A transparency disclosure index measuring disclosures: Chinese ...

J. Account. Public Policy 29 (2010) 259–280

Contents lists available at ScienceDirect

J. Account. Public Policy journal homepage: www.elsevier.com/locate/jaccpubpol

A transparency Disclosure Index measuring disclosures: Chinese listed companies q Yan-Leung Cheung a,*, Ping Jiang b, Weiqiang Tan a a b

School of Business, Hong Kong Baptist University, Hong Kong School of International Trade and Economics, University of International Business and Economics, Beijing, China

a r t i c l e

i n f o

Article history:

JEL classification: G3 G34 Keywords: Corporate Governance Disclosure China OECD Principles

a b s t r a c t This study addresses the question whether transparency matters among Chinese listed companies. We construct a comprehensive scorecard, based on the OECD Principles of Corporate Governance, in order to assess the transparency of 100 major Chinese listed companies. Based on the scorecard, we construct a Transparency Index (TI) which is used to assess these major Chinese listed companies during 2004–2007. The results reveal that there is a positive and significant relation between company transparency and market valuation. When we further split the TI into Mandatory and Voluntary Disclosure Indexes, it is found that market valuation is only related to the Voluntary Disclosure Index. Finally, we find that more profitable, overseas-listed, and companies with a separate CEO and board chairman tend to disclose more on a voluntary basis. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Corporate disclosure has been identified as one of the most fundamental elements contributing to good Corporate Governance.1 Availability of information is essential to minimize the information asymmetry between insiders and outsiders and to allow general investors to assess company performance. Information disclosure can be divided into two categories: mandatory and voluntary. There are strict regulatory requirements for mandatory information disclosure of listed companies. These q

This paper has benefited from comments from Martin Loeb (the editor) and two anonymous referees. * Corresponding author. Tel.: +852 3411 7520; fax: +852 3411 8051. E-mail address: [email protected] (Y.-L. Cheung). 1 A literature review on the importance of corporate disclosure can be found in Healy and Palepu (2001).

0278-4254/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jaccpubpol.2010.02.001

260

Y.-L. Cheung et al. / J. Account. Public Policy 29 (2010) 259–280

include some basic accounting, financial, and operational information. Some jurisdictions may require companies to disclose non-financial information as well, including Corporate Governance practices. More voluntary disclosure increases the transparency of the company that reduces the information asymmetry between insiders and outsiders. This could promote the management accountability and reduce the monitoring costs of investors. While this type of disclosure may not be mandatory, it is recommended as best practice. The voluntary aspect allows management discretion in deciding the content of information to disclose. An important question is if disclosure is important and if investors are able to make better investment decisions with more information? Will companies be rewarded for more disclosure, particularly for more voluntary disclosure? In this study, we propose to examine the disclosure practices of Chinese listed companies. This study has three objectives. The first is to examine the relationship between company disclosure and market valuation. We develop a Transparency Index to measure the quality of disclosure of Corporate Governance practices of Chinese listed companies. The Transparency Index is based on the five OECD Principles of Corporate Governance (OECD, 2004).2 The five OECD Principles are: the rights of shareholders, equitable treatment of shareholders, the role of stakeholders, disclosure and transparency, and board responsibilities and composition. Cheung, Jiang, Limpaphayom and Lu (2008) design a Corporate Governance Index consisting of 86 criteria with which to assess the Corporate Governance practices of Chinese listed companies.3 Within the Corporate Governance Index, there are 56 criteria (see Appendix 1) that are related to information disclosure. We use these 56 criteria to construct the Transparency Index which is used in our analyses. Our second objective is to examine the difference between mandatory and voluntary information disclosure. Among the 56 criteria, there are criteria that companies are required to disclose according to the regulatory framework and others that are up to management discretion. We split the Transparency Index into two sub-groups. The first sub-group consists of disclosure items the company must disclose according to the regulatory framework and the second sub-group consists of disclosure items the company discloses on a voluntary basis. We are thus able to assess the impact of these two levels of disclosure requirements on company valuation. The final objective of our study is to identify characteristics of Chinese listed companies that are more transparent.4 This has implications for investors in making stock selections. China is the largest emerging economy in the world. It has experienced huge economic growth during the last two decades. Using purchasing power parity exchange rates, the Chinese economy is the second largest in the world and at current growth rates it may become the largest in less than 10 years (Allen, Qian, and Qian, 2005). The combined market capitalization of the two Chinese mainland stock exchanges makes them the third largest in Asia, following Tokyo and Hong Kong. Recently, there has been rising interest in investing in China and increasing numbers of Chinese firms that are cross-listed in Hong Kong, the US and Singapore, making their stocks available to international investors. International fund management houses have been launching various investment products that either invest only in China or have an explicit policy of investing a fixed portion of their portfolio in China.5 There is

2 Since its introduction in 1999, the OECD’s Principles have been widely used and are becoming the most recognized international benchmark for governance practices. Currently, the OECD’s Principles are widely used by investors, stakeholders, companies, and policy makers worldwide. One of the reasons for its popularity is that the OECD’s Principles were not created to be prescriptive or to dictate corporate governance practices. Rather, the principles are flexible and adaptable to each nation’s unique culture, history, legal system, and level of economic development. The Principles provide a framework guiding the development of corporate governance infrastructure and practices within an economy. 3 Nankai University has been compiling Corporate Governance Index for all A-share companies listed in China. However, they only disclose scores for the top 100 companies. They do not disclose score for other companies. The major limitation of this study is that we cannot have the full coverage of all A-share companies listed in China. 4 Ahmed and Courtis (1999) perform a meta-analysis of 29 companies on the relationship between corporate characteristics and their disclosure level. They confirm a significant and positive association between disclosure levels and corporate size, listing status and leverage. 5 According to the Hong Kong Investment Funds Association (www.hkifa.org.hk), the gross sales of China equity funds have increased from US$184 million in 2002 to US$659 million in 2005 and eventually to US$10,755 million in 2007. This represents a 58-fold increase from 2002 to 2007.


261

a need for the international investment community to acquire a better understanding of the disclosure practices of Chinese enterprises.6 This paper makes two contributions to the literature. First, this study develops a comprehensive set of disclosure criteria recommended by the OECD Principles of Corporate Governance (OECD, 2004). These criteria are then used to create a TI to measure the overall quality of disclosure practices of the 100 major Chinese listed companies from 2004 to 2007.7 The information that is used for assessment includes annual reports, articles of association, memorandums of association, notices to call shareholders’ meetings, annual general meeting minutes, company websites, analyst reports, and other sources available to the general public. Second, the disclosure practice of our sample companies is rated several times during 2004–2007. The 4-year data set and the use of a fixed effects regression model for panel data minimizes the endogeneity problem found in cross sectional studies based on a single year’s data. Our empirical results offer evidence that information disclosure matters for Chinese listed companies. Specifically, we observe that there is a positive and statistically significant relation between company valuation and the quality of disclosure practice, as measured by the Transparency Index, even after the inclusion of variables to control for firm characteristics, board characteristics, and ownership structure. We further find that company valuation is positively associated with companies that disclose more information on a voluntary basis. Finally, Chinese listed companies that are larger and dually listed overseas tend to be more transparent on a voluntary basis. The paper is organized as follows. Section 2 reviews previous research on the factors that affect voluntary disclosure by companies. Section 3 gives a brief description of the development of China’s equity market. Description of the data and empirical methods are presented in Section 4. Section 5 presents and discusses the empirical findings and the last section concludes the study.

