Ownership structure and voluntary disclosure: A synthesis of empirical studies

Abstract

In this article, we meta-analyse 69 empirical studies assessing the association between corporate voluntary disclosure and ownership concentration and types, and how institutional characteristics and research design moderate these relationships. Our overall analyses show that state, foreign and institutional ownerships have a positive effect but managerial ownership and ownership concentration have a negative effect on voluntary disclosure. Since the overall effect may conceal the underlying factors that cause heterogeneity in the effect size distribution, we select two important institutional factors: country-level investor protection and the equity market development, and research design and journal quality, to explain the mixed and conflicting findings. Our results emphasise the need to consider legal and institutional characteristics, and researcher induced-artefacts, in understanding the role of ownership structure and identity in corporate voluntary disclosure.

Keywords

Investor protection meta-analysis ownership structure ownership type voluntary disclosure

1. Introduction

Corporate voluntary disclosure is an important means for management to communicate firm performance and governance to outsider stakeholders and has remained an important field of empirical and theoretical studies since the 1970s. According to the Financial Accounting Standards Board (FASB, 2001), the term voluntary disclosure describes disclosures, primarily outside the financial statements, that are not explicitly required by generally accepted accounting principles (GAAP) or specific country rules (García-Meca and Sánchez-Ballesta, 2010). When a firm makes the decision to disclose information voluntarily, it assumes that benefits will outweigh costs. Such benefits may come in the form of reduced cost of financing investment opportunities (e.g. cost of equity), lower transactions costs for investors by reducing information asymmetry between the contracting parties and more efficient functioning of capital markets (Healy and Palepu, 2001; Leuz and Verrecchia, 2000).

The literature on voluntary corporate disclosure over the last 40 years has predominantly focused on examining how firms’ specific characteristics are associated with the extent of corporate voluntary disclosures.¹ In recent years, corporate voluntary disclosure studies have shifted towards understanding the role of governance structure in voluntary disclosure primarily using contracting costs and signalling theories (García-Meca and Sánchez-Ballesta, 2010). It has been argued that superior governance structure ensures better monitoring of managers, reduces agency conflicts between contracting parties and improves financial reporting transparency (Fama and Jensen, 1983; Jensen and Meckling, 1976; Watts and Zimmerman, 1990). Other researchers have employed legitimacy theory, political economy theory, stakeholder theory and proprietary cost theory to explain why firms disclose certain types of information voluntarily (Birt et al., 2006; Henderson et al., 2004).

Corporate ownership structure, as part of the governance mechanism, has received increasing attention in recent years (Connelly et al., 2010; Hope, 2013). Ang et al. (2000) and Armstrong et al. (2010) highlight the agency conflicts existing between different groups of equity ownership and their impact on the demand for accounting information. They suggest that the expected relations between the demand for quality accounting information and ownership structure are driven primarily by economic incentives, and that such demand varies cross-sectionally with ownership structure. This finding is due to the emerging insight that different categories of owners may have different preferences and priorities with respect to corporate risk, stability, growth and performance (Douma et al., 2006; Gedajlovic et al., 2005).

Most prior studies document a positive relation between higher quality governance mechanisms and voluntary disclosure (e.g. Ajinkya et al., 2005; Karamanou and Vafeas, 2005; Laksmana, 2008), but with regard to the effect of ownership structure on voluntary disclosure, the empirical results are mixed and, in some cases, conflicting (Arcay and Vazquez, 2005; Armstrong et al. 2010; Brown et al., 2011; García-Meca and Sánchez-Ballesta, 2010; Luo et al., 2006). In their reviews, Artiach and Clarkson (2011) and Brown et al. (2011) attribute the mixed evidence across studies to the variations in the number of disclosure items, the mode and nature of disclosure (e.g. from total voluntary disclosures, to specific issues such as earnings forecasts and corporate social and environmental information), the type of disclosure required by owners and the research setting being examined. The conflicting results could also be due to treating ownership as a homogeneous group by ignoring the identity of the owners, who have different motivations affecting corporate decision-making (Hautz et al., 2013; Heugens et al., 2009).

In recent years, there have been calls for better understanding of the important role of corporate ownership structure, given its strong influence on governance and monitoring globally (Judge, 2011, 2012). In the accounting discipline, García-Meca and Sánchez-Ballesta (2010: 622) undertook a review of 19 studies that examined the relationship between corporate voluntary disclosure and ownership concentration and found a negative association between the two variables.² However, although they did not consider other types of ownership structure, they suggested that further research into other forms of ownership would be very desirable to provide more insights into the ownership and voluntary disclosure relationship.

Following the suggestions outlined above from the literature, in this article, we address this gap in the literature by presenting a review of studies that have examined the relation between corporate ownership concentration and types (namely, institutional, foreign, state and managerial ownerships) and voluntary disclosure and assess the underlying factors that give rise to such divergence in results. We do this by employing meta-analysis techniques, because such procedures are effective ways of drawing valid inferences from a diverse evidence base that has produced mixed findings, since narrative reviews may not capture the underlying reasons for differences in results (Stanley and Doucouliagos, 2012). In particular, we assess the role of institutional characteristics, such as country-specific investor protection levels and capital market development, in the relationship between ownership types and voluntary disclosure. Prior studies suggest that legal and institutional attributes affect financial reporting behaviour (e.g. Bushman and Piotroski, 2006; La Porta et al., 2000; Leuz et al., 2003). To control for the simultaneous effect of investor protection and market development, along with other moderating variables, namely, journal quality, index construction and substantial/family ownership (FO), we employed a more advanced multivariate meta-regression methodology (meta-analytic regression analysis (MARA)), which has not been utilised in prior meta-analysis papers in accounting.

Based on 69 studies, our analyses show that voluntary disclosure, on average, is lower in countries with more concentrated ownership, which is consistent with Leuz et al. (2003) and García-Meca and Sánchez-Ballesta (2010). However, this negative relationship between ownership concentration and voluntary disclosure is not affected by the level of investor protection and other moderating factors, but it is greater in countries with lower capital market development. These results suggest that concentrated owners are in a more entrenched position to expropriate minority shareholders in countries where the equity market is not well developed. Regarding ownership types, we find that institutional owners have a beneficial effect on the extent of voluntary disclosure, as is evident from a significant positive association between the two constructs. The positive effect is more evident in countries where the level of investor protection is low and the capital market is less developed, suggesting that institutional investors demand more public information from firms operating in low investor protection and low capital market development countries, and thus play an effective monitoring role. We also find foreign investors and state ownership are positively associated, but managerial ownership is negatively associated, with voluntary disclosure.

Our study contributes to the debate on the relationship between voluntary disclosure and ownership attributes by offering a quantitative generalisation drawn from a large sample of empirical studies dealing with this topic. We provide evidence that ownership type is not homogeneous, due to different investment objectives and horizons, and as a result, such relationships between voluntary disclosure and ownership structure are affected, both in direction and strength. It is, therefore, important for investors and policymakers to understand the role of ownership identity as a monitoring mechanism in corporate voluntary disclosure policy, rather than focusing on ownership concentration only. Our study also shows that jurisdictional differences (particularly, investor protection level and equity market development), research design and construct measurement are important factors to consider. The in-depth analyses of a large number of studies facilitate theory development and provide guidance to future researchers in conducting empirical studies on this issue.

The remainder of the article is organised as follows: the next section develops the hypotheses. Section 3 explains the research methods employed in this study, including the procedures adopted in collecting the data and their main characteristics. Section 4 presents the results and Section 5 concludes the article.

2. Hypotheses development

Owners differ in terms of power, wealth, competence and non-ownership ties to the firm. These differences affect their objectives and the way they exercise their ownership rights and, therefore, have important consequences for management behaviour with respect to corporate reporting policy (Connelly et al., 2010; Pedersen and Thomsen, 2003). This section reviews the theoretical foundations for the association between ownership concentration and four ownership types (institutional, foreign, state and managerial ownerships) and voluntary disclosure.

2.1. Ownership concentration

Theoretical justifications for information disclosure and ownership concentration derive from two perspectives. Morck (2000), using agency theory, argues that when a firm has widely dispersed ownership, a ‘free rider problem’ will emerge because dispersed owners lack both the means and motives to address managerial agency problems. When information asymmetry and interest misalignment between the owner/principal and the manager/agent are present, problems associated with managerial opportunism will be greater, causing agency costs to rise (Fama and Jensen, 1983; Jensen and Meckling, 1976). To reduce such costs, firms disclose additional information (García-Meca and Sánchez-Ballesta, 2010).

On the one hand, controlling shareholders are typically assumed to have fewer agency conflicts with managers and boards of directors compared with widely owned firms. This is due to there being little separation between ownership and control, hand-selection of directors and managers and more intense monitoring of managers (Armstrong et al., 2010). However, high ownership concentration may increase agency conflicts between majority and minority shareholders if controlling shareholders’ interests converge with management objectives (Fama and Jensen, 1983; Ghazali and Weetman, 2006; Pound, 1988). As such, the demand for high-quality public disclosures and financial reporting to monitor management appears less important for firms with controlling shareholders than for firms with dispersed ownership that rely more heavily on outside directors to monitor management. This situation increases the potential for expropriation of minority shareholders and subsequent withdrawal of relevant information.

Proprietary cost theory has also been employed to explain the relationship between various types of voluntary information and ownership concentration. Proprietary costs arise when private information, if released, may harm the firm’s competitive position. For instance, large investors may prefer to limit disclosure-related costs, such as competitive and preparation costs (Verrecchia, 1983), to increase short-term profitability. However, proprietary costs vary depending on the level of competition and entry barriers within the market that affect voluntary disclosure (Birt et al., 2006; Clarkson et al., 1994).

Based on the above theoretical predictions, the following hypothesis is proposed in this research:

H1. There is a negative association between ownership concentration and corporate voluntary disclosure.

