down the quality of accounting information (Fan and Wong, 2002).Controlled ownership would even lead to the unfair situation where minority interest is expensed out in favour of the maximization of the owner's individual wealth.High ownership concentration and the prevalence of family owned business in many emerging market countries would invariably mean that public disclosure is less developed in these countries because insiders are closely informed about the company's financial position and activities (Ghazali and Weetman, 2006).On the other hand, dispersed ownership is an effective method in reducing the power of concentrated ownership.Dispersed ownership will monitor management through demand for more disclosure.Hence, based on the literatures, it is proposed that: Hypothesis 1: Ceteris paribus, there is negative relationship between concentration of ownership and extent of risk disclosure Hypothesis 2: Ceteris paribus, there is positive relationship between dispersion of ownership and extent of risk disclosure

3.3.Foreign Ownership
Andrew et al. (1989) in their explanation of size factors noted that large companies in developing countries are normally foreign owned and because of their greater visibility, they are more likely to be subjected to scrutiny by the host government.Thus, greater social commitment and disclosure in the annual reports is one way of overcoming possible criticisms.Moreover, the foreign firms tend to bring in their own culture and management style to the local host.Investors to Malaysia come from developed nations such as the US and the UK and they will place a high premium on transparency.For these reasons, disclosure of risks in the annual report will be higher in the case of firms that are controlled by foreign ownership.Therefore, it is proposed that: Hypothesis 3: Ceteris paribus, there is positive relationship between the level of foreign ownership and extent of risk disclosure

3.4.Government Ownership
In Malaysia, government ownership is a particular feature.Ownership by government institutions or government-controlled bodies may exert pressure on the affected companies to disclose additional information because the government is accountable to the public at large (Ghazali and Weetman, 2006).Conversely, some researchers argue that companies with government ownership may not provide full disclosure because of separate monitoring by the government.In fact, companies controlled by the government may have little incentive to disclose detailed information because of the government's backing and guaranteed returns (Naser and Nuseibah, 2003).
According to Eng and Mak (2003), agency cost is higher in government owned or linked companies.This is due to the conflict of objectives between the profit takers and those with the nation's interest in mind.Due to this dilemma, the need to communicate with the rest of the shareholders is greater.
As the ultimate body in the nation, the government must play the role of a good model for other companies to follow.Due to this, government controlled companies may disclose more to the shareholders.Based on the literatures, it is proposed that: Hypothesis 4: Ceteris paribus, there is positive relationship between the level of government ownership and extent of risk disclosure

4.Methodology
The unit of analysis for this study is the annual report of the public listed company on the Bursa Malaysia.The sample of companies was drawn from the annual reports of listed companies on the Main and Second Board of Bursa Malaysia for the year 2005.A total of 100 companies were selected randomly, which comprise of 70 from the main board and 30 companies from the second board.Refer to appendix C for the list of companies.

4.1.Method of Analysis
The method used in this study to analyze risk disclosure is content analysis.It was chosen since the study focuses on the extent or amount and not the quality of the risk disclosures.Content analysis is also the most common and widely used method in assessing disclosure [Gray et al., (1995), Hackston and Milne (1996), Haniffa and Cooke (2002), Raar (2002) and Amran (2006)].Weber (1990) defined content analysis as a research method that uses a set of procedures to make valid inferences from text.Weber added that the rule of this inferential process varies based on the interest of the investigator.This research technique enables a replicable and valid inference from data according to the context (Krippendorff, 1980).In order to ensure the replicable manner of inference, a set of interrogation instrument, checklist and decision rules, was developed.The checklist and decision rules used in this study are the ones developed by Linsley and Shrives (2006).Please refer to appendix A and B for further details.The same method was replicated in order to classify whether the information in the annual report is about risk or not.As highlighted in the earlier section, this study focuses only on the non-financial section or the narrative part of the annual report.
To ensure reliability of the coded output, the coder underwent a short period of training to master the checklist and the decision rules.The coder was also exposed to different examples of the various types of risk information.Later, the coder's understanding and skill was tested by using inter-rater or inter-observer method where two coders are involved in analyzing the same set of material.In this case, the two persons involved are the coder and the researcher and they analyzed five sets of annual report.The results of the content analysis done by both coders were than correlated to determine the extent of agreement.The result showed that there were no significant differences between the scores.Gray et al. (1995b) raised a big concern on the unit of analysis used to determine the amount of disclosure.Milne and Adler (1999) proposed the use of -sentence‖ as a basis for coding which is far more reliable than other units of analysis.Further, although most of the studies use sentences for coding, the use of word or area of page (e.g.tenths or one hundredths) to measure the disclosure amount is common.Using word or areas of page as a basis to measure disclosures complicates reliability.Milne and Adler (1999) and Linsley and Shrives, (2006) criticized the use of words since, by themselves, words do not convey any meaning unless referred to the sentences for their proper contexts.Moreover, it is recondite to decide which words are considered as risk disclosure (Linsley and Shrives, 2006).
Likewise, using a plastic grid sheet over a body of text and trying to code the contents of each square would also result in meaningless measures.This method may have the advantage of including charts or graphs into the analysis but it is also exposed to lots of noise introduced when unnecessary pictures or different fonts, column or page sizes are used in the annual report.Hackston and Milne (1996) made use of all three measures and found that they produced the same results-significant correlation between the three measures.Hence, based on the above argument, it appears that using sentences as a basis to code and count the content of risk disclosure could serve the purpose of this study.The same method had also been employed by Linsley and Shrives (2006).
The measurement for the other variables used in this study can be found in the following table.