2. Literature review Disclosure is important to mitigate asymmetric information and agency problems. A comprehensive literature review on disclosure is provided by Healy and Palepu (2001). Among many other research questions, they discuss in detail the following issues: (i) the role of regulation; (ii) management reporting decisions, i.e., voluntary disclosure; and (iii) the consequences of disclosure. Our paper addresses these three issues. Regulatory authorities around the world have set up various regulations governing corporate disclosure. Two questions are raised regarding regulations: what is the economic rationale behind regulations and how effective are regulations in solving the information and agency problem. Leftwich (1980), Watts and Zimmerman (1986), and Beaver (1998) propose that regulators should reduce the information gap by providing a ‘level playing field’ between sophisticated and unsophisticated investors by creating minimum disclosure requirements. Research on management reporting decisions has focused on two areas: positive accounting theory and voluntary disclosure. The positive accounting empirical literature focuses on whether management makes changes to the accounting method or uses accrual estimates for private benefit, such as for determining compensation bonuses. The second area focuses on the motivation of management to voluntarily disclose. Researchers discuss various factors that affect management disclosure decisions. For example, according to Healy and Palepu (1993, 1995), managers who anticipate debt or equity issuing have an incentive to voluntarily disclose. By voluntarily disclosing, there is a reduction in asymmetric information and a consequent reduction in the firm’s cost of external financing. Skinner 6 The importance of disclosure in China can be illustrated by an event reported in Business Week on December 1, 2003. In 1998, Zhu Kuan, a company controlled by the government of the city of Zhuhai, defaulted on US$750 million borrowed from Standard Chartered, Morgan Stanley, Lehman Brothers and others. In 2003, during negotiations for a workout, creditors discovered that the Zhuhai government had transferred land worth US$125 million out of Zhu Kuan’s control and back into the hands of the city (land that the creditors had assumed to serve as collateral for their loans). 7 Our sample companies consist of the 2004 and 2007 Fortune 100 largest listed Chinese companies. The Fortune ranking is based on the total revenue of all Chinese listed companies.

262


(1994, 1997) finds that firms with bad earnings news are more than twice as likely to pre-disclose the bad news than are firms with good news. In addition, he finds that firms with negative earnings are more likely to be subject to litigation. Aboody and Kasznik (2000) show that prior to stock option award periods, firms often delay disclosure of good news and accelerate the release of bad news, in order to increase management stock-based compensation. Both the positive accounting theory and voluntary disclosure research have examined the economic consequences of changes in disclosure. The former has focused on the effects of changes in accounting standards or methods while the latter has focused on the capital markets response to changes in corporate disclosure. In the positive accounting theory literature, studies indicate that there is generally no significant relation between stock returns at the announcement of the accounting standard change and contracting or political cost explanations (Holthausen, 1981). In the voluntary disclosure literature, studies argue that there are three potential outcomes for firms making extensive voluntary disclosure: improved liquidity (Diamond and Verrecchia, 1991; Kim and Verrecchia, 1994; Healy, Hutton, and Palepu, 1999; and Welker, 1995); reduction in cost of capital (Botosan, 1997; and Botosan and Plumlee, 2002); or increased information intermediation (Bhushan, 1989a,b; Lang and Lundholm, 2000; Healy et al., 1999; Francis, Hanna and Philbrick, 1998). The major limitation to studying voluntary disclosure is the difficulty in measuring the extent of voluntary disclosure. Currently, researchers are using several proxies for voluntary disclosure. For example, Lang and Lundholm (1993, 2000) use metrics based on the AIMR database, and Botosan (1997) and Miller (2002) use self-constructed measures. Each approach has its limitations. The accuracy of management forecasts can be easily verified by outsider investors through actual realized earnings. This is not true for most other voluntary disclosure. Therefore, the research using management forecasts as the basis for voluntary disclosure cannot generalize to other forms of voluntary disclosure. The AIMR database is based on an annual survey which produces firm rankings of voluntary disclosure. It provides a more general measure of voluntary disclosure. However, the AIMR rankings rely heavily on analysts’ selection and judgment, which may bring bias to the data. Critiques of self-constructed measures argue that these metrics typically rely on annual reports and may ignore other sources of information such as press briefings, analyst meetings, etc. This study contributes to the literature by constructing a Transparency Index based on the OECD Principles of Corporate Governance (2004), a comprehensive and widely accepted Corporate Governance guideline. We further decompose the index into a Mandatory Disclosure Index and a Voluntary Disclosure Index. While most of the prior research focuses on financial disclosure, such as information in financial reports or information related to earnings, our index is unique in that it includes many Corporate Governance related disclosure items beyond the financial disclosure related items. Our rating also includes other channels of disclosure such as press briefings, analyst conferences, and company websites. In addition, we use firm fixed effect regressions in our analysis to reduce the possible endogeneity problem in the existing literature. The key research question of this study is whether there is a relation between disclosure practice and market valuation for Chinese listed companies. If a company discloses more information that facilitates the monitoring of management decisions by shareholders, then this should reduce monitoring costs and reduce the company’s cost of capital. If transparency is desired by investors, then companies will be given a premium for more disclosure on a voluntary basis. This should then be reflected in company valuation. One would thus expect company valuation to be positively associated with voluntary disclosure.

3. Institutional background Development of China’s financial system is essential to the success of its economic reform. In the early 1990s, China became the first transition country to establish a formal stock market exchange. The Shanghai Stock Exchange and the Shenzhen Stock Exchange were formally opened in December 1990 and in July 1991, respectively. There are various types of shares issued in China. These include


263

State Shares (owned by the state), legal person shares (owned by other institutions, mostly stateowned), individual shares (owned by individual Chinese investors, also known as A-shares), and domestically-listed foreign-held shares (available only to foreigners, known as B-shares).8 State-owned shares (state and legal person shares) account for two-thirds of all shares issued and are non-tradable.9 There are a few Chinese companies listed on overseas exchanges. Chinese companies have been permitted to list on the Stock Exchange of Hong Kong as H-shares since 1993. A few Chinese companies issue N-shares that are traded on US stock markets in the form of American Depository Receipts (ADRs). In 2006, there were 1427 companies listed on both stock exchanges, with total market capitalization of more than RMB 8.9 trillion (US$ 1.1 trillion). The development of China’s stock market has been characterized by excessive administrative control and intervention, lack of transparency, and an underdeveloped legal and regulatory framework. More recently, there have been numerous efforts launched by the Chinese State Regulatory Commission (CSRC) to develop and better regulate the stock market. We highlight four major reform initiatives in the following paragraphs. First, China’s first Securities Law was adopted in December 1998 and took effect in July 1999. This represents a major step towards a well regulated stock market. Among other provisions, the Law imposes greater disciplines on the participants, including listed companies, their directors and senior management, and financial intermediaries. Importantly, it imposes a tighter disclosure requirement on listed companies for better investor protection. Second, there were several major amendments to the 1985 Accounting Law, aimed at increasing the transparency of accounting information. The Law stipulates higher standards for accounting information released by listed companies. The amendments also strengthen disclosure requirements in other areas, such as board decisions related to dividends, explanations to declare no dividends, and more disclosure on large transactions. The Law also requires companies to publish annual reports on-line, in addition to the newspaper announcements that were formerly required. Third, the Guidelines for Corporate Governance of Listed Companies were issued by CSRC in December 2003. The objective is to enhance good Corporate Governance practices in China. The guidelines are based on the OECD Principles of Corporate Governance and the Chinese authority has taken steps to enhance its enforcement through special inspections. Finally, the Chinese government announced a pilot program to start the split share structure reform on April 29, 2005. The split share structure was the outcome of partial privatization of China’s companies implemented in the early 1990s, in which the Chinese government still holds the controlling shares of listed companies after they are listed. Split share structure refers to the two classes in the ownership structure of Chinese listed companies: domestic A-shares with identical cash flow and voting rights are divided into tradable and non-tradable shares. Non-tradable shares are typically held by the Chinese government and its affiliates, and account for about two-thirds of the total number of A-shares outstanding. The non-tradable shares cannot be traded in the stock market so they do not stand to gain from good firm performance. As a result, the controlling shareholders are not interested in firm value maximization, while tradable shareholders, due to their relatively small ownership stakes, have no or very limited influence on corporate decisions The Chinese government has recognized that the predominance of non-tradable shares in the stock market constitutes a major problem for its future development. With the split share structure reform, nontradable shareholders gain tradability by compensating existing tradable shareholders with bonus shares, cash, etc. The reform could change the ownership structure of the Chinese listed companies and their incentives to push companies towards a better Corporate Governance structure (including better transparency).