2.2. Institutional ownership

The corporate governance literature suggests that institutional investors play an active role in the monitoring and control of firms that benefit the company by reducing agency costs and information asymmetry (Donnelly and Mulcahy, 2008). Generally, large institutional investors enjoy greater voting power, allowing them to undertake corrective actions when deemed necessary, because they possess financial know-how and are better able to determine their information needs and expect firms to meet their expectations (Donnelly and Mulcahy, 2008; Hope, 2013). Ajinkya et al. (2005) argue that institutional investors desire and demand more voluntary disclosures, and that such disclosures, especially earnings forecasts, are closely watched by market participants. They further suggest that these institutions, as outsiders, might not be able to directly oversee the activities of managers, but could elicit greater transparency by demanding more information from the firm, given their decision-oriented focus. Analytical studies such as that by Diamond and Verrecchia (1991) demonstrate that substantial shareholdings of institutional investors also encourage more dissemination of private information to reduce information asymmetry. Laidroo (2009), in contrast, suggests that when the ownership of institutional investors is high, such investors can negotiate with firms to have direct access to private information, which reduces the companies’ need for dissemination of voluntary information to the general public. From a supply perspective, firms increase voluntary disclosure as a signalling mechanism to potential investors in the stock market prior to raising new or additional funds, and institutions prefer to buy stocks in firms that have sustained disclosure enhancements (Bushee and Noe, 2000; Firth, 1980).

Thus, the following hypothesis is tested in this research:

H2. There is a positive association between institutional ownership and corporate voluntary disclosure.

2.3. Foreign ownership

The corporate governance literature commonly distinguishes between foreign and domestic owners (e.g. Douma et al., 2006). Mangena and Tauringana (2007) argue that foreign investors incur more information and agency costs when investing in a foreign country. In a similar vein, Huafang and Jianguo (2007) suggest that due to geographical separation, local accounting standards, language barriers between shareholders and management and the lack of knowledge of local operating environments, foreign investors face high levels of information asymmetry compared to domestic investors. This phenomenon is commonly known in accounting and finance literature as ‘home bias’. Furthermore, Khanna et al. (2004) provide evidence that foreign investors from countries with better disclosure standards demand greater disclosure to reduce information asymmetry. Accordingly, management may increase voluntary disclosure to reduce home bias among foreign investors. However, it has been argued that if foreign owners become substantial shareholders, firms will have less need for equity finance from the local market since funds are acquired from foreign sources. Therefore, the demand for public information may decline due to the ability of foreign investors to extract such information privately from local owners (Laidroo, 2009).

Using legitimacy theory, Haniffa and Cooke (2005) suggest that the presence of foreign investors may result in a higher legitimacy gap. Therefore, management uses voluntary social and environmental disclosure as a proactive legitimacy strategy to encourage foreign flows of capital and satisfy foreign ethical investors.

Thus, the following hypothesis is tested in this research:

H3. There is a positive association between foreign ownership and corporate voluntary disclosure.

2.4. State ownership

There is no consensus concerning theoretical predictions about the association between state ownership and voluntary disclosure. The resource dependence theory suggests that the presence of the state as a major shareholder reduces a firm’s external dependencies, and thus the firm has less incentive to make voluntary disclosures beyond those required by law (Wang et al. 2008). Bushman et al. (2004) find that state ownership and financial transparency are negatively related, a result consistent with the suppression of information transparency by state-owned firms (because of expropriation activities) or information transparency not being required (because of direct governance over firm management). Similarly, using property rights theory, Hope (2013) suggests that managers in state-owned companies lack incentives to increase corporate profitability and improve corporate reporting policy, since the government owns a large proportion of such companies’ shares.

On the other hand, Williams (1999), using Bourgeois theory of political economy, suggests that government plays an important role in reducing the adverse effects of imperfect competition and externalities by establishing an adequate corporate reporting policy. This theory suggests that when the state has shareholdings in the company, the socio-political and economic system of a nation interacts to shape the perceptions of organisations on the need to release social and environmental information that meet stakeholders’ social expectations. Similarly, Naser et al. (2006), using legitimacy and stakeholder theories, argue that government may put pressure on firms (via government-appointed board members) to disclose more social and environmental information, rather than only financial information, in order to improve social perceptions of firms in which it has significant ownership.

We thus propose the following hypothesis in this research:

H4. There is an association between state ownership and corporate voluntary disclosure.

2.5. Managerial ownership

Managerial ownership (i.e. ownership by executive directors and chief executive officer ownership) also affects a firm’s disclosure policy. Jensen and Meckling (1976) suggest that if directors have shareholdings in the company, their interests will be more in line with those of other shareholders. Therefore, they will have long-term interests in the company and less incentive to expropriate minority shareholders. Accordingly, it is expected that managerial ownership will have a positive effect on voluntary disclosure.

However, beyond a certain limit of managerial ownership, the entrenchment effect takes place (Jensen and Ruback, 1983). Shleifer and Vishny (1997) suggest that when a large portion of the company’s shares are owned by managers, they have incentives to reduce corporate transparency in order to preserve their strong voting power and serve their own interests. For instance, block manager owners have direct access to information and may control a firm’s operating and financial policies by taking executive roles on the board of directors, a strategy that reduces their incentives to disseminate information voluntarily (Chau and Gray, 2010). When the proportion of insider ownership becomes greater, the monitoring capacity of outsiders is reduced, which results in less voluntary disclosure by firms.

Thus, the following hypothesis is proposed in this research:

H5. There is an association between managerial ownership and corporate voluntary disclosure.

3. Methods

3.1. Studies included in meta-analysis

In order to obtain relevant meta-analytic papers, we searched several accounting and finance journals, including search of databases such as ScienceDirect, EJS (EBSCO), Blackwell, Springer, Sage, Emerald, Inderscience, Macmillan, ABI Inform and Social Science Research Network (SSRN). The keywords used include voluntary disclosure, ownership concentration, ownership structure and terms such as managerial, institutional, state and foreign ownerships. We also consulted accounting and corporate finance journals that typically publish studies dealing with corporate governance, corporate ownership and voluntary disclosure. After excluding 29 studies for not meeting the criteria³ required for meta-analysis, the final sample comprised 69 studies. A total of 34 papers were published in ranked journals (A* and A),⁴ 31 empirical studies appeared in other refereed journals (B and C), 2 studies were published in low or un-ranked journals and only 2 papers are unpublished. Table 1 shows the sample selection process. Appendix 1 shows the journal ranking and Appendix 2 summarises the results of the 69 studies by year of publication, country setting, number of companies sampled, reporting years and the effect size measure (r) for each ownership attribute: ownership concentration, institutional, foreign, state and managerial ownerships.

Table 1.

Sample selection.

	Number of studies	Percentage
Initial sample	98	100
Criteria leading to exclusion of studies
Studies providing only descriptive statistics	3	3.061
Other measures of disclosure	2	2.040
Corporate governance variables not included in our meta-analysis	24	10.204

Final sample	69	84.695
	Number of studies	Percentage
Publication quality
Top-ranked journals	34	49.275
Other referred journals	31	44.927
Low-quality journals	2	2.899
Unpublished studies	2	2.899
Total	69	100

3.2. Meta-analysis procedures

We approached our study in two ways. First, we used meta-analysis techniques developed by Hunter and Schmidt (1990), Rosenthal (1991) and Hunter and Schmidt (2000) to compute effect size. This meta-analysis approach reports findings in terms of bivariate effect sizes. Second, to assess the association between the effect size and moderating variables, we performed meta-analytic regression analysis, popularly known as MARA (see Lipsey and Wilson, 2001). The advantage of meta-analytic regression is that it allows us to assess whether jurisdiction-level moderating effects on the associational strength of the focal relationship still hold in the presence of plausible control variables. Miller and Cardinal (1994) suggest that analysing (moderator) variables simultaneously in a multiple regression format allows proper assessment of the relative explanatory power of each variable. This is because these variables directly compete against one another in the same statistical analyses.

3.2.1. Effect size computation

The meta-analysis technique requires computing the effect size⁵ to measure the standardised magnitude of the association between the dependent and independent variables. The effect size is an index used to represent and standardise the findings of the included studies (Lipsey and Wilson, 2001). It provides information about the magnitude of the association between the five ownership forms (ownership concentration, foreign investors, institutional shareholders, state ownership and managerial ownership) and voluntary disclosure level for each study included in the meta-analysis.

In computing the effect size from the reported statistics, a range of procedures were used. When a study reports Pearson’s r coefficient, such a statistic was used to calculate the effect size between the voluntary disclosure score and explanatory variable. Whenever other statistics are reported, such as Student’s t and Z value, we transformed them into r following Rosenthal (1991).⁶ We then implemented three steps as suggested by Hunter and Schmidt (2000): (1) the mean correlation $(\bar{r})$ ,⁷ (2) the observed variance $(S_{r}^{2})$ and the sampling error variance $(S_{e}^{2})$ ⁸ and (3) the variance used to estimate a confidence interval $(S_{r}^{2} / K)$ . The estimates of population mean $(\bar{r})$ and the standard deviation $\sqrt{S_{r}^{2} / K}$ are normally used to construct a 95% confidence interval.⁹ A 95% confidence interval that does not include 0 indicates that there is a true association between the variables of interest (Dalton et al., 1999). We then computed a chi-square statistic to test for moderating variables and determine whether the observed variance is trivial or higher than expected.¹⁰ If the computed chi-square statistic is trivial compared to the tabulated one, the association is considered un-moderated and homogeneous, and the variation across studies is only due to some statistical (random) errors. However, if the computed chi-square statistic is significant, further analyses are conducted to test for moderators and reduce heterogeneity across studies by sub-grouping the sample according to data characteristics (moderators) and study features.

3.2.2. Variables intervening in the association between ownership attributes and voluntary disclosure

The literature documents inconsistency in findings concerning the relationship between different types of ownership and voluntary disclosure. Various factors espoused in the literature could contribute to the results’ heterogeneity, such as legal and institutional characteristics, level of investor protection and market development, among others. Using meta-analysis, it is possible to quantitatively determine if these factors, referred to as moderating factors, have influenced the relationship between ownership forms and voluntary disclosure. In our study, we employed two country-specific moderating factors, investor protection and the level of market development, as well as other researcher-induced factors, to investigate whether the conflicting results between ownership concentration, ownership types and voluntary disclosure could be explained by these factors, or whether the results are just random errors. These two characteristics have generally been used in the literature for explaining differences in voluntary disclosure among countries (Jaggi and Low, 2000; La Porta et al., 2000). We outline how these two factors would act as moderating variables in the association between ownership concentration, ownership types and voluntary disclosure.