4.2.Data Analysis
This study uses multiple regressions in assessing the variability of the extent of risk disclosure.This statistical method has been widely used in previous researches (Hackston and Milne, 1998;Cooke, 1998;Haniffa andCooke, 2002 andAmran, 2006).Based on the above discussion of dependent and independent variables, the following regression model was developed: Total sentence of risk management disclosure = α 0 -β 1 CONS + β 2 DISP + β 3 GOV + β 4 FOR + β 5 SIZE + β 6 LEV + ε Where: Error term The correlation matrix was reviewed and the variance inflation factors (VIF) computed to detect whether there was multicollinearity problem.Further analysis to see whether the multiple regression assumptions have been violated was also carried out.The normality, linearity and homoscedasticity assumptions were determined based on the analysis of residuals, plots of the studentized residuals against predicted values, and Q-Q plot.
The above analyses showed that the untransformed data violated the multiple regression assumption on normality, linearity and homoscedasticity and none on multicollinearity effects.In addressing the above violation and in order to ensure the rigorousness of the regression test, the data was then transformed into normal data and re-checked for violation.The problem was found to have ceased.

5.Results
Table 2 provides descriptive statistics of the total sentences of risk management disclosure and the continuous independent variables.It is noticeable that total sentences for risk management disclosure ranges from the minimum of 3 sentences to the maximum of 78 sentences.On average, an annual report has 20 sentences devoted to the discussion of risk.The minimum of 3 sentences are attributed mainly to the Second Board companies.On average, a company spends only one page in their annual report as depicted by the mean.The measure of dependent variables was found not to be normally distributed as indicated by the standard test on skewness and kurtosis 3 .Similarly, the continuous independent variables were found not to be normally distributed.As such, the dependent and continuous independent variables were transformed to normal scores before conducting the regression analysis

Table 2 INSERT HERE
The next test conducted in order to ensure the rigorousness of the regression was the multicollinearity, homoscedasticity and linearity tests (Haniffa and Cooke, 2005).Table 3 presents the correlation matrix for the dependent variable and continuous independent variables.The table provides strong justifications that multicollinearity is not a problem 4 and so are the homoscedasticity and linearity.

Table 3 INSERT HERE
Table 4 summarizes the results of the multivariate regression model using normal scores.The adjusted R 2 for the regression model was 0.500.This implies that the amount of variability in risk disclosure explained by the variable chosen in this study is quite large.The regression result shows that dispersed ownership has significant influence on the extent of risk disclosure as compared to concentrated ownership.In addition, the result shows that there is a negative relationship between concentrated ownership and extent of disclosure.Government ownership is shown to be significant at 10%.This indicates that a company with higher government ownership tends to disclose more on risk.Size variable is significant at 5% level.The result is expected as it is consistent with Linsley and Shrives (2006) and most of the other disclosure based studies conducted in the Malaysian context (Thompson andZakaria, 2004, Amran, 2006).Leverage, which is the popular proxy for risk is shown as insignificant.However, the positive correlation between the two variables is consistent with the hypothesis.