8

In 2001, B-shares were made available to domestic investors. The shares held by state or government agencies are called State Shares. Before the ownership structure reform of China in 2005, State Shares were non-tradable on any stock market as were legal person shares, which are held by domestic institutions or legal entities. 9

264


4. Data and methodology Our sample consists of the 2004–2007 Fortune 100 largest listed Chinese companies.10 The Fortune ranking is based on the total revenue of all Chinese firms listed around the world, including Mainland China, Hong Kong, New York, London, Singapore, and the NASDAQ. The centerpiece of this study is the construction of a Transparency Index, designed to measure the quality of disclosure practices of these large Chinese listed companies. The OECD Principles (OECD, 2004) give the best practice recommendations in five categories: rights of shareholders, equitable treatment of (minority) shareholders, the role of stakeholders, disclosure and transparency, and board responsibilities and composition. A set of criteria is developed to measure the quality of disclosure from these five Corporate Governance Principles. A total of 56 criteria (questions and sub-questions) form the scorecard used to assess each firm in our sample. To reduce the subjectivity of assessment, all criteria for every company are rated twice and cross checked by different raters. This study contributes to the existing disclosure literature by adding a quantitative dimension to the disclosure measures. Companies that omit or do not comply with a specific scoring criterion receive a ‘poor’ score (score = 1). Meeting the minimum compliance standard earns a firm a score of ‘fair’ (score = 2), while firms that exceed the minimum requirements and/or meet international standards receive a higher score (score = 3). This is a distinctive feature of this study, as most previous research has only checked for the presence of a specific disclosure measure. Our data sources include annual reports, articles of association, memorandums of association, notices to call shareholders’ meetings, annual general meeting minutes, company websites, analyst reports, and other sources available to the general public. We calculate the Transparency Index as the equally weighted score of all 56 criteria.11 Firms with a better quality of disclosure practices have higher scores. Among the 56 criteria, 24 (questions and sub-questions) are included in various regulations of the CSRC, stock exchanges, and other authorities.12 These criteria are considered as mandatory disclosure requirements. We classify the other criteria (32) that are not included in any regulations as voluntary disclosure criteria. This divides the Transparency Index into two sub-indexes – Voluntary Disclosure Index and Mandatory Disclosure Index. The list of criteria is attached in the Appendix. Included in our sample are dually or overseas-listed Chinese companies. As there is variation in the mandatory disclosure requirements of different jurisdictions, these companies are under different regulatory frameworks that require different disclosure requirements. We identify three groups of dually or overseas-listed companies in the sample. One group issues both A- and B-shares, a second group issues only A-shares and is dually listed in the US, and the third group issues H-shares in Hong Kong. In 2004, 9 (7 in 2005, 7 in 2006, and 3 in 2007) Chinese companies issued both A- and B-shares. Companies that issue both A- and B-shares are under the regulatory framework of the CSRC and there is no difference in disclosure requirement between the two markets.13 However, A-share companies 10 These include A-share and H-share companies in our sample. These companies are representative that the market capitalization of A-share companies of our sample account for 74% of the total market capitalization of A-share market during the four years period ranging from 64% in 2004 to 84% in 2007. The average market capitalization of H-share companies of our sample account for 63% of the total market capitalization of H-share market. 11 There is substantial variation in the weighting schemes used in the literature. For example, Mitton (2004) assigns a composite score to each emerging market firm based on management discipline, transparency, independence, accountability, responsibility, fairness, and social responsibility. The first six areas have a weight of 15 %, while social responsibility has a weight of 10 %. Gompers, Ishii and Metrick (2003) code each of 24 factors as either 1 (yes) or 0 (no). The sum of the points yields a Corporate Governance Index. Brown and Caylor (2006) also use equal weights for their factors. We try different weighting schemes and the results are qualitatively the same. 12 The regulations of the CSRC, stock exchanges, and other authorities which we are refering to include: Opinion on the Norm of Shareholder Meeting of Listed Companies, Securities Law of the People’s Republic of China, Notice on Strengthening financial disclosure of Listed Company, Shanghai Stock Exchange Stock Listing Rules (Revised in 2000), Guiding Opinion on Establishing Independent Director System in Listed Companies, Shenzhen Stock Exchange Stock Listing Rules, the Circular by CSRC on Content and Format Standards of Information Disclosure for Securities Issuing Companies No. 2 – Content and Format of Annual Report, Guidelines for Corporate Governance of Listed Companies. 13 According to the disclosure requirements issued by the CSRC ‘‘Listing of Foreign Investment Shares inside China by a Company Limited by Shares Provisions Implementing Rules” (www.eduzhai.net/yingyu/615/763/yingyu_246464.html) and ‘‘Regulations of the State Council of foreign capital stocks listed in China by Joint-stock Companies”, www.eduzhai.net/yingyu/615/763/ yingyu_246429.html. The A-share and B-share companies have the same disclosure requirement.


265

are required to prepare accounting information under the Chinese GAAP and audited by local CPA firms, while B-share companies under the IAS and audited by international accounting firms. A description of the different auditing practices between the A-share and B-share can be found in Sami and Zhou (2004). Chinese listed companies that were dually listed in China and the US as American Depository Receipts (ADRs) totaled 8 in 2004 (7 in 2005, 7 in 2006, and 8 in 2007). ADRs are certificates for shares in foreign companies that are held in the foreign branches of US banks. In practice, if a US entity was to purchase one thousand shares of a Chinese ‘‘H-share”; its broker in Hong Kong would buy the shares and deposit them into a Hong Kong branch of a US bank. The American buyer would then receive an ADR certificate indicating ownership. These ADRs are not actively traded on US exchanges and companies trading ADRs are not required to meet US disclosure requirements. For example, ‘‘Level I” ADRs are traded in the over-the-counter market. They are required to provide summaries of public reporting documents required by the firm’s domestic market. These companies are also under the disclosure requirements of the CSRC. There were 28 companies in 2004 (30 in 2005, 33 in 2006, and 33 in 2007) listed in Hong Kong as H-shares. These companies are subject to Hong Kong market’s listing rules which are different from those of China. The Code of Corporate Governance Practices (2004) of the Hong Kong market has two levels of disclosure requirements for listed companies.14 All Hong Kong listed companies are required to comply with the first level of disclosure. In their annual reports they must state whether they have complied with the Code Provision and provide explanations for any non-compliance. The second level has recommendations for voluntary disclosure. Using the Hong Kong market’s disclosure requirements, the 56 disclosure items of the Transparency Index are classified into two sub-groups: 32 criteria of mandatory disclosure and 24 criteria of the voluntary disclosure.15 The Transparency Index of H-share companies is assessed using the Hong Kong regulatory framework.16 Other data sources include accounting information and firm performance data, which are obtained from DataStream. All data are matched according to each sample firm’s fiscal year. Firm ownership, board composition, and other firm data are obtained from annual reports. To make sure that the results are not driven by firm heterogeneity, we add control variables that cover firm characteristics (return on assets, firm size, and debt-equity ratio), ownership structure (percent of shares held by the State, whether the top five shareholders are holding over 50% of the total shares), board characteristics (percentage of independent directors on the Board) and other Corporate Governance variables (CEO duality; presence of an audit committee, compensation committee, and nomination committee; whether the company is dually or overseas-listed). 5. Empirical analysis 5.1. Descriptive statistics Table 1 presents descriptive statistics of the Transparency Index and the two sub-indexes during the 4-year period. The results show that there was no substantial improvement in the corporate transparency of China listed companies from the year 2004 to 2007. The average Transparency Index is around 55.8. An interesting change was seen in the Voluntary Disclosure Index which increased from 20.92 in 2004 to 53.95 in 2007, while the Mandatory Disclosure Index showed some improvement from 62.43 to 67.03. This implies that Chinese listed companies are improving substantially in their voluntary disclosure, if we use 50 as an acceptable score. It should be noted that the average Mandatory Disclosure Index is around 66.16 with a standard deviation of 9.9. This reflects there exists some variations in mandatory disclosure requirements among Chinese listed companies. This could be explained by the scoring scheme. A score ranging from 1 to 3 is used to assess the level of disclosure with 14

The two levels of disclosures are the Code Provisions and Recommended Best Practices. There are more mandatory disclosure requirements under Hong Kong’s regulatory framework than that of China. For example Hong Kong listed companies are required to provide announcements in both Chinese and English but this is an option for listed companies in China (for example, question D.8 viii). 16 To test for robustness, the H-share companies are assessed by the CSRC requirements that apply for A-share companies. We have repeated our analysis and obtained similar results (available upon request). 15

266


Table 1 Summary Statistics of Transparency Index, Voluntary Disclosure Index, and Mandatory Disclosure Index. This table reports the summary statistics of the Transparency Index, Voluntary Disclosure Index and Mandatory Disclosure Index. Year