Ownership concentration determines the power and incentives of large owners to enforce their objectives. In jurisdictions where markets are underdeveloped, concentrated ownership gives such owners power to get involved in governance and in determining strategies. This power and the incentives will be less under high legal enforcement and high market development settings, since minority investors will be more protected by law and market forces (Pedersen and Thomsen, 1997). For instance, Leuz et al. (2003) suggest that the level of outside investor protection determines the extent of financial information reported to outsiders, noting that weak legal protection appears to result in lower disclosure levels. In addition, a strong equity market is generally characterised by low ownership concentration and high volume of transactions, implying better disclosure and production of information (Salter, 1998). This implies that ownership concentration will have a negative effect on voluntary disclosure in low investor protection and low market development settings.

Different types of owners have different preferences and objectives, and it has been suggested that legal characteristics and market forces affect manager–owner behaviour with respect to disclosure strategy (Beyer et al., 2010). First, institutional investors generally tend to have multiple ties with the firm in which they own shares, since they will also act as creditors with the same firm (Heugens et al., 2009). Large institutional investors enjoy greater voting power and are better able to determine the investors’ information needs (Donnelly and Mulcahy, 2008). In a low disclosure environment, institutional owners would play an important role in improving reporting quality, because investor protection regulations and market forces are less able to control management discretion over corporate reporting policy.

Second, as the primary objective of foreign investors is to protect their investments by realising higher performance, they generally demand more information compared to their domestic counterparts due to information asymmetry (home bias). However, if they find a favourable legal environment that protects their financial interests, they may put pressure on management to go beyond the higher profitability objective to improve corporate reporting policy. Furthermore, foreign investors in a less developed market encourage management to improve corporate reporting policy, since they bring good governance experience from their home markets to ensure that managers operate in line with their expectations (Chibber and Majumdar, 1999; Heugens et al., 2009). Therefore, we expect foreign investors will have a stronger association with voluntary disclosure in a weak legal setting.

Third, the association between state ownership and voluntary disclosure is expected to be affected by the level of investor protection and capital market development in a country. If the prevailing rules and institutions influencing corporate transparency are weak, managers may be less inclined to disclose information beyond mandatory requirements because private investors (who generally demand more information) will have less influence on management dominated by state enterprises. In contrast, in a developed capital market, where there is already a high level of corporate disclosure (due to strong lobbying and competition), firms with substantial state ownership have an incentive to match their private shareholder-dominated counterparts in mitigating reputational loss.

Finally, managerial investors are viewed as insider owners, who combine a substantial equity stake with direct managerial control over the firm (Heugens et al., 2009). They generally invest a part of their wealth in the firms they manage; therefore, they will be more risk-averse than more diversified investors (Fama and Jensen, 1983). This risk aversion may be higher under stringent legal rules (e.g. strong investor protection, high market development), given the increased level of litigation risk that may arise.

Other researcher-induced factors considered in this study are measurement of disclosure index construction, publication quality and measurement of the explanatory variables. The first of these addresses how disclosure construction¹¹ affects the ownership structure and voluntary disclosure relationship. According to Artiach and Clarkson (2011), ‘different disclosure indices often develop within different studies that address relatively similar issues because there is no agreed upon theory to guide the choice or number of items to include within the index’ (p. 26). Accordingly, disclosure index construction is expected to moderate the relationship between ownership structure and types and voluntary disclosure. Publication quality may also moderate the examined relationships, since quality journals conduct a more thorough and rigorous review process (Moller and Jennions, 2001). Finally, meta-analytic accounting literature posits that the measurement of explanatory variables may also affect the examined relationship (Ahmed and Courtis, 1999; García-Meca and Sánchez-Ballesta, 2010).

3.2.3. Measuring moderating factors, additional tests and file drawer problem

For investor protection level, our classification is based on the ranking provided by the World Bank¹² and World Governance Index. High investor protection countries are those which have an investor protection rank lower than the median rank within the sample. Similarly, countries classified as part of the high (low) market development group are those possessing a market development ratio¹³ higher (lower) than the median ratio.

The moderating effect of disclosure index construction was examined by dividing our sample into two groups: (1) low number of items group (disclosure index including a number of items lower than the median) and (2) high number of items group (disclosure index higher than the median). For journal quality, we divided our overall sample into two groups: high-quality journal papers and low-quality journal papers. The first group includes all published papers that appear in A*- and A-ranked journals in the Australian Business Deans Council (ABDC) list (2013), while the second group includes all studies identified in SSRN and published in other journals.¹⁴ Finally, ownership concentration was sub-grouped into percentage of shares owned by substantial shareholders (state ownership) and FO groups. It should be noted here that managerial, government, foreign and institutional ownerships were measured using the percentage of shares owned by managers, government, foreign shareholders and institutions, respectively, so no further grouping was considered necessary.

As published studies may remain undetected even after a thorough search, the reliability of the results could be questioned. Rosenthal (1979) refers to this problem as the ‘file drawer problem’. To determine whether the conclusions drawn from our meta-analysis are likely to be influenced by any such publication bias, we applied Orwin’s (1983) method, which in turn is modified and based on Rosenthal’s (1979) approach. This method requires the estimation of the fail-safe¹⁵ N. The fail-safe N is calculated when significant associations are reported for the meta-analysis.

3.2.4. MARA

Meta-regression represents an extension of sub-group meta-analysis. However, instead of taking into account only one moderator at a time, it tests for all moderators (Higgins and Green, 2011). In meta-regression, the dependent variable is the weighted effect size of an examined relationship, and the explanatory variables can be continuous variables or dummy variables depending on the study characteristics (Hedges and Olkin, 1985). MARA represents a special meta-analytic approach using weighted least squares (WLS) regression analysis in order to assess the association between the effect size and the moderating variables, in an attempt to model and explore the heterogeneity in the effect size distribution (Lipsey and Wilson, 2001).

Therefore, the MARA model in our meta-analytic review is specified as follows

Y_{i} = α_{0} + α_{1} I P R_{i} + α_{2} M K T D_{i} + α_{3} I D C_{i} + α_{4} J Q_{i} + α_{5} E X V M_{i} + ε_{i}

where $Y_{i}$ is the weighted effect size¹⁶ for study_i, $I P R_{i}$ is the investor protection rank for country_i in study_i, $M K T D_{i}$ is the market development index for country_i in study_i, $I D C_{i}$ is the index disclosure construction: dummy variable (1 if the number of items exceeds the median and 0 otherwise), $J Q_{i}$ is the journal quality: dummy variable (1 if quality journal and 0 otherwise), $E X V M_{i}$ is the explanatory variable measure: dummy variable (1 if substantial ownership and 0 if FO).

4. Results

The univariate meta-analysis results are presented in Tables 2 to 4. We first present the overall results, and if the homogeneity test is rejected, we present further analyses according to moderating variables explained earlier. In Table 5, we present the MARA results.

Table 2.

Ownership concentration and voluntary disclosure.

	Sample (n)	Number of studies (K)	Mean correlation $(\bar{r})$	Z-statistic	Observed variance $S_{r}^{2}$	Estimated error variance $S_{e}^{2}$	Percentage explained $S_{e}^{2} / S_{r}^{2}$	95% confidence interval	$χ_{K - 1}^{2}$	Fail-safe N 0.05
Ownership concentration
Overall meta-analysis	8626	39	−0.125	−4.332***	0.032	0.004	13.140	−0.182, −0.068	290.981***	194
Moderating factors
High investor protection	4217	19	−0.102	−2.713***	0.026	0.004	16.480	−0.175, −0.028	115.322***	46
Low investor protection	3596	18	−0.112	−2.570***	0.034	0.005	14.330	−0.197, −0.026	125.592***	45
High market development	3521	17	−0.086	−2.036**	0.030	0.005	15.590	−0.169, −0.003	109.033***	13
Low market development	4187	18	−0.130	−3.335***	0.027	0.004	15.050	−0.207, −0.053	119.588***	41
Additional analysis
High number of items	3704	20	−0.132	−2.727***	0.047	0.005	11.080	−0.227, −0.037	180.552***
Low number of items	4922	19	−0.120	−3.549***	0.021	0.004	17.200	−0.186, −0.054	110.475***	32
FO	706	4	−0.119	−2.400**	0.009	0.005	55.400	−0.217, −0.021	7.225^†	3
SO	7920	35	−0.125	−3.999***	0.035	0.004	12.233	−0.187, −0.064	283.783***	164
Ranked journals	5610	23	−0.164	−5.122***	0.024	0.005	16.390	−0.227, −0.101	140.339***	122
Other journals	3016	16	−0.053	−1.042	0.041	0.005	12.760	−0.152, 0.046	125.414***	−

Dependent variable: voluntary disclosure; FO: family ownership; SO: percentage shares owned by substantial shareholders; high (low) investor protection settings are those that have an investor protection rank lower (higher) than the median; high (low) market development settings are those that have a market development index, respectively, higher (lower) than the median; high (low) group includes disclosure indices comprising number of items higher (lower) than the median value.

†

p < .10; *p < .05; **p < .01; ***p < .001.

Table 3.

Institutional ownership, foreign ownership and voluntary disclosure.