6.Discussion
In the wake of recent corporate scandals that involve fraudulent commission and omission, stakeholders have begun to demand from companies ever more transparency of their dealings.One aspect of this disclosure pertains to risk information.Risk disclosure is undeniably beneficial to stakeholders and this fact has been attested to by previous studies.Linsmier et al. (2002), for example, have proven that such information does influence shareholders' decision.4 The rule of thumb for checking problems of multicollinearity is when the correlation is >0.800 (Gujarati, 2003, Haniffa andCooke, 2005).
Given this importance of risk disclosure to the stakeholders, it is equally important to find out the factors that influence the extent of such disclosure.
This study was done on the premise that ownership structure impacts the company's disclosure practices.Specifically, this study looks at concentration of ownership and foreign and government ownerships for possible impacts on the extent of disclosure.Several hypotheses were developed using the perspective of the Agency theory.
The overall regression result indicates that ownership structure does have a significant influence on the extent of risk disclosure.It confirms that highly concentrated ownership would lead to high agency problem, which then leads to less disclosure.This is evident from the negative relationship depicted by the regression result and is consistent with many earlier propositions (Ishak and Napier, 2006;Fan and Wong, 2002).Companies with high concentrated ownership may be averse to balanced power sharing.A shareholder with high ownership may yield high controlling power and operate the company independently.Where the agency conflict lies between majority and minority shareholders, the latter may be deprived from making their best decision.Where ownership is more dispersed, there would be less agency conflict and greater risk disclosure may ensue.This is again supported by the regression result.Company with dispersed ownership would reduce the controlling power embedded in any single shareholder, thereby engendering better corporate governance practices and systems, including better reporting.
A company with significant government ownership is expected to have more extensive risk disclosure since it will be the focal point for emulation by other companies.Furthermore, the government has to walk its talk on transparency and what better way there is than practicing it in its own holding company.As borne by earlier findings (Shrives and Linsley, 2006;Beretta and Bozzolan, 2004;Ahn &Lee, 2004 andClarkson et al., 1994), size does in fact matter.
From the above discussion, a conclusion can be drawn that in order to promote greater transparency in countries where many of the big listed companies are family owned, more stringent laws that mandates adequate risk disclosure is clearly warranted.This would ensure that the needs of all stakeholders be properly met, even the minority shareholders.

7.Conclusion
The primary intention of this study is to test whether ownership structure has any impact on the extent of risk disclosure.The companies involved in this study are listed companies on the Bursa Malaysia.Based on the regression result, ownership structure does influence the extent of risk disclosure.The findings indicate that companies with high concentrated ownership disclose less.This is consistent with the explanation given by the Agency theory.
Based on the findings, it is wise for the relevant regulatory bodies to think about the proper mechanism that could help alleviate this problem.By introducing laws to mandate greater transparency, for example, they will be helping minority shareholders and other stakeholders make more informed decisions from the additional materials disclosed.

7.1.Limitation
This study is not without its limitations.Being based purely on content analysis, it has the disadvantage of possibly overlooking the quality of the reporting.Even though content analysis is one of the most acceptable and oft used approaches, undeniably the quality of expressions used to report the risks are as equally important.Hence, we recommend that studies be undertaken in the future to analyze risk disclosure from the qualitative aspect.company in the future or of the management of any such opportunity, prospect, hazard, harm, threat or exposure. 3) The risk definition just stated shall be interpreted such that "good" or "bad" "risk" and uncertainties will be deemed to be contained within the definition.

4)
The type of risk disclosure shall be classified according to the Appendix A.

5)
If a sentence has more than one possible classification, the information will be classified into the category that is most emphasized within the sentences.

6)
Tables (quantitative and qualitative) that provide risk information should be interpreted as one line equals one sentence and classified accordingly.

7)
Any disclosure that is repeated shall be recorded as a risk disclosure sentence each time it is discussed.

Table 2 .
Descriptive Statistics of Total Sentences and Continuous Independent Variables

Table 3 .
Correlation Matrix Correlation is significant at the 0.05 level.** Correlation is significant at the 0.01 level . *

Table 4 .
Regression Results