Index

Mean

25th Percentile

Median

75th Percentile

2004

Transparency Index Voluntary Disclosure Index Mandatory Disclosure Index

51.057 20.920 62.437

SD 5.843 6.726 9.116

47.561 16.563 56.466

51.220 20.000 61.979

56.098 25.625 69.271

2005


51.224 30.418 62.872

5.676 6.935 8.241

47.670 26.250 57.609

50.365 30.625 63.778

53.780 36.250 68.478

2006


60.709 41.192 69.187

9.400 13.884 6.775

53.623 30.208 64.817

60.009 38.021 70.390

68.388 52.951 74.632

2007


59.280 53.951 67.030

13.687 19.004 12.165

53.125 42.105 58.667

60.938 56.250 67.025

68.750 68.421 74.617

All


55.808 40.986 66.161

10.230 15.443 9.873

48.750 30.556 59.483

54.688 38.021 65.945

62.500 50.000 72.845

The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. The Mandatory Disclosure Index is based on mandatory disclosure related questions from the OECD Principles on Corporate Governance. We report the mean, standard deviation, median, as well as the 25th percentile and 75th percentile values by year.

a score of 2 assigned to indicate compliance with the mandatory requirement. A score of 3 is assigned to companies that disclose more than required. For example, question B.1 assesses the quality of the disclosure on related-party transactions of the companies in the annual reports. According to the disclosure requirements for the Chinese listed companies on the related-party transactions (article 46, CSRC, 2003), companies are required to disclose the information on the substantial related-party transactions during the reporting period. These include the identity of the related party, transaction content, pricing rule, transaction amount, the market share, the settlement details, and the impacts on the company. However, there are circumstances (articles IV and V, CSRC, 2003) upon the approval of the stock exchanges that companies may not compile the disclosure requirements. These conditions are either companies find these rules are not applicable (article IV) or for the reason of commercial secrets (article V). There are differences among companies in disclosing the details on connected party transactions. Some companies insert a table in the annual reports that gives the details of the individual connected party transaction. In 2004, among the 100 companies, there were seven companies that scored 1, 18 companies scored 3, and the rest scored 2.17 This explains the possibility of non-compliance of the mandatory disclosure requirements by Chinese listed companies and also the variation in the Mandatory Disclosure Index among Chinese listed companies. Tables 2a and 2b show the descriptive statistics of company characteristics. It can be seen from Panel A that the average market-to-book ratio (MTBV) is 2.36 for the 2004 sample and 3.90 for the 2007 sample. The average total assets are 132.7 billion RMB (16.1 billion US$), increasing from 58.7 billion RMB (7.2 billion US$) in 2004 to 182.1 billion RMB (22.2 billion US$) in 2007. The average return on assets (ROA) is 6.4%, improving from 5.1% in 2004 to 7.3% in 2007. A similar pattern is found in the return on equity (ROE). These statistics show that there is improvement in the financial performance of Chinese listed companies during the sample period. In Panel B, other characteristics of the companies in our sample are as follows: independent nonexecutive directors make up an average of one-third of the boards of director; the CEO is also the Chairman of the Board for 14% of our sample companies; the State controls an average of about 48% of the shares of the listed companies in our sample; and 45% of the sample companies are either 17 Wuliangye Yibin Corporation Limited (stock code: 000858) was among the seven companies that scored 1. They have just reported the amount, the related party, but they do not disclose the date and the pricing rule in the annual report.

267

Y.-L. Cheung et al. / J. Account. Public Policy 29 (2010) 259–280 Table 2a Summary statistics of firm characteristics. Year

Variable

Panel A: Firm characteristics 2004 Total asset Market value Total sales Tobin’s Q MTBV ROA ROE Leverage

N

Mean

SD

25th Percentile

Median

75th Percentile

96 94 96 93 94 96 96 95

58.727 38.833 24.445 1.179 2.360 0.051 0.116 0.958

131.112 106.953 54.723 0.560 1.566 0.047 0.083 1.488

7.286 5.076 7.305 0.798 1.500 0.020 0.080 0.199

10.712 9.190 11.145 1.082 2.110 0.050 0.118 0.482

24.269 18.298 16.381 1.496 2.850 0.080 0.155 0.956

2005

Total asset Market value Total sales Tobin’s Q MTBV ROA ROE Leverage

97 97 97 97 97 97 97 97

72.004 37.743 33.351 0.929 1.775 0.060 0.139 0.829

151.508 102.602 73.520 0.463 1.075 0.060 0.099 0.826

7.585 3.770 10.296 0.692 1.210 0.020 0.095 0.245

14.768 7.205 16.218 0.840 1.540 0.048 0.137 0.561

31.161 25.710 23.760 1.145 2.050 0.088 0.184 1.080

2006


97 97 97 97 97 97 97 97

216.755 101.925 44.802 1.972 3.896 0.071 0.143 0.831

794.595 270.349 129.880 1.681 2.850 0.053 0.158 1.683

7.414 11.820 4.368 0.924 1.916 0.034 0.095 0.195

19.662 22.081 12.007 1.377 3.055 0.053 0.140 0.444

61.084 53.522 37.489 2.442 4.798 0.101 0.206 0.721

2007


98 98 98 97 98 95 94 98

182.118 95.678 42.378 1.954 3.896 0.073 0.125 0.919

789.710 274.834 129.188 1.589 2.860 0.220 0.141 1.695

7.294 12.402 2.915 0.924 1.916 0.056 0.082 0.225

18.433 22.532 11.407 1.396 3.135 0.087 0.123 0.506

55.336 51.764 30.227 2.448 5.231 0.141 0.190 0.954

All


388 386 388 384 386 385 384 387

132.719 68.846 36.290 1.512 2.989 0.064 0.131 0.884

572.293 208.725 102.509 1.298 2.419 0.119 0.124 1.462

7.413 7.307 6.866 0.800 1.540 0.030 0.086 0.207

13.348 15.030 12.910 1.186 2.180 0.060 0.131 0.497

48.827 34.475 26.369 1.688 3.454 0.101 0.183 0.954

This table reports the summary statistics of financial characteristics of firms in the sample. We report the mean, standard deviation, median as well as the 25th percentile, and 75th percentile values by year. Total assets, market value and total sales are in billion RMB and measured at the fiscal year end. Tobin’s Q is defined as (short-term debt + long term debt + market value of equity)/total asset. MTBV is defined as market value of equity/total equity. ROA is return on assets and equals net income divided by total assets. ROE is return on equity and equals net income divided by total equity. ROS is return on sales and equals net income divided by total sales. Leverage is total debt divided by total equity.

B-share listed or listed overseas in exchanges such as Hong Kong and New York. The correlation table of main dependent variables is reported in Appendix 2. We do not spot any serious multicollinearity problem. 5.2. Is more disclosure associated with higher market valuation? This subsection addresses the relationship between the Transparency Index and company valuation. This study uses Tobin’s Q and market-to-book ratio (MTBV) as proxies for company valuation.

268


Table 2b Summary statistics of firm characteristics. Year

Variable

N

Panel B: Corporate Governance characteristics 2004 State Shares 97 Concentrate Dummy 97 Board Independence 97 Duality Dummy 97 Committee 97 Overseas Listing 97 2005

2006

2007

All

Mean

SD

25th Percentile

Median

75th Percentile

0.510 0.835 0.312 0.175 0.536 0.474

0.266

0.372

0.579

0.706

0.076

0.273

0.333

0.333

0.261

0.400

0.583

0.704

0.063

0.286

0.333

0.364

0.309

0.000

0.535

0.661

0.058

0.333

0.333

0.364

0.246

0.345

0.545

0.655

0.068

0.333

0.333

0.364

0.274

0.319

0.558

0.685

0.068

0.313

0.333

0.364

State Shares Concentrate Dummy Board Independence Duality Dummy Committee Overseas Listing

96 98 98 98 98 98

0.514 0.827 0.324 0.133 0.592 0.510


98 100 98 92 100 100

0.403 0.880 0.345 0.120 0.660 0.430


99 99 99 99 99 99

0.483 0.768 0.353 0.131 0.626 0.384


390 394 392 386 394 394

0.477 0.827 0.334 0.140 0.604 0.449

This table reports the summary statistics of Corporate Governance characteristics of firms in the sample. We report the mean, standard deviation, median, as well as the 25th percentile and 75th percentile values by year. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise.