	Sample (n)	Number of studies (K)	Mean correlation $(\bar{r})$	Z-statistic	Observed variance $S_{r}^{2}$	Estimated error variance $S_{e}^{2}$	Percentage explained $S_{e}^{2} / S_{r}^{2}$	95% confidence interval	$χ_{K - 1}^{2}$	Fail-safe N 0.05
Institutional ownership (A)
Overall meta-analysis	4419	19	0.149	3.265***	0.039	0.004	10.320	0.059, 0.239	184.116***	79
Moderating factors
High investor protection	1360	9	0.108	1.735^†	0.035	0.006	18.630	−0.014, 0.229	48.302***	10
Low investor protection	3059	10	0.168	2.625***	0.041	0.003	7.530	0.046, 0.293	132.756***	38
High market development	1095	8	0.131	1.986**	0.035	0.007	20.150	0.001; 0.261	39.708***	10
Low market development	2689	8	0.166	2.150**	0.048	0.003	5.860	0.015, 0.318	136.576***	22
Additional tests
High number of items	1686	10	0.247	2.929***	0.071	0.005	7.300	0.082, 0.413	136.992***	40
Low number of items	2733	9	0.089	2.590***	0.011	0.003	30.600	0.021, 0.156	29.408***	9
Ranked journals	2360	10	0.207	2.810***	0.054	0.004	7.130	0.062; 0.352	140.227***	42
Other journals	2059	9	0.083	2.048**	0.015	0.004	29.070	0.003, 0.162	30.962***	9
Foreign ownership (B)
Overall meta-analysis	4502	20	0.142	3.524***	0.032	0.004	13.200	0.062, 0.220	151.654***	75
Moderating factors
High investor protection	1901	9	0.224	3.745***	0.032	0.004	13.300	0.106, 0.340	67.669***	53
Low investor protection	2601	11	0.082	1.751^†	0.042	0.004	17.400	−0.009, 0.173	63.260***	13
High market development	1973	10	0.049	1.077	0.021	0.005	24.500	−0.040, 0.137	40.840***	–
Low market development	2246	9	0.110	2.500**	0.017	0.004	22.500	0.023, 0.195	39.929***	11
Additional tests
High number of items	1805	9	0.101	2.046*	0.032	0.006	18.700	0.004, 0.251	58.959***	7
Low number of items	2697	10	0.160	2.751**	0.032	0.003	10.000	0.046, 0.278	89.783***	26
Ranked journals	2341	9	0.201	2.892***	0.043	0.004	8.200	0.064, 0.336	110.288***	37
Other journals	2161	11	0.078	2.305**	0.013	0.005	40.200	0.012, 0.143	27.351***	10

Dependent variable: voluntary disclosure; high (low) investor protection settings are those that have an investor protection rank lower (higher) than the median; high (low) market development settings are those that have a market development index, respectively, higher (lower) than the median; high (low) group includes disclosure indices comprising number of items higher (lower) than the median value.

†

p < .10; *p < .05; **p < .01; ***p < .001.

Table 4.

State ownership, managerial ownership and voluntary disclosure.

	Sample (n)	Number of studies (K)	Mean correlation $(\bar{r})$	Z-statistic	Observed variance $S_{r}^{2}$	Estimated error variance $S_{e}^{2}$	Percentage explained $S_{e}^{2} / S_{r}^{2}$	95% confidence interval	$χ_{K - 1}^{2}$	Fail-safe N 0.05
State ownership (A)
Overall meta-analysis	5265	22	0.170	4.277***	0.034	0.004	11.330	0.092, 0.247	194.132***	117
Moderating factors
High investor protection	3004	11	0.268	5.661***	0.024	0.003	12.700	0.175, 0.361	86.596***	26
Low investor protection	2261	11	0.038	0.963	0.017	0.005	27.270	−0.040, 0.117	40.340***	–
High market development	2537	10	0.315	8.093***	0.015	0.003	20.990	0.239, 0.392	47.641***	60
Low market development	2707	11	0.033	0.906	0.015	0.004	27.560	−0.038, 0.104	39.913***	–
Additional tests
High number of items	2081	11	0.192	2.622***	0.059	0.005	8.300	0.048, 0.335	132.587***	38
Low number of items	3787	11	0.132	3.303***	0.018	0.003	15.770	0.211, 0.054	69.771***	26
Ranked journals	2512	11	0.178	2.794***	0.044	0.004	9.210	0.053, 0.302	119.441***	40
Other journals	2753	11	0.163	3.368***	0.026	0.004	14.730	0.068, 0.257	74.669***	37
Managerial ownership (B)
Overall meta-analysis	7108	28	−0.126	−3.734***	0.038	0.004	11.912	−0.192, −0.060	234.821***	126
Moderating factors
High investor protection	3687	14	−0.158	−4.162***	0.020	0.004	17.790	−0.088, 0.233	78.692***	56
Low investor protection	3421	14	−0.091	−1.665^†	0.042	0.004	9.520	−0.166, 0.016	147.108***	–
High market development	3267	14	−0.205	−5.271***	0.021	0.004	18.490	−0.281, −0.189	75.728***	50
Low market development	3177	12	−0.011	−0.267	0.022	0.004	17.220	−0.094, 0.072	69.709***	−
Additional tests
High number of items	2967	14	−0.193	−3.744***	0.037	0.004	11.740	−0.293, −0.091	119.273***	41
Low number of items	4471	14	−0.065	−1.636	0.022	0.003	13.860	−0.143, 0.013	100.999***	−
Ranked journals	3728	14	−0.115	−3.105***	0.035	0.004	10.180	−0.253, −0.057	137.523***	37
Other journals	3380	14	−0.094	−2.156**	0.026	0.004	15.330	−0.179, −0.008	91.310***	25

†

p < .10; *p < .05; **p < .01; ***p < .001.

Table 5.

Meta-analytic regression meta-analysis.

	Ownership concentration	Institutional ownership	Foreign ownership	State ownership	Managerial ownership
Constant	1.344 (1.300)	3.680 (0.020)	0.121 (0.060)	−0.096 (−0.010)	−0.109 (−0.100)
Legal characteristics
IPR	−0.015 (−1.520)	2.322 (1.650)	0.020 (1.070)	0.062 (0.310)	−0.004 (−0.270)
MKTD	−0.005 (−0.720)	0.190 (0.150)	0.006 (0.510)	−0.028 (−0.280)	−0.011^† (−1.850)
Other moderators
Index construction	−0.692 (−0.980)	−1.203 (−0.370)	−1.378 (−1.160)	9.146 (0.840)	1.234 (1.650)
Journal quality	−0.954 (−1.340)	−1.911 (−0.750)	−0.367 (−0.300)	5.819 (0.490)	0.013 (0.020)
EXVM	1.055 (0.710)
Number of studies	37	19	18	21	28
F-statistic	1.200	0.342	0.660	0.749	0.146
Adjusted R²	2.710	4.800	10.100	11.500	11.620

IPR: investor protection rank for country_i in study_i; MKTD: the degree of market development for country_i in study_i; index construction: dummy variable (1 for high number of items and 0 for low number of items); journal quality: dummy variable (1 for high-ranked journals and 0 otherwise); EXVM: explanatory variable measure (1 for substantial ownership and 0 for family ownership).

The number of studies used in MARA is less than those used in the overall meta-analysis because, first, some studies examine several contexts and second, some legal characteristics are not available for certain countries.

t-statistics are reported between parentheses.

†

p < .10; *p < .05; **p < .01; ***p < .001.

4.1. Ownership concentration

The relationship between ownership concentration and voluntary disclosure in 39 studies (n = 8626) is shown in Table 2. The overall meta-analysis shows that ownership concentration is significantly negatively associated with voluntary disclosure, with a mean effect size $(\bar{r})$ of −0.125 (Z = −4.332) and a 95% confidence interval ranging from −0.182 to −0.068. This result is in line with agency theory predictions, which suggest that firms characterised by high ownership concentration are expected to have less information asymmetry resulting from less disperse ownership, since dominant shareholders have direct access to the information needed. This implies less voluntary disclosure. However, the calculated observed variance $(S_{r}^{2})$ shows a high degree of variation (heterogeneity) across the 39 studies. Since heterogeneity is significantly high and the explanatory power of the observed variance is low, sub-group meta-analysis is conducted to reduce it by introducing the moderating variables as explained earlier. The results show that the low investor protection group has a $(\bar{r})$ of −0.112 (Z = −2.570), while it is −0.102 (Z = −2.713) for the high investor protection group, indicating that investor protection level does not moderate the association because the difference in means is not significant. These findings are in line with García-Meca and Sánchez-Ballesta (2010), suggesting that the negative association between ownership concentration and voluntary disclosure exists regardless of the level of investor protection. In contrast, the negative association is more pronounced in low market development ( $(\bar{r}) = - 0.130$ ; Z = −3.335) compared to high market development group ( $(\bar{r}) = - 0.086$ ; Z = −2.036). The difference in means is 0.044, which is highly significant (t-statistic = 11.44),¹⁷ confirming the moderating effect of market development on the examined association. A plausible explanation for the result is that less developed markets are characterised by more concentrated ownership where there are fewer alternative sources of capital, so the use of voluntary disclosure to attract funds is less relevant. The implication is that less agency conflicts between management and dominant shareholders reduce the need for public disclosures of voluntary information.

4.2. Institutional ownership

The relationship between institutional ownership and voluntary disclosure is significantly positive, with a mean effect size $(\bar{r})$ of 0.149 (Z = 3.265, n = 4419) (Table 3(A)). This result supports the theoretical prediction that instit$utional investors are effective monitors, resulting in better disclosure. The sub-group meta-analysis shows a significant relationship between disclosure and institutional ownership in low investor protection countries, with ( $(\bar{r}) = 0.168$ ; Z = 2.625), while it is less significant in high investor protection countries ( $(\bar{r}) = 0.108$ ; Z = 1.735). This implies that in weak investor protection countries, the presence of institutional investors has a more positive effect on voluntary disclosure (the difference between the two $(\bar{r}) = - 0.060$ ; t-statistic = −9.301). Similarly, the association between institutional ownership and voluntary disclosure is more pronounced in low market development settings (0.166; Z = 2.150) compared to the high market development group (0.131; Z = 1.986). The difference in means between these two groups is 0.035, and this is significant (t-statistic = 4.581). This implies that institutional ownership plays a more active monitoring role in promoting corporate disclosure policy when firms operate in less developed markets.

4.3. Foreign ownership

Table 3(B) shows a significant effect size $(\bar{r})$ of 0.142 (Z = 3.524) based on 20 studies (n = 4,502). This result further lends credence to the increasing belief in the corporate governance literature that the investor’s identity matters, since the presence of foreign investors is positively and significantly associated with disclosure levels. This result provides support for agency theory, which suggests that firms with high levels of foreign ownership are expected to experience more information asymmetry due to home bias faced in foreign markets, which, in turn, increases the need to communicate more voluntary information. Foreign investors also demand greater levels of voluntary disclosure to justify their investment in overseas firms.

The level of investor protection has a moderating effect on the relationship, with a mean correlation $(\bar{r})$ of 0.224 (Z = 3.745) for countries with high investor protection levels, while it is less significant for low investor protection settings (0.082; Z = 1.751). The difference in means between these two groups accounts for 0.142, and this is highly significant (p < 0.01). This could be due to the fact that most countries that have high investor protection regulations are also economically developed, where a high level of disclosure standard is maintained. Market development level also moderates the association between foreign ownership and voluntary disclosure, since the difference between low and high developed capital markets is significant. The presence of foreign investors encourages good governance and disclosure practices derived from their home markets, which translates into more voluntary disclosure, especially in less developed markets characterised by weak disclosure traditions.