Tobin’s Q is defined as (short-term debt + long-term debt + market value of equity)/total asset and MTBV as market value of equity/total equity. Endogeneity is always a concern for studies dealing with the relationship between company valuation and the Transparency Index. For example, a firm that has more information disclosure is more likely to make a high profit; it may be that it is the high profit that investors value rather than the amount of information disclosure. To avoid misspecification of the equation used to explain how investors value information disclosure, we tackle this problem using two approaches. First, we include a comprehensive set of control variables to mitigate the omitted-variable bias and the possibility that our results are affected by an endogeneity problem. Second, we apply the panel regression model with fixed effect to minimize the endogeneity problem. Our control variables cover performance (asset size and return on asset); risk factors (leverage); ownership structure (state ownership and the shareholdings of the top five shareholders); board characteristics (the proportion of independent non-executive directors; the presence of an audit committee, compensation committee, and nomination committee; and CEO duality); other variables

269


(Overseas Listing and listing time). The set of control variables is listed in Panel B of Table 2b. The regression model is given by:

Tobin’s Q ðor MTBVÞ ¼ a þ b1 TI þ b2 Firm Size þ b3 Leverage þ b4 ROA þ b5 State Shares þ b6 Concentrate Dummy þ b7 Board Independence þ b8 Duality Dummy þ b9 Committee þ b10 Oversea Listing þ e

ð1Þ

Table 3 reports the results of the regression model using Tobin’s Q and MTBV as the dependent variables. A positive relation is found between Tobin’s Q and the Transparency Index in Model 1. When we progressively add control variables into the regression model, including company characteristics and Corporate Governance variables, we also find a positive and significant relation between the Transparency Index and Tobin’s Q in Model 2. Interestingly, the result seems to be stronger for companies with less state ownership and more independent directors. The control variables of company

Table 3 Regression results of firm value on the Transparency Index. Dependent variable

(1) Tobin’s Q

(2) Tobin’s Q

(3) MTBV

(4) MTBV

Transparency Index

0.009*** [0.001]

0.005* [0.070] 0.001 [0.502] –0.011 [0.589] 0.025*** [0.001] 0.304** [0.017] 0.019 [0.880] 1.218** [0.018] 0.082 [0.546] 0.058 [0.397] 0.125 [0.211] 0.669*** [0.007] 370 0.236

0.022*** [0.007]

0.016* [0.067] 0.001 [0.539] 0.110* [0.095] 0.077*** [0.002] 0.082 [0.843] 0.076 [0.848] 2.231 [0.183] 0.399 [0.375] 0.377* [0.088] 0.107 [0.743] 1.878** [0.021] 371 0.145

Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Board Independence Duality Dummy Committee Overseas Listing Constant Observations R-squared

0.587*** [0.000] 384 0.060

1.320*** [0.003] 386 0.041

This table reports the firm fixed effects for panel data regression results of firm value on the Transparency Index and control variables. The dependent variable is industry adjusted Tobin’s Q and MTBV. Tobin’s Q is defined as (short-term debt + long-term debt + market value of equity)/total asset. MTBV is defined as market value of equity/total equity. The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent nonexecutive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise. p-Values are reported in parentheses. * Indicate significance at 10% levels. ** indicate significance at 5% levels. *** indicate significance at 1% levels.

270


Table 4 Regression results of firm value on Voluntary and Mandatory Disclosure Index. Dependent variable

(1) Tobin’s Q

(2) Tobin’s Q

(3) MTBV

(4) MTBV

Voluntary Disclosure Index

0.006*** [0.009] 0.002 [0.483]

0.004* [0.081] 0.003 [0.346] 0.001 [0.824] 0.008 [0.790] 0.037*** [0.001] 0.390** [0.022] 0.184 [0.298] 0.258 [0.722] 0.068 [0.732] 0.012 [0.900] 0.072 [0.627] 0.334 [0.373] 370 0.138

0.020*** [0.000] 0.006 [0.479]

0.017*** [0.004] 0.005 [0.523] 0.001 [0.892] 0.115* [0.091] 0.091*** [0.000] 0.351 [0.371] 0.037 [0.928] 0.515 [0.760] 0.359 [0.438] 0.250 [0.268] 0.042 [0.903] 0.874 [0.316] 371 0.159

Mandatory Disclosure Index Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Board Independence Duality Dummy Committee Overseas Listing Constant Observations R-squared

0.421* [0.073] 384 0.037

0.820 [0.130] 386 0.069

This table reports the firm fixed effects for panel data regression results of firm value on both the Voluntary Disclosure Index and the Mandatory Disclosure Index and control variables. The dependent variable is industry adjusted Tobin’s Q and MTBV. Tobin’s Q is defined as (short-term debt + long-term debt + market value of equity)/total asset. MTBV is defined as market value of equity/total equity. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. The Mandatory Disclosure Index is based on mandatory disclosure related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise. p-Values are reported in parentheses. * Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels.

characteristics are adjusted to control for variation among different industries. Similar results are found for Models 3 and 4 using MTBV as the dependent variable. The results also show that more profitable companies tend to exhibit a higher degree of transparency in both regression models. The findings support the hypothesis that company valuation is positively and significantly associated with more disclosure, as measured by the Transparency Index. 5.3. Mandatory Disclosure Index and Voluntary Disclosure Indexes We further investigate which kinds of disclosure affect market valuation. We replace the Transparency Index in our regressions with the two sub-indexes: the Voluntary Disclosure Index (VDI) and the

271

Y.-L. Cheung et al. / J. Account. Public Policy 29 (2010) 259–280 Table 5 Robustness test with subsample. Dependent variable

(1) Tobin’s Q

Panel A: Subsample of companies listed in China stock market Transparency Index 0.005 [0.159] Voluntary Disclosure Index Mandatory Disclosure Index Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Board Independence Duality Dummy Committee Overseas Listing Constant Observations R-squared

0.001** [0.047] 0.001 [0.981] 0.031*** [0.001] 0.278** [0.049] 0.025 [0.857] 1.133* [0.095] 0.079 [0.627] 0.062 [0.397] 0.160 [0.170] 0.622** [0.050] 300 0.285

(2) Tobin’s Q

(3) MTBV

(4) MTBV

0.018* [0.080] 0.004* [0.089] 0.001 [0.647] 0.001 [0.150] 0.005 [0.803] 0.029*** [0.001] 0.322** [0.018] 0.015 [0.916] 1.090 [0.107] 0.062 [0.705] 0.055 [0.447] 0.101 [0.400] 0.435 [0.210] 300 0.285

Panel B: Subsample of companies non-listed in China stock market Transparency Index 0.021*** [0.003] Voluntary Disclosure Index 0.011* [0.069] Mandatory Disclosure Index 0.010 [0.137] ** 0.001** Firm Size-ind adjusted 0.001 [0.015] [0.015] 0.296*** Leverage-ind adjusted 0.292*** [0.002] [0.002] 0.092*** ROA-ind adjusted 0.091*** [0.000] [0.000] State Shares 0.025 0.026 [0.932] [0.930] Concentrate Dummy 0.049 0.061 [0.834] [0.803] Board Independence 0.878 0.912 [0.152] [0.164] Duality Dummy 0.052 0.061 [0.676] [0.629] Committee 0.014 0.025 [0.908] [0.839] 0.842* Constant 0.912** [0.029] [0.056] Observations 70 70 R-squared 0.449 0.501

0.002 [0.456] 0.102 [0.147] 0.078*** [0.007] 0.114 [0.803] 0.148 [0.751] 2.862 [0.198] 0.222 [0.683] 0.428* [0.078] 0.154 [0.689] 2.271** [0.027] 301 0.166

0.021*** [0.003] 0.001 [0.882] 0.001 [0.929] 0.098 [0.156] 0.077*** [0.007] 0.119 [0.788] 0.115 [0.800] 1.960 [0.377] 0.082 [0.878] 0.361 [0.133] 0.118 [0.766] 1.626 [0.154] 301 0.205

0.018 [0.348]

0.001 [0.373] 0.148 [0.400] 0.087* [0.052] 0.820 [0.222] 1.034 [0.368] 1.407 [0.218] 0.039 [0.900] 0.183 [0.364] 1.811 [0.256] 70 0.400

0.020 [0.204] 0.012 [0.472] 0.001 [0.404] 0.125 [0.450] 0.088** [0.043] 0.851 [0.199] 0.837 [0.469] 1.024 [0.403] 0.070 [0.827] 0.229 [0.274] 0.934 [0.564] 70 0.431

272


Table 5 (continued) This table reports the firm fixed effects regression results of Tobin’s Q on the Transparency Index, the Voluntary Disclosure Index and the Mandatory Disclosure Index and control variables with subsample. The dependent variable is industry adjusted Tobin’s Q and MTBV. Tobin’s Q is defined as (short-term debt + long-term debt + market value of equity)/total asset. MTBV is defined as market value of equity/total equity. The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. The Mandatory Disclosure Index is based on mandatory disclosure related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise. p-Values are reported in parentheses. Panel A presents the regression result with the companies which should list in China stock market. Panel B presents the regression result with the companies not listed in China stock market. * Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels.

Mandatory Disclosure Index (MDI). The results are reported in Table 4. The Voluntary Disclosure Index is positive and significantly related to the Tobin’s Q and MTBV in all regression models. The Mandatory Disclosure Index, however, is not found to be related to the company valuations in any of the models. Our findings show that investors are more responsive to higher voluntary disclosure than mandatory disclosure. The result is stronger among more profitable companies.

5.4. Robustness test To check the robustness of our findings, we address three questions. The first question relates to the companies that are only listed in the overseas market. We notice that there are 45% of the sample companies that are listed in overseas markets. These companies are either dually listed or only listed in the overseas markets. During the 4-year period, there are 85 companies that are only listed in Hong Kong as H-share. The rest of the sample is listed in China as A-share. These H-share companies are under a more stringent regulatory framework, compared with the A-share market. Therefore, we need to address whether our results are driven by the H-share companies. To check the robustness of our findings, we separate our sample into two sub-groups. The first sub-group includes Chinese companies that are listed in China and the second contains those that are listed in Hong Kong. Table 5 shows the regression results of the two sub-samples. Panel A include companies that are listed in China as A-shares. The result shows that Transparency Index is significantly positive related to both Tobin’s Q (marginally at 16%) and MTBV. When we split Transparency Index into Voluntary Disclosure and Mandatory Disclosure Indexes, we find both company valuations are positive and significant related to the VDI, but not with the MDI. The result for H-share Company is presented in Panel B that shows Tobin’s Q is positive and significant related to Transparency Index and VDI and the result is weaker for MTBV. This provides evidence to support that our findings are applicable for A-share companies and not driven by H-share companies. The second question addresses the reverse causality between the Transparency Index and the company valuation. One major problem troubling all Corporate Governance studies is the potential for reverse causality problem. The instrumental variable approach is widely used as a means to resolve the problem. However, it is difficult to find an instrument highly correlated with the variable of interest but uncorrelated with the error term of the true structural model. Nevertheless, we acknowledge the problem and make four attempts to identify the instrument to address the problem. These include Overseas Listings, Overseas Listings and Board Independence, Corporate Governance variables, and all control variables as instruments.

273


In order to be a valid instrumental variable, the instrumental dummy needs to satisfy two conditions in the regression model. First, the covariance between instrument dummy and the residual from regression model should equal 0. The second condition is that the covariance between the instrumental dummy and the variable of interest should not equal 0. With the valid instrumental variable, we use a two-stage least squares model to check whether the results are induced by endogeneity problem. The Sargan statistic is to ensure the validity of the instrumental variable. The results of the two-stage least squares are presented in Tables 6a–6d. Panel A shows the result using Overseas Listing as the instrument variable that Transparency Index is positive and significant related to both company valuations. The Sargan statistic is insignificant that indicates the validity of using Overseas Listing as the instrumental variable. Panel B shows the result of using Oversea Listing and Board Independence as instrumental variables that both company valuations are positive and

Table 6a Robustness test for the issue of reverse causality. Panel A: 2SLS regression with the variable Overseas Listing as instrumental variables Dependent variable (1) Tobin’s Q Transparency Index Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Board Independence Duality Dummy Committee Constant Observations

0.197* [0.055] 0.001 [0.471] 0.084 [0.356] 1.488 [0.380] 0.591 [0.298] 0.446 [0.219] 2.351 [0.361] 0.540 [0.264] 0.456 [0.284] 10.077** [0.044] 370

(2) MTBV 0.376** [0.043] 0.003 [0.198] 0.139 [0.406] 0.496 [0.872] 1.499 [0.139] 0.151 [0.820] 5.206 [0.265] 1.444 [0.103] 0.916 [0.242] 19.278** [0.033] 371

This table reports the firm instrumental variables (2SLS) regression results of firm value on the Transparency Index, the Voluntary Disclosure Index, and the Mandatory Disclosure Index and control variables. The dependent variable is industry adjusted Tobin’s Q and MTBV. Tobin’s Q is defined as (short-term debt + long-term debt + market value of equity)/total asset. MTBV is defined as market value of equity/total equity. The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. The Mandatory Disclosure Index is based on mandatory disclosure related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise. In this table, we use the variable Overseas Listing as instrumental variables to the Transparency Index. p-Values are reported in parentheses. * Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels.

274


Table 6b Robustness test for the issue of reverse causality. In this table, we use the variables Overseas Listing and Board Independence as instrumental variables to Voluntary Disclosure Index and Mandatory Disclosure Index. Panel B: 2SLS regression with the variables Overseas Listing and Board Independence as instrumental variables Dependent variable (1) (2) Tobin’s Q MTBV 0.074*** [0.004] 0.010 [0.842] 0.000 [0.699] 0.120* [0.064] 0.565 [0.584] 0.092 [0.801] 0.425 [0.123] 0.343 [0.240] 0.149 [0.543] 2.316 [0.508] 370

Voluntary Disclosure Index Mandatory Disclosure Index Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Duality Dummy Committee Constant Observations

0.136*** [0.002] 0.048 [0.569] 0.002 [0.198] 0.056 [0.614] 1.617 [0.359] 0.628 [0.305] 0.232 [0.625] 1.015** [0.042] 0.227 [0.590] 2.532 [0.675] 371

p-Values are reported in parentheses. Indicate significance at 10% levels. Indicate significance at 5% levels. *** Indicate significance at 1% levels. *

**

Table 6c Robustness test for the issue of reverse causality. In this table, we use the control variable State Share, Concentrate Dummy, Board Independence, and Duality Dummy as instrumental variables. Panel C: 2SLS regression with part of control variables as instrumental variables Dependent variable (1) (2) Tobin’s Q Tobin’s Q Transparency Index

0.095*** [0.001]

Voluntary Disclosure Index Mandatory Disclosure Index Firm Size-ind adjusted Leverage-ind adjusted ROA-ind adjusted Constant Sargan statistic Observations p-Values are reported in parentheses. Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels. *

0.034 [0.248] 0.122** [0.040] 0.013 [0.986] 5.698*** [0.001] 6.629 370

(3) MTBV

(4) MTBV

0.139*** [0.004] 0.058*** [0.001] 0.022 [0.579] 0.025 [0.441] 0.128** [0.039] 0.317 [0.699] 4.260* [0.100] 4.449 370

0.040 [0.414] 0.082 [0.404] 3.073** [0.017] 8.134*** [0.002] 5.648 371

0.102*** [0.001] 0.030 [0.649] 0.004 [0.939] 0.044 [0.667] 2.701** [0.045] 2.564 [0.545] 2.013 371

275

Y.-L. Cheung et al. / J. Account. Public Policy 29 (2010) 259–280 Table 6d Robustness test for the issue of reverse causality. In this table, we use all control variables as instrumental variables. Panel D: 2SLS regression with all control variables as instrumental variables Dependent variable (1) (2) Tobin’s Q Tobin’s Q Transparency Index

0.112*** [0.000]

Voluntary Disclosure Index Mandatory Disclosure Index Constant Sargan statistic Observations

6.573*** [0.000] 10.660 370

(3) MTBV

(4) MTBV

0.174*** [0.000] 0.063*** [0.000] 0.031 [0.391] 4.965** [0.029] 9.277 370

10.119*** [0.000] 10.616 371

0.123*** [0.000] 0.014 [0.824] 4.500 [0.240] 5.510 371

p-Values are reported in parentheses. Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels. *

significant related to VDI but not to MDI.18 The Sargan statistic is also not significant. The results of using Corporate Governance and all control variables as instrument are shown in Panels C and D that report similar findings. This supports the original regression results do not suffer from endogeneity. The third question addresses the construction of the Voluntary Disclosure Index. Should mandatory scores above the required level be treated as voluntary scores? How does disclosure above the mandatory level affect the findings? One could argue that mandatory scores above the required level should be treated as voluntary scores. When the manager has discretion in disclosing above the normal mandatory requirement, this overage could be counted as a voluntary score because the extra mandatory disclosure is a form of voluntary disclosure. We reassess the sample companies by putting a cap score of 2 for all mandatory criteria that meet the regulatory requirements and scores above 2 are added to the Voluntary Index. Under the new scoring method, the maximum Mandatory Disclosure Index is restricted to 50. We repeat the regression analysis in Table 5 using the new mandatory disclosure and Voluntary Disclosure Indexes. We expected to find a stronger result for the VDI because the new scoring method gives a bonus to companies disclosing above the mandatory level. The additional result also shows positive and significant associations between Tobin’s Q and the new VDI in all four regression models (results not reported here). The relationships between the MDI and Tobin’s Q are not significant.