4.4. State ownership

The relationship between state ownership and voluntary disclosure in 22 studies (n = 5265) shows a significant association, with a $(\bar{r})$ of 0.170 (Z = 4.277) (Table 4(A)). This means that voluntary disclosure is higher in firms when state ownership increases. Such a positive association is in contrast with arguments from agency theory, resource dependence theory and property rights, which posit lower disclosure levels when state ownership is greater due to lack of incentives for management in state-owned companies. We then further examined the nature of disclosure items and underlying theories used to assess state ownership and voluntary disclosure and find that most studies have used voluntary environmental and social disclosures using legitimacy theory which posit a positive relationship between the two variables.

Consistent with our expectation, the positive association between voluntary disclosure levels and state ownership is greater in high investor protection and high market development countries. As stated earlier, firms with substantial state ownership tend to disclose more social and environment items, and the demand for such disclosure tends to be higher where the capital market is developed and legal enforcement is better. This could be seen as state ownership having a complementary role in voluntary disclosure.

4.5. Managerial ownership

Managerial ownership is negatively associated ( $\bar{r} = - 0.126$ ; Z = −3.724, n = 7108) with voluntary disclosure, and this is consistent with agency theory prediction, suggesting that firms with high levels of insider ownership concentration are expected to be entrenched and increase the potential for expropriation of minority shareholders. The dominance of managerial ownership having private access to information reduces managerial incentives to disseminate information voluntarily (Chau and Gray, 2010). This could also be consistent with the substantive argument that as higher managerial ownership aligns interests with shareholders, this results in less need for greater levels of disclosure. We also find the negative association is more pronounced in a high investor protection and high market development setting. One plausible explanation for this outcome is that manager investors are viewed as insider and risk-aversive owners. This means they will be more conservative in communicating voluntary disclosure, especially where stringent legal rules apply (e.g. strong investor protection, high market development), given the increased level of litigation risk that may be incurred in this setting (Beyer et al., 2010).

4.6. Additional tests and file drawer problem

Additional tests examined whether disclosure index construction and publication quality moderate the above associations. With respect to the construction of disclosure index, we find that this factor moderates the association between institutional, foreign, state and managerial ownerships and voluntary disclosure. We also find that the quality of journals moderates all the associations considered in univariate tests.

To test for the stability of the results, we computed the fail-safe N for each significant association. The number of unreported needed studies with insignificant results required to reduce the mean effect size to a specified criterion is generally high. It ranges from 3 to 194 studies, indicating that our meta-analytic findings have not been compromised by the file drawer problem.

4.7. Meta-regression (MARA)

In Table 5, we report the MARA results for each ownership form after controlling for all moderating variables. For managerial ownership, the degree of market development negatively moderates the relationship between management ownership and voluntary disclosure, with a coefficient of −0.011 (t-statistic = −1.850). For the remaining regressions, the moderating variables do not affect the overall relationships between ownership concentration and forms (institutional, foreign and state ownerships) and voluntary disclosure. It should be noted that the MARA methodology controls for underlying correlations among the independent variables. While we find a high level of correlation among the moderating variables, raising concerns about multicollinearity, the variance inflation factor (VIF) statistics are generally lower than the well-accepted cut-off point of 10 (Neter et al., 1989). Furthermore, the difference in the results could also be due to the use of different weightings for the effect size in MARA (Heugens et al., 2009).

5. Discussion and conclusion

Many studies have addressed the effect of ownership structure on voluntary disclosure over the last decade. They have, however, at best produced mixed empirical evidence. The conflicting results could be due to treating ownership as a homogeneous group by ignoring the identity of the owners, who have different motivations affecting corporate decision-making (Hautz et al., 2013). As such, there have been calls to better understand the role of ownership attributes in corporate reporting policy (Armstrong et al., 2010; Brown et al., 2011; García-Meca and Sánchez-Ballesta, 2010). To contribute to the ongoing debate on the role of ownership on disclosure policy, we investigated, using well-accepted meta-analysis techniques, the association between voluntary disclosure and ownership concentration and different ownership types: state, institutional, managerial and foreign ownerships. Furthermore, we tested the moderating effects of the levels of investor protection, market development and a number of other variables, on these associations.

Based on 69 empirical studies undertaken over the last 13 years, our primary analysis shows that ownership concentration and managerial ownership are negatively associated with voluntary disclosures, while state, institutional and foreign ownership types are positively related to voluntary disclosures. Testing for the moderating effects, we document that the negative association between voluntary disclosure and concentrated ownership is greater in low market development settings. In low capital market development settings, concentrated owners become more entrenched and extract information from management for private benefit, thus reducing voluntary information disclosure.

The negative relationship between managerial ownership and disclosures is more evident in countries with strong legal enforcement mechanisms (investor protection and market development). In high investor protection settings, managerial incentives to voluntarily disclose information could be constrained by the potential for litigation and managerial reputation loss, implying that shareholder litigation risk and corporate governance mechanisms are important determinants of voluntary disclosure strategies (Beyer et al., 2010).

The higher level of correlations between voluntary disclosure, and institutional and foreign ownership in weak legal enforcement settings and less developed markets, suggests that these investor categories act as effective monitors and play a substitution role for weak legal enforcement. Finally, state ownership is associated with more voluntary disclosures in high investor protection and high market development settings. This is consistent with arguments that most firms with substantial state owners disclose more social and environment items, and investors’ demand for such disclosure tends to be higher where the capital market is developed and legal enforcement is better.

Overall, our meta-analysis provides cross-country evidence that ownership identity, and legal and institutional characteristics, are important in understanding the relationship between voluntary disclosure and ownership types. One important implication of our meta-analytic findings is that since potential owners differ in terms of wealth constraints, competence, preferences and non-ownership ties to the firm, these differences affect the way owners exercise their ownership rights, which therefore has important consequences for corporate reporting policy. Another important implication is that there is no universal association between ownership structure and voluntary disclosure, since such a relationship depends on the interplay between country- and firm-level governance mechanisms.

Footnotes

Appendix

Appendix 2.

Studies included in the meta-analysis.