5.5. Which kind of firms disclose more? We explore which company characteristics are associated with transparency. We focus on the overall transparency and voluntary disclosure of companies. First, we examine the companies with higher Transparency Index scores. Second, we examine the companies with higher Voluntary Disclosure Indexes. We perform a logit regression of the transparency dummy on various control variables. The transparency dummy (TI_Dummy) is equal to 1 if the transparency of the company is among the top 25% of the sample and zero otherwise.19 The results are shown in Table 7. We find that more profitable, overseas-listed, and companies with sub-committee tend to more transparent as measured the Transparency Index. 18 A group of instrument variables has been used by Bushman, Chen, Engel, and Smith (2004) to investigate how ownership concentration, directors’ and executive’s incentives, and board structure affect with earning management activities. 19 The cutoff point for the transparency dummy is 25%. We also ran the regression using cutoff points of 10% and 20%. The results are consistent with those using 25% as the cutoff point.

276


Table 7 Logit regression of Transparency Index and Voluntary Disclosure Index. This table reports the logit regression results of the transparency dummy and voluntary disclosure dummy on control variables. The dependent variable is TI_Dummy and VDI_Dummy. TI_Dummy is equal to 1 if the Transparency Index of the firm is among the top 25% of the sample and zero otherwise. VDI_Dummy is equal to 1 if the Voluntary Disclosure Index of the firm is among the top 25% of the sample and zero otherwise. The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise. Dependent variable

(1) TI_Dummy

(2) VDI_Dummy

Firm Size-ind adjusted

0.001 [0.188] 0.023 [0.765] 0.027*** [0.001] 0.026 [0.964] 0.190 [0.615] 0.119 [0.949] 0.401 [0.244] 0.536* [0.065] 0.738*** [0.009] 1.820*** [0.010] 373 0.067

0.001 [0.571] 0.118* [0.068] 0.027*** [0.002] 0.249 [0.689] 0.006 [0.989] 2.992 [0.144] 1.272*** [0.000] 0.308 [0.284] 1.469*** [0.000] 1.261* [0.097] 373 0.150

Leverage-ind adjusted ROA-ind adjusted State Shares Concentrate Dummy Board Independence Duality Dummy Committee Overseas Listing Constant Observations Pseudo R2 p-Values are reported in parentheses. Indicate significance at 10% levels. Indicate significance at 5% levels. *** Indicate significance at 1% levels. *

**

We further examine companies that disclose more on a voluntary basis. We perform a logit regression of the voluntary disclosure dummy (VDI_Dummy) on various control variables. The voluntary disclosure dummy is equal to 1 if the Voluntary Disclosure Index of the company is among the top 25% of the sample, and zero otherwise. Similar to the result for Transparency Index that we find that more profitable and overseas-listed companies tend to disclose more on a voluntary basis. In addition, companies, in which the CEO is also the Chairman of the Board, tend to be less transparent on a voluntary basis. 6. Concluding remarks Based on the OECD Principles of Corporate Governance, this study designs a comprehensive measure to assess the transparency of Chinese listed companies. The scorecard is composed of 56 criteria that are classified into five categories: rights of shareholders, equitable treatment of (minority) shareholders, the role of stakeholders, disclosure and transparency, and board responsibilities and composition. Each company is assessed each year. The 4-year time data is used in the fixed effect for panel data regression model to examine the relationship between company transparency and market

277


Table A1 Transparency and Disclosure Criteria include 56 questions and sub-questions. The table presents the mandatory requirements for Hong Kong and China. There are 24 and 32 mandatory disclosure items for China and Hong Kong respectively ( defines the questions related to mandatory disclosure). No.

Mandatory disclosure in China

Survey question

HK

Section A. Rights of shareholders A.1 Is the decision on the remuneration of board members or executives approved by the shareholders annually? A.2 How is the remuneration of the board presented? A.3 Quality of Notice to call a Shareholders Meeting in the past year (i) Appointment of directors, providing their names and background (ii) Appointment of auditors, providing their names and fees (iii) Dividend policy, providing the amount and explanation A.4.1 (i) Did the CEO/Managing Director attend at least one AGM in the past 2 years? A.4.2 (ii) Is a name list of board attendance available? A.5 Do AGM minutes record that there was an opportunity for shareholders to ask questions/raise issues in the past year? (i) Is there a record of answers and questions? (ii) Is any resolution being solved? Section B. Equitable treatment of shareholders B.1 Does the company provide rationales/explanations for related-party transactions affecting the corporation? B.2 Have there been any non-compliance cases regarding related-party transactions in the past year? B.3 Does the company facilitate voting by proxy? B.4.1 Does the notice to shareholders specify the documents required to give proxy? B.4.2 Is there any requirement for a proxy appointment to be notarized? B.5 How many days in advance does the company send out the notice of general shareholder meetings? Section C. The role of stakeholders in Corporate Governance C.1 Does the company explicitly mention the safety and welfare of its employees? C.2 Does the company explicitly mention the role of key stakeholders such as customers or the community at large (or creditors or suppliers)? C.3 Does the company explicitly mention environmental issues in its public communications? Section D. Disclosure and Transparency D.1 Does the company have a transparent ownership structure? (i) Breakdown of shareholdings (ii) Is it easy to identify beneficial ownership? (iii) Are director shareholdings disclosed? (iv) Is management shareholding disclosed? D.2 Assess the quality of the annual report. In particular, the following: (i) Financial performance (ii) Business operations and competitive position (iii) Board member background (iv) Basis of the board remuneration (v) Operating risks D.3 Is there any statement requesting the directors to report their transactions of company stock? D.4 Does the company have an internal audit operation established as a separate unit in the company? D.5 Does the company perform an annual audit using independent and reputable auditors? D.6 Are there any accounting qualifications in the audited financial statements apart from the qualification of Uncertainty of Situation? D.7 Does the company offer multiple channels of access to information? (i) Annual report (ii) Company website (iii) Analyst briefing (iv) Press conference/press briefing D.8 Does the company have a website, disclosing up-to-date information? (i) Business operation (ii) Financial statement (iii) Press release (continued on next page)

278


Table A1 (continued) No.

Mandatory disclosure in China

Survey question

HK (iv) Shareholding structure (v) Organizational structure (vi) Corporate group structure (vii) Annual report downloadable (viii) Provided in both Chinese and English

Section E. Responsibilities of the board E.1.1 Does the company have its own written Corporate Governance rules? E.1.2 Does the board of directors provide a code of ethics or statement of business conduct for all directors and employees? E.1.3 Does the company have a corporate vision/mission? E.2 Assess the quality and content of the Audit Committee Report in the annual report. (i) Attendance (ii) Internal control (iii) Management control (iv) Proposed auditors (v) Financial report review (vi) Legal compliance (vii) Conclusions or opinions E.3 Does the company state in its annual report the definition of ‘independence’? E.4 Does the company provide contact details for a specific investor relations person? E.5 Does the company have a board of director’s report?

valuation. The results reveal that there is a positive and significant relation between company transparency, as measured by the Transparency Index, and market valuation. When we further split the Transparency Index into Mandatory and Voluntary Disclosure Indexes, according to regulatory requirements, it is found that market valuation is only related to the Voluntary Disclosure Index, but not to the Mandatory Disclosure Index. This implies that investors desire transparency in Chinese listed companies and reward companies for more voluntary disclosure. However, the insignificant relation for the Mandatory Disclosure Index does not imply that mandatory disclosure requirements are not important. Companies are expected to comply with the mandatory disclosure requirements and there are consequences for non-compliance. The difference is that companies are not further rewarded for complying with the mandatory requirements. The final part of our study attempts to identify which companies tend to be more transparent and disclose more information on a voluntary basis. We find that profitable, overseas-listed, and companies with sub-committee tend to more transparent as measured the Transparency Index. In addition, we also find that more profitable, overseas-listed, and companies with a separate CEO and board chairman tend to disclose more on a voluntary basis. This study has three policy implications. First, we benchmark the disclosure practices of Chinese listed companies against an international standard of Corporate Governance. This has important implications for company management wanting to raise funds in the international capital market. Second, this study assesses the impact of disclosure practices on company valuation. The results are useful for management in designing their company disclosure policy. Finally, our findings can assist Chinese policy makers in formulating disclosure requirements for listed companies. Appendix 1 See Table A1. Appendix 2 See Table A2.