Studies	Country	Number of firms	Reporting years	Disclosure proxy	Number of items	Effect size (Pearson’s coefficient)					Sources of information
						OC	FO	GO	IO	MO
Depoers (2000)	France	102	1995	TS	65	−0.401					Table 4, p. 253
Chen and Jaggi (2000)	Hong Kong	87	1993–94	TS	30	0.002					Table 3, p. 302
Haniffa and Cooke (2002)	Malaysia	167	1995	TS	65	0.309	0.054		0.247		Table 3, p. 336
Ferguson et al. (2002)	Hong Kong	142	1995	TS	93			0.374			Table 6, p. 141
Toms (2002)	UK	126	1997	CSED	(a)					−0.195	Table 3, p. 272
Eng and Mak (2003)	Singapore	158	1995	TS	55	0.068		0.369		−0.288	Table 3, p. 337
Oyelere et al. (2003)	New Zealand	229	1998	IBD	(a)	−0.150					Table 7, p. 51
Leung and Horwitz (2004)	Hong Kong	376	1996	TS	44					−0.120	Table 3, p. 367
Makhija and Patton (2004)	Czech	43	1993	TS	66	0.130		0.164	0.452		Table 8, p. 482
Xiao et al. (2004)	China	300	2001	IBD	18		0.040	0.070	0.060		Table 4, p. 212
Cormier et al. (2005)	Germany	43	1992–1998	CSED	37	−0.104	−0.166				Table 3, p. 25
Firer and Williams (2005)	Singapore	390	2000	ICD	53	−0.234		0.166		−0.348	Table 3, p. 12
Hanifa and Rashid (2005)	Malaysia	100	2001	IBD	(a)		0.321		0.462		Table 4, model 1, p. 36
Haniffa and Cooke (2005)	Malaysia	139	2002	CSED	41		0.258				Table 5, p. 413
Mangena and Pike (2005)	UK	262	2002	TS	113				0.212	−0.191	Table 5, p. 338
Barako et al. (2006)	Kenya	43	1993–2001	TS	47	−0.377	0.317		0.633		Table 6, p. 120
Birt et al. (2006)	Australia	286	2001–2003	TS	(a)	−0.103					Table 6
Brammer and Pavelin (2006)	UK	447	2000	CSED	(a)	−0.140					Table 1, p. 1181
Cheng and Courtenay (2006)	Singapore	104	2000	TS	72			0.273	−0.289		Table 3(B), p. 272
Ghazali and Weetman (2006)	Malaysia	87	2001	CSED	53	−0.142		0.147		−0.374	Table 7, p. 241
Luo et al. (2006)	Singapore	172	1994–2000	TS	82			0.454		−0.310	Table 3(B), p. 511
Oliveira et al. (2006)	Portugal	56	2003	ICD	32	−0.470					Table 8, p. 26
Naser et al. (2006)	Qatar	21	2001	CSED	34	0.259		0.207	0.065		Table 4, p. 14
Abdelsalam and Street (2007)	UK	115	2006	IBD	11	−0.002					Table 2, p. 123
Cerbioni and Parbonetti (2007)	SC	54	2002–2004	ICD	28	−0.002					Table 6, p. 812
Hashim and Saleh (2007)	Malaysia	107	2003	TS	60	0.087					Table 7, p. 143
Huafang and Jianguo (2007)	China	559	2002	TS	30	0.163	0.153	0.049	−0.014	0.043	Table 3, p. 613
Lim et al. (2007)	Australia	181	2001	CSED	21	−0.181					Table 3, p. 571
Mangena and Tauringana (2007)	Zimbabwe	51	2002	TS	86	−0.141	0.325		0.243		Table 5, p. 74
White et al. (2007)	Australia	96	2005	ICD	78	−0.061					Table 3, p. 529
Buniamin et al. (2008)	Malaysia	243	2005	CSED	(a)					−0.140	Table 5, p. 79
Donnelly and Mulcahy (2008)	Ireland	51	2002	TS	50				−0.110	−0.160	Table 3, p. 422
Kelton and Yang (2008)	USA	284	2003	IBD	36	−0.176				−0.050	Table 5(D), p. 77
Li et al. (2008)	UK	100	2005	ICD	61	−0.442					Table 2, p. 145
Lim et al. (2008)	Malaysia	743	2003	CSED	27			0.374			Table 3, p. 80
Trabelsi et al. (2008)	Canada	108	2002	IBD	(a)	−0.279					Table 7, p. 141
Wang et al. (2008)	China	110	2005	CSED	17		0.130	0.040			Table 3(B), p. 23
Apostolou and Nanopoulos (2009)	Greece	72	2004	TS	56	0.046					Table 10, p. 410
Bogdan et al. (2009)	Romania	15	2006	TS	84		0.328	0.116	0.605		Table 8, p. 21
Jiang and Habib (2009)	New Zealand	116	2001–2005	TS	39	−0.196		0.014		−0.133	Table 9, p. 293
Laidroo (2009)	Baltics	52	2000–2005	TS	DQ	−0.180	0.110	0.200	0.180	0.260	Table 3, p. 23
Reverte (2009)	Spain	46	2005–2006	CSED	4	−0.342					Table 4, p. 362
Rashid and Lodh (2008)	Bangladesh	21	2003–2007	CSED	21	0.160			−0.140	0.020	Table 7, p. 226
Said et al. (2009)	Malaysia	150	2006	CSED	41	0.206	0.178	0.317		−0.086	Table 7, p. 222
Yau et al. (2009)	Malaysia	60	2003	ICD	21			0.261			Table 7, p. 30
Akhtaruddin and Haron (2010)	Malaysia	124	2003	TS	64	−0.312					Table 3, p. 75
Bokpin and Isshaq (2009)	Ghana	27	2002–2007	TS	98		−0.098				Table 3, p. 695
Broberg et al. (2010)	Sweden	199	2002 and 2005	TS	68		0.102			−0.198	Table 5, p. 364
Broberg et al. (2010)	Sweden	199	2003 and 2005	ICD	6		0.088			−0.058	Table 5, p. 364
Chau and Gray (2010)	Hong Kong	273	2002	TS	85	−0.203					Table 4, p. 101
Da Silva Monteiroand Aibar-Guzmán (2010)	Portugal	109	2002–2004	CSED	16		−0.059				Table 4, p. 197
Elsayed and Hoque (2010)	Egypt	100	2004	TS	70			−0.451			Table 5, p. 27
Felo (2010)	USA	429	2001	TS	98	0.114					Table 3(1), p. 30
Khan (2010)	Bangladesh	30	2007–2008	CSED	60		0.258				Table 7, p. 99
Khodadadi et al. (2010)	Iran	106	2001–2005	TS	33				0.150		Table 5, p. 161
Li and Zhang (2010)	China	692	2008	CSED	36			0.030	0.086		Tables 3 and 4, p. 638
Nor et al. (2010)	Malaysia	93	2005–2006	ICD	(a)	−0.175	−0.036	0.045	0.061	0.005	Table 4, p. 366
Orens et al. (2010)	SC	668	2002	IBD	92	−0.374					Table 5, p. 20
Siregar and Bachtiar (2010)	Indonesia	87	2003	CSED	41		−0.003				Table 2, p. 248
Hidalgo et al. (2011)	Mexico	100	2005–2007	ICD	58	−0.070				−0.110	Table 3, p. 8
Jiang et al. (2011)	New Zealand	103	2001–2005	TS	39				0.079	0.057	Table 2(C), p. 9
Samaha and Dahawy (2011)	Egypt	100	2006	TS	80	−0.485		0.123		−0.059	Table 4, p. 77
Taliyang et al. (2011)	Malaysia	150	2009	ICD	36					−0.255	Table p. 31
Mert (2011)	Turkey	173	2009	IBD	60				0.074	0.054	Table 6 (OLS)
Li et al. (2012)	UK	100	2005	ICD	61					−0.604	Table 4, p. 106
Samaha and Abdallah (2012)	Egypt	66	2011	IBD	72	0.042				0.526	Table 5, p. 165 (WCD-T)
Samaha et al. (2012)	Egypt	61	2008	IBD	87			−0.300		0.179	Table 8, p. 161
Khan et al. (2013)	Bangladesh	116	2005–2009	CSED	20		0.462			−0.330	Table 3, p. 10
Mallin et al. (2013)	USA	135	2005–2007	CSED	121	−0.176				0.007	Table p. 9

TS: total score; CSED: corporate social and environmental disclosure; ICD: intellectual capital disclosure; IBD: Internet-based disclosure; SC: several countries; OC: ownership concentration; FO: foreign ownership; GO: government ownership; IO: institutional ownership; MO: managerial ownership; (a): dummy variable (1 if the firm makes such disclosure and 0 otherwise); DQ: disclosure quality including informativeness, relevance and precision; OLS: ordinary least squares; WCD-T: web-based corporate disclosure total score.

Acknowledgements

The authors gratefully acknowledge the constructive comments and suggestions from Darren Henry, Paul Kim, Robert Rosenthal, seminar participants at La Trobe University, two anonymous referees and editors of this journal, Peter Clarkson and Baljit Sidhu.

Final transcript accepted 25 February 2016 by Peter Clarkson (AE Accounting).

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

1.

Marston and Shrives (1991) and provide reviews of these studies. Ahmed and Courtis (1999), using meta-analytic techniques, find that firm-specific attributes such as firm size, leverage and international cross-listing are positively associated with corporate disclosures (voluntary and aggregate).

2.

Although they examined 27 studies, there were 19 studies on ownership concentration and voluntary disclosure (p. 620). They also examined the association between board independence and voluntary disclosure using 19 prior studies and find a positive association between the variables. , who focus on the distribution of ownership shares (ownership concentration) and voluntary disclosure, have examined two forms of ownership concentration (insiders vs outsiders) with only seven studies.

3.

The criteria considered for the inclusion of studies are as follows: (1) the study should deal with voluntary disclosure with at least one of the above ownership types examined as an explanatory variable and (2) sufficient data are provided in the empirical paper included in the meta-analysis to compute the effect size (e.g. Pearson’s coefficient, t-statistic, Z-statistic, p-value). provides some examples of papers excluded for not meeting these criteria.

4.

For the purpose of ranking, we used the Australian Business Deans Council (ABDC) list of journals (2013), in order to avoid confusion due to the availability of several university-specific and commercial journal rankings.

5.

See for different methods of computing effect size from reported statistics in published and working papers.

6.

Ahmed and Courtis (1999); Ahmed et al. (2013) and Trotman and Wood (1991) followed the same procedures in prior meta-analysis in accounting and auditing $r = \sqrt{t^{2} / (t^{2} + d f)}$ , where df is the degree of freedom and $r = Z / \sqrt{N}$ . We note that this formula will always generate a positive number. According to , cf. Chapter 3), if the regression coefficient is negative, it is necessary to keep the negative sign.

7.

For $\bar{r} = \sum (r_{i} \cdot N_{i}) / \sum N_{i}$ , where n_i is sample size of the study and r_i is Pearson’s correlation coefficient.

8.

The observed variance $(S_{r}^{2})$ and the sampling error variance $(S_{e}^{2})$ are calculated using the following formulas: $S_{r}^{2} = \sum N_{i} (r_{i} - \bar{r})^{2} / \sum N_{i}$ , $S_{e}^{2} = {(1 - {\bar{r}}^{2})}^{2} K / \sum N_{i}$ , where $K$ is the number of individual studies included in the meta-analysis.

9.

$[\bar{r} - (\sqrt{S_{r}^{2} / K}) Z_{0.975}; \bar{r} + (\sqrt{S_{r}^{2} / K}) Z_{0.975}] = [\bar{r} - (\sqrt{S_{r}^{2} / K}) (1.96); \bar{r} + (\sqrt{S_{r}^{2} / K}) (1.96)]$ .

10.

$χ_{K - 1}^{2} = N S_{r}^{2} / {(1 - {\bar{r}}^{2})}^{2}$ , all other notations are defined.

11.

Empirical disclosure literature focuses on the assessment of the level of voluntary disclosure, either in annual reports (total score, intellectual capital information or social and environmental disclosure) or on firms’ websites (Internet-based disclosure). The total score refers to those studies using a mix of information dealing with financial, social and environmental and intellectual capital information.

12.

13.

Average of the ratio of stock market capitalisation to gross domestic product for the period 2008–2012 (World Development Indicators at ).

14.

We have taken appropriate measures not to double count the same paper (published with a different name).

15.

The fail-safe N is calculated using equation (5): $K_{0} = K [E S_{k} / E S_{0} - 1]$ , where $K_{0}$ represents the fail-safe N or the number of non-significant, unpublished studies, K is the number of studies included in the meta-analysis, $E S_{k}$ is effect size of studies included in the analysis and $E S_{0}$ is the criterion effect size of 0.05 significance level, which will reduce the effect size to a specified criterion.

16.

Hedges (1982) and show that the optimal weight to be used in MARA is the inverse variance weight, which is calculated as follows: $1 / \sum N_{i} (r_{i} - \bar{r})^{2}$ .

17.

We calculated mean difference and t-statistics for all moderating variables but not reported.

References

*Abdelsalam

Street

(2007) Corporate governance and the timeliness of corporate internet reporting by UK listed companies. Journal of International Accounting, Auditing and Taxation 16: 111–130.

Ahmed

Courtis

(1999) Associations between corporate characteristics and disclosure levels in annual reports: A meta-analysis. The British Accounting Review 31: 35–61.

Ahmed

Chalmers

Khlif

(2013) A meta-analysis of IFRS adoption and financial reporting quality. The International Journal of Accounting 48: 173–217.

Ajinkya

Bhojraj

Sengupta

(2005) The association between outside directors, institutional investors, and the properties of management earnings forecasts. Journal of Accounting Research 43: 343–375.

*Akhtaruddin

Haron

(2010) Board ownership, audit committees’ effectiveness and corporate voluntary disclosures. Asian Review of Accounting 18: 68–82.

Ang

Cole

Lin

(2000) Agency costs and ownership structure. Journal of Finance 55: 81–106.

*Apostolou

Nanopoulos

(2009) Voluntary accounting disclosure and corporate governance: Evidence from Greek listed firms. International Journal of Accounting and Finance 1: 395–414.

*Arcay

Vazquez

(2005) Corporate characteristics, governance rules and the extent of voluntary disclosure in Spain. Advances in Accounting 21: 299–331.

Armstrong

Guay

Weber

(2010) The role of information and financial reporting, corporate governance and debt contracting. Journal of Accounting and Economics 50: 179–234.

10.

Artiach

Clarkson

(2011) Disclosure, conservatism and the cost of equity capital: A review of the foundation literature. Accounting & Finance 51: 2–49.

11.

*Barako

Hancock

Izan

(2006) Factors influencing voluntary corporate disclosure by Kenyan companies. Corporate Governance: An International Review 14: 107–125.

12.

Beyer

Cohen

Lys

. (2010) The financial reporting environment: Review of the recent literature. Journal of Accounting and Economics 50: 296–343.

13.

*Birt

Bilson

Smith

. (2006) Ownership, competition, and financial disclosure. Australian Journal of Management 31: 235–263.

14.

*Bogdan

Popa

Pop

CM, et al

. (2009) Voluntary disclosure and ownership structure: An exploratory study of Romanian listed companies. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1345267

15.

*Bokpin

Isshaq

(2009) Corporate governance, disclosure and foreign share ownership on the Ghana stock exchange. Managerial Auditing Journal 24: 688–703.

16.

*Brammer

Pavelin

(2006) Voluntary environmental disclosures by large UK companies. Journal of Business Finance & Accounting 33: 1168–1188.

17.

*Broberg

Tagesson

Collin

(2010) What explains variation in voluntary disclosure? A study of the annual reports of corporations listed on the Stockholm stock exchange. Journal of Management & Governance 14: 351–377.

18.

Brown

Beekes

Verhoeven

(2011) Corporate governance, accounting and finance: A review. Accounting & Finance 51: 96–172.

19.

*Buniamin

Alrazi

Johariand

. (2008) An investigation of the association between corporate governance and environmental reporting in Malaysia. Asian Journal of Business and Accounting 1: 65–88.

20.

Bushee

Noe

(2000) Corporate disclosure practices, institutional investors, and stock return volatility. Journal of Accounting Research 38: 171–202.

21.

Bushman

Piotroski

(2006) Financial reporting incentives for conservative accounting: The influence of legal and political institutions. Journal of Accounting and Economics 42: 107–148.

22.

Bushman

Chen

Engel

. (2004) Financial accounting information, organizational complexity, and corporate governance systems. Journal of Accounting and Economics 37: 167–201.

23.

*Cerbioni

Parbonetti

(2007) Corporate governance and intellectual capital disclosure: An analysis of European biotechnology companies. European Accounting Review 16: 791–826.

24.

*Chau

Gray

(2010) Family ownership, board independence and voluntary disclosure: Evidence from Hong Kong. Journal of International Accounting, Auditing and Taxation 19: 93–109.

25.

*Chen

CJP

Jaggi

(2000) Association between independent non-executive directors, family control and financial disclosures in Hong Kong. Journal of Accounting and Public Policy 19: 285–310.

26.

*Cheng

ECM

Courtenay

(2006) Board composition, regulatory regime and voluntary disclosure. The International Journal of Accounting 41: 262–289.

27.

Chibber

Majumdar

(1999) Foreign ownership and profitability: Property rights, control, and the performance of firms in Indian industry. Journal of Law and Economics 42: 209–238.

28.

Clarkson

Kao

Richardson

(1994) The voluntary inclusion of forecasts in the MD&A section of annual reports. Contemporary Accounting Research 11: 423–450.

29.

Connelly

Hoskisson

Tihanyi

. (2010) Ownership as a form of corporate governance. Journal of Management Studies 47: 1561–1589.

30.

*Cormier

Magnan

Van Velthoven

(2005) Environmental disclosure quality in large German companies: Economic incentives, public pressures or institutional conditions? The European Accounting Review 14: 3–39.

31.

Dalton

Daily

Johnson

. (1999) Number of directors and financial performance: A meta-analysis. Academy of Management Journal 42: 674–686.

32.

*Da Silva Monteiro

Aibar-Guzmán

(2010) Determinants of environmental disclosure in the annual reports of large companies operating in Portugal. Corporate Social Responsibility and Environmental Management 17: 185–204.

33.

*Depoers

(2000) A cost-benefit study of voluntary disclosure: Some empirical evidence from French listed companies. The European Accounting Review 9: 245–263.

34.

Diamond

Verrecchia

(1991) Disclosure, liquidity, and the cost of capital. Journal of Finance 46: 1325–1359.

35.

*Donnelly

Mulcahy

(2008) Board structure, ownership, and voluntary disclosure in Ireland. Corporate Governance: An International Review 16: 416–429.

36.

Douma

George

Kabir

(2006) Foreign and domestic ownership, business groups, and firm performance: Evidence from a large emerging market. Strategic Management Journal 27: 637–657.

37.

*Elsayed

Hoque

(2010) Perceived international environmental factors and corporate voluntary disclosure practices: An empirical study. The British Accounting Review 42: 17–35.

38.

*Eng

Mak

(2003) Corporate governance and voluntary disclosure. Journal of Accounting and Public Policy 22: 325–345.

39.

Fama

Jensen

(1983) Separation of ownership and control. Journal of Law and Economics 26: 301–325.

40.

*Felo

(2010) Corporate reporting transparency, board independence and expertise, and CEO duality. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1373942

41.

*Ferguson

Lam

KCK

Lee

(2002) Voluntary disclosure by state-owned enterprises listed on the stock exchange of Hong Kong. Journal of International Financial Management & Accounting 13: 125–152.

42.

Financial Accounting Standards Board (FASB) (2001) Improving business reporting: Insights into enhancing voluntary disclosures. Business Reporting Research Project, January. Norwalk, CT: FASB.

43.

*Firer

Williams

(2005) Firm ownership structure and intellectual capital disclosures. South African Journal of Accounting Research 19: 1–18.

44.

Firth

(1980) Raising finance and firms’ corporate reporting policies. Abacus 16: 100–115.

45.

García-Meca

Sánchez-Ballesta

(2010) The association of board independence and ownership concentration with voluntary disclosure: A meta-analysis. European Accounting Review 19: 603–627.

46.

Gedajlovic

Yoshikawa

Hashimoto

(2005) Ownership structure, corporate strategy and firm performance in Japanese manufacturing industries. Organization Studies 31: 271–300.

47.

*Ghazali

Weetman

(2006) Perpetuating traditional influences: Voluntary disclosure in Malaysia. Journal of International Accounting, Auditing and Taxation 15: 226–248.

48.

Greene

(2008) Econometric Analysis. 6th ed. London: Prentice Hall.

49.

*Hanifa

Rashid

(2005) The determinants of voluntary disclosures in Malaysia: The case of internet financial reporting. UNITAR e-Journal 2: 22–42.

50.

*Haniffa

Cooke

(2002) Culture, corporate governance and disclosure in Malaysian corporations. Abacus 38: 317–349.

51.

*Haniffa

Cooke

(2005) The impact of culture and governance on corporate social reporting. Journal of Accounting and Public Policy 24: 391–430.

52.

*Hashim

Saleh

(2007) Voluntary annual report disclosures by Malaysian multinational corporations. Malaysian Accounting Review 6: 129–156.

53.

Hautz

Mayer

MCJ

Stadler

(2013) Ownership identity and concentration: A study of their joint impact on corporate diversification. British Journal of Management 24: 102–126.

54.

Healy

Palepu

(2001) Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of Accounting and Economics 31: 405–440.

55.

Hedges

(1982) Estimating effect size from a series of independent experiments. Psychological Bulletin 92: 490–499.

56.

Hedges

Olkin

(1985) Statistical Methods for Meta-Analysis. Orlando, FL: Academic Press.

57.

Henderson

Peirson

Harris

(2004) Financial Accounting Theory. Frenchs Forest, NSW, Australia: Pearson Prentice Hall Australia.

58.

Heugens

PMAR

van Essen

van Oosterhout

(2009) Meta-analyzing ownership concentration and firm performance in Asia: Towards a more fine-grained understanding. Asia Pacific Journal of Management 26: 481–512.

59.

*Hidalgo

García-Meca

Martínez

(2011) Corporate governance and intellectual capital disclosure. Journal of Business Ethics 100: 483–495.

60.

Higgins

Green

(2011) Cochrane handbook for systematic reviews of interventions. Available at: http://handbook.cochrane.org/front_page.htm

61.

Hope

(2013) Large shareholders and accounting research. China Journal of Accounting Research 6: 3–20.

62.

*Huafang

Jianguo

(2007) Ownership structure, board composition and corporate voluntary disclosure: Evidence from listed companies in China. Managerial Auditing Journal 22: 604–619.

63.

Hunter

Schmidt

(1990) Methods of Meta-Analysis: Correcting Error and Bias in Research Findings. Beverly Hills, CA: SAGE.

64.

Hunter

Schmidt

(2000) Fixed effects vs. random effects meta-analysis models: Implications for cumulative research knowledge. International Journal of Selection and Assessment 8: 275–292.

65.

Jaggi

Low

(2000) Impact of culture, market forces, and legal system on financial disclosures. The International Journal of Accounting 35: 495–519.

66.

Jensen

Ruback

(1983) The market for corporate control. Journal of Financial Economics 11: 5–50.

67.

Jensen

Meckling

(1976) Theory of the firm: Managerial behaviour, agency costs, and ownership structure. Journal of Financial Economics 3: 305–360.

68.

*Jiang

Habib

(2009) The impact of different types of ownership concentration on annual report voluntary disclosures in New Zealand. Accounting Research Journal 22: 275–304.

69.

*Jiang

Habib

(2011) Ownership concentration, voluntary disclosures and information asymmetry in New Zealand. The British Accounting Review 43: 1–15.

70.

Judge

(2011) The pivotal role of ownership. Corporate Governance: An International Review 19: 505–506.

71.

Judge

(2012) Owner type as emerging area of governance research. Corporate Governance: An International Review 20: 231–232.

72.

Karamanou

Vafeas

(2005) The association between corporate boards, audit committees, and management earnings forecasts: An empirical analysis. Journal of Accounting Research 43: 453–486.

73.

*Kelton

Yang

(2008) The impact of corporate governance on internet financial reporting. Journal of Accounting and Public Policy 27: 62–87.

74.

*Khan

Muttakin

Siddiqui

(2013) Corporate governance and corporate social responsibility disclosures: Evidence from an emerging economy. Journal of Business Ethics 112: 207–223.

75.

*Khan

(2010) The effect of corporate governance elements on corporate social responsibility (CSR) reporting: Empirical evidence from private commercial banks of Bangladesh. International Journal of Law and Management 52: 82–109.

76.

Khanna

Palepu

Srinivasan

(2004) Disclosure practices of foreign companies interacting with US markets. Journal of Accounting Research 42: 475–508.

77.

Khlif

Souissi

(2010) The determinants of corporate disclosure: A meta-analysis. International Journal of Accounting and Information Management 18: 198–219.

78.

*Khodadadi

Khazami

Aflatooni

(2010) The effect of corporate governance structure on the extent of voluntary disclosure in Iran. Business Intelligence Journal 3: 151–164.

79.

*Laidroo

(2009) Association between ownership structure and public announcements’ disclosures. Corporate Governance: An International Review 17: 13–34.

80.

Laksmana

(2008) Corporate board governance and voluntary disclosure of executive compensation practices. Contemporary Accounting Research 25: 1147–1182.

81.

La Porta

Lopez-de-Silanes

Shleifer

. (2000) Investor protection and corporate governance. Journal of Financial Economics 58: 3–27.

82.

*Leung

Horwitz

(2004) Director ownership and voluntary segment disclosure: Hong Kong evidence. Journal of International Financial Management & Accounting 15: 235–260.

83.

Leuz

Verrecchia

(2000) The economic consequences of increased disclosure. Journal of Accounting Research 38: 91–124.

84.

Leuz

Nanda

Wysocki

(2003) Earnings management and investor protection: An international comparison. Journal of Financial Economics 69: 505–527.

85.

*Li

Pike

Haniffa

(2008) Intellectual capital disclosure and corporate governance structure in UK firms. Accounting and Business Research 38: 137–159.

86.

*Li

Pike

Haniffa

(2012) The effect of audit committee characteristics on intellectual capital disclosure. The British Accounting Review 44: 98–110.

87.

*Li

Zhang

(2010) Corporate social responsibility, ownership structure, and political interference: Evidence from China. Journal of Business Ethics 96: 631–645.

88.

*Lim

Matolcsy

Chow

(2007) The association between board composition and different types of voluntary disclosure. European Accounting Review 16: 555–583.

89.

*Lim

Talha

Mohamed

. (2008) Corporate social responsibility disclosure and corporate governance in Malaysia. International Journal of Behavioural Accounting and Finance 1: 67–89.

90.

Lipsey

Wilson

(2001) Practical Meta-Analysis. Thousand Oaks, CA: SAGE.

91.

*Luo

Courtenay

Hossain

(2006) The effect of voluntary disclosure, ownership structure and proprietary cost on the return–future earnings relation. Pacific-Basin Finance Journal 14: 501–521.

92.

*Makhija

Patton

(2004) The impact of firm ownership structure on voluntary disclosure: Empirical evidence from Czech annual reports. Journal of Business 77: 457–491.

93.

*Mallin

Michelon

Raggi

(2013) Monitoring intensity and stakeholders’ orientation: How does governance affect social and environmental disclosure? Journal of Business Ethics 114: 29–43.

94.

*Mangena

Pike

(2005) The effect of audit committee shareholding, financial expertise and size on interim financial disclosures. Accounting and Business Research 35: 327–349.

95.

*Mangena

Tauringana

(2007) Disclosure, corporate governance and foreign share ownership on the Zimbabwe stock exchange. Journal of International Financial Management & Accounting 18: 53–85.

96.

Marston

Shrives

(1991) The use of disclosure indices in accounting research: A review article. British Accounting Review 25: 195–210.

97.

*Mert

(2011) The impact of corporate governance on internet financial reporting: Evidence from Turkey. International Journal of Business Research 11. Available at: http://www.highbeam.com/doc/1G1-272511073.html

98.

Miller

Cardinal

(1994) Strategic planning and firm performance: A synthesis of more than two decades of research. Academy of Management Journal 37: 1649–1665.

99.

Moller

Jennions

(2001) Testing and adjusting for publication bias. Trends in Ecology & Evolution 16: 580–586.

100.

Morck

(2000) Concentrated Corporate Ownership. Chicago, IL: University of Chicago Press.

101.

*Naser

Al-Hussaini

Al-Kwari

. (2006) Determinants of corporate social disclosure in developing countries: The case of Qatar. Advances in International Accounting 19: 1–23.

102.

Neter

Wasserman

Kutner

(1989) Applied Linear Regression Models. Homewood, IL: Irwin.

103.

*Nor

Saleh

Jaffar

. (2010) Corporate governance and R&D reporting in Malaysian MESDAQ market. International Journal of Economics and Management 4: 350–372.

104.

*Oliveira

Rodrigues

Craig

(2006) Firm-specific determinants of intangibles reporting: Evidence from the Portuguese stock market. Journal of Human Resource Costing & Accounting 10: 11–33.

105.

*Orens

Aerts

Cormier

(2010) Web-based non-financial disclosure and cost of finance. Journal of Business Finance & Accounting 37: 1057–1093.

106.

Orwin

(1983) A fail-safe N for effect size in meta-analysis. Journal of Educational Statistics 8: 157–159.

107.

*Oyelere

Laswad

Fisher

(2003) Determinants of internet financial reporting by New Zealand companies. Journal of International Financial Management & Accounting 14: 26–63.

108.

Pedersen

Thomsen

(1997) European patterns of corporate ownership: A twelve-country study. Journal of International Business Studies 28: 759–778.

109.

Pedersen

Thomsen

(2003) Ownership structure and value of the largest European firms: The importance of owner identity. Journal of Management & Governance 7: 27–55.

110.

Pound

(1988) Proxy contests and the efficiency of shareholder oversight. Journal of Financial Economics 20: 237–265.

111.

*Rashid

Lodh

(2008) The influence of ownership structures and board practices on corporate social disclosures in Bangladesh. In: Tsamenyi

Uddin

(eds) Corporate Governance in Less Developed and Emerging Economies (Research in Accounting in Emerging Economies, vol. 8). Bingley: Emerald Group Publishing Limited, pp. 211–237.

112.

*Reverte

(2009) Determinants of corporate social responsibility disclosure ratings by Spanish listed firms. Journal of Business Ethics 88: 351–366.

113.

Rosenthal

(1979) The ‘file drawer problem’ and tolerance for null results. Psychological Bulletin 86: 638–641.

114.

Rosenthal

(1991) Meta-Analytic Procedures for Social Research (Applied Social Research Methods Series, vol. 6). Newbury Park, CA: SAGE.

115.

*Said

Zainuddin

Haron

(2009) The relationship between corporate social responsibility disclosure and corporate governance characteristics in Malaysian public listed companies. Social Responsibility Journal 5: 212–226.

116.

Salter

(1998) Corporate financial disclosure in emerging markets: Does economic development matter? The International Journal of Accounting 33: 211–234.

117.

*Samaha

Abdallah

(2012) Further evidence on web-based corporate disclosures in developed versus developing countries: A comparative analysis of nature and determinants in Egypt and the United Kingdom. International Journal of Disclosure and Governance 9: 148–180.

118.

*Samaha

Dahawy

(2011) An empirical analysis of corporate governance structures and voluntary corporate disclosure in volatile capital markets: The Egyptian experience. International Journal of Accounting, Auditing and Performance Evaluation 7: 61–93.

119.

*Samaha

Dahawy

Abdel-Meguid

. (2012) Propensity and comprehensiveness of corporate internet reporting in Egypt: Do board composition and ownership structure matter? International Journal of Accounting and Information Management 20: 142–170.

120.

Shleifer

Vishny

(1997) A survey of corporate governance. Journal of Finance 52: 737–783.

121.

*Siregar

Bachtiar

(2010) Corporate social reporting: Empirical evidence from Indonesia Stock Exchange. International Journal of Islamic and Middle Eastern Finance and Management 3: 241–252.

122.

Stanley

Doucouliagos

(2012) Meta-Regression Analysis in Economics and Business. Oxford: Routledge.

123.

*Taliyang

Abdul Latif

Mustafa

(2011) The determinants of intellectual capital disclosure among Malaysian listed companies. International Journal of Management and Marketing Research 4: 25–33.

124.

*Toms

(2002) Firm resources, quality signals and the determinants of corporate environmental reputation: Some UK evidence. British Accounting Review 34: 257–282.

125.

*Trabelsi

Labelle

Dumontier

(2008) Incremental voluntary disclosure on corporate websites, determinants and consequences. Journal of Contemporary Accounting & Economics 4: 120–155.

126.

Trotman

Wood

(1991) A meta-analysis of studies on internal control judgements. Journal of Accounting Research 29: 180–192.

127.

Verrecchia

(1983) Discretionary disclosure. Journal of Accounting and Economics 5: 179–194.

128.

*Wang

Sewon

Claibornem

(2008) Determinants and consequences of voluntary disclosure in an emerging market: Evidence from China. Journal of International Accounting, Auditing and Taxation 17: 14–30.

129.

Watts

Zimmerman

(1990) Positive accounting theory: A ten year perspective. The Accounting Review 65: 131–156.

130.

*White

Lee

Tower

(2007) Drivers of voluntary intellectual capital disclosure in listed biotechnology companies. Journal of Intellectual Capital 8: 517–537.

131.

Williams

(1999) Voluntary environmental and social accounting disclosure practices in the Asia-Pacific region: An international empirical test of political economy theory. The International Journal of Accounting 34: 209–238.

132.

World Bank (2013) World development indicators. Available at: http://data.worldbank.org/data-catalog/world-development-indicators/wdi-2013

133.

*Xiao

Yang

Chow

(2004) The determinants and characteristics of voluntary-based disclosures by listed Chinese companies. Journal of Accounting and Public Policy 23: 191–225.

134.

*Yau

Chun

Balaraman

(2009) Intellectual capital reporting and corporate characteristics of public-listed companies in Malaysia. Journal of Financial Reporting