1 1. 2.

Transparency Index Voluntary Disclosure Index

3.

Mandatory Disclosure Index

4.

Firm Size-ind adjusted

5.

Leverage-ind adjusted

6.

ROA-ind adjusted

7.

State Shares

8.

Concentrate Dummy

9.

Board Independence

10.

Duality Dummy

11.

Committee

12.

Overseas Listing

1.000 0.825*** (0.000) 0.731*** (0.000) 0.092* (0.071) 0.082 (0.106) 0.119** (0.020) 0.023 (0.645) 0.052 (0.306) 0.119** (0.018) 0.107** (0.035) 0.219*** (0.000) 0.187*** (0.000)

2

3

4

5

6

7

8

9

10

11

12

1.000 0.290*** (0.000) 0.112** (0.028) 0.063 (0.215) 0.147*** (0.004) 0.022 (0.672) 0.046 (0.361) 0.080 (0.114) 0.144*** (0.005) 0.210*** (0.000) 0.269*** (0.000)

1.000 0.014 (0.789) 0.085* (0.097) 0.077 (0.134) 0.042 (0.409) 0.098* (0.052) 0.152*** (0.003) 0.012 (0.814) 0.112** (0.026) 0.073 (0.149)

1.000 0.012 (0.820) 0.050 (0.328) 0.061 (0.230) 0.003 (0.947) 0.046 (0.367) 0.010 (0.848) 0.104** (0.041) 0.162*** (0.001)

1.000 0.053 (0.302) 0.078 (0.128) 0.185*** (0.000) 0.142*** (0.005) 0.065 (0.210) 0.085* (0.095) 0.024 (0.639)

1.000 0.020 (0.701) 0.014 (0.790) 0.035 (0.494) 0.040 (0.441) 0.016 (0.761) 0.028 (0.581)

1.000 0.449*** (0.000) 0.011 (0.829) 0.029 (0.566) 0.043 (0.395) 0.097* (0.056)

1.000 0.028 (0.582) 0.044 (0.385) 0.068 (0.180) 0.088* (0.080)

1.000 0.027 (0.600) 0.118** (0.020) 0.028 (0.580)

1.000 0.013 (0.806) 0.091* (0.073)


Table A2 Correlation matrix. This table reports the correlation matrix of dependent variables. The Transparency Index is based on transparency related questions from the OECD Principles on Corporate Governance. The Voluntary Disclosure Index is based on voluntary disclosure related questions from the OECD Principles on Corporate Governance. The Mandatory Disclosure Index is based on mandatory disclosure related questions from the OECD Principles on Corporate Governance. Firm Size is the industry adjusted market capitalization of the firm. Leverage is calculated as total debt divided by total equity adjusted by the corresponding industry’s ratio. ROA is the ratio of net income to total assets of the firm adjusted by the corresponding industry’s ratio. State Shares is the fraction of shares owned by the state divided by total shares of the firm. Concentrate Dummy equals 1 if the top five shareholders hold more than 50% of the total shares of the firm. Board Independence is the percentage of independent non-executive directors on the board. Duality Dummy is dummy variable, equals 1 if the CEO is also the Chairman of the Board, and 0 otherwise. Committee is dummy variable, equals 1 if the firm has at least one of the following three committees: audit committee, nomination committee and remuneration committee, and 0 otherwise. Overseas Listing is dummy variable, equals 1 if the firm issued in B-shares market, or is cross-listed in the Hong Kong Stock Exchange, the US stock exchanges via ADR, the London Stock Exchange, or on the Singapore Stock Exchange, and 0 otherwise.

1.000 0.251*** (0.000)

1.000

p-Values are reported in parentheses. Indicate significance at 10% levels. ** Indicate significance at 5% levels. *** Indicate significance at 1% levels. *

279

280


References Aboody, D., Kasznik, R., 2000. CEO stock options awards and the timing of corporate voluntary disclosures. Journal of Accounting and Economics 29, 73–100. Ahmed, K., Courtis, J., 1999. Associations between corporate characteristics and disclosure levels in annual reports: a metaanalysis. British Accounting Review 31, 35–61. Allen, F., Qian, J., Qian, M., 2005. Law, finance, and economic growth in China. Journal of Financial Economics 77, 57–116. Beaver, W., 1998. Financial Reporting: an Accounting Revolution. Prentice-Hall, Englewood Cliffs, NJ. Bhushan, R., 1989a. Collection of information about publicly traded firms: theory and evidence. Journal of Accounting and Economics 11, 183–207. Bhushan, R., 1989b. Firm characteristics and analyst following. Journal of Accounting and Economics 11 (2–3), 255–275. Botosan, C.A., 1997. Disclosure level and the cost of equity capital. The Accounting Review 72 (3), 323–350. Botosan, C.A., Plumlee, M.A., 2002. A re-examination of disclosure level and expected cost of capital. Journal of Accounting Research 40 (1), 21–40. Brown, L., Caylor, M., 2006. Corporate governance and firm operating performance. Journal of Accounting and Public Policy 25, 409–434. Bushman, R., Chen, Q., Engel, E., Smith, A., 2004. Financial accounting information, organizational complexity and corporate governance systems. Journal of Accounting and Economics 37, 167–201. Cheung, Y.L., Jiang, P., Limpaphayom, P., Lu, T., 2008. Does corporate governance matter in China? China Economic Review 19, 460–479. China Securities Regulatory Commission, 2003. Circular on the Revision of the Content and Format Guidelines for Public Offering Companies’ Information Disclosure for No. 2 Annual Report Content and Format. Code of Corporate Governance Practices, 2004. Listing Rules (Appendix 14). Hong Kong Exchange. Diamond, D., Verrecchia, R., 1991. Disclosure, liquidity, and the cost of capital. The Journal of Finance 66, 1325–1355. Francis, J., Hanna, J., Philbrick, D., 1998. Management communications with securities analysts. Journal of Accounting and Economics 24, 363–394. Gompers, P., Ishii, J., Metrick, A., 2003. Corporate governance and equity prices. Quarterly Journal of Economics 118, 107–155. Healy, P., Palepu, K., 1993. The effect of firms’ financial disclosure strategies on stock prices. Accounting Horizons 7, 1–11. Healy, P., Palepu, K., 1995. The challenges of investor communications: the case of CUC International, Inc. Journal of Financial Economics 38, 111–141. Healy, P., Palepu, K., 2001. Information asymmetry, corporate disclosure, and the capital markets: a review of the empirical disclosure literature. Journal of Accounting and Economics 31, 405–440. Healy, P., Hutton, A., Palepu, K., 1999. Stock performance and intermediation changes surrounding sustained increases in disclosure. Contemporary Accounting Research 16, 485–520. Holthausen, R., 1981. Evidence on the effect of bond covenants and management compensation contracts on the choice of accounting techniques: the case of the depreciation switch-back. Journal of Accounting and Economics 3 (1), 73–109. Kim, O., Verrecchia, R., 1994. Market liquidity and volume around earnings announcements. Journal of Accounting and Economics 17, 41–68. Lang, M., Lundholm, R., 1993. Cross-sectional determinants of analysts ratings of corporate disclosures. Journal of Accounting Research 31, 246–271. Lang, M., Lundholm, R., 2000. Voluntary disclosure and equity offerings: reducing information asymmetry or hyping the stock? Contemporary Accounting Research 17 (4), 623–662. Leftwich, R., 1980. Market failure fallacies and accounting information. Journal of Accounting and Economics 2, 193–211. Miller, G., 2002. Earnings performance and discretionary disclosure. Journal of Accounting Research 40 (1), 173–204. Mitton, T., 2004. Corporate governance and dividend policy in emerging markets. Emerging Markets Review 5, 409–426. Organization for Economic Cooperation and Development, 2004. OECD Principles of Corporate Governance. OECD Publications, France. Sami, H., Zhou, H., 2004. Information in different segments of the Chinese stock market. The International Journal of Accounting 39, 403–427. Skinner, D., 1994. Why firms voluntarily disclose bad news. Journal of Accounting Research 32, 38–61. Skinner, D., 1997. Earnings disclosures and stockholder lawsuits. Journal of Accounting and Economics 23, 249–283. Watts, R., Zimmerman, J., 1986. Positive Accounting Theory. Prentice-Hall, Englewood Cliffs, NJ. Welker, M., 1995. Disclosure policy, information asymmetry and liquidity in equity markets. Contemporary Accounting Research 11, 801–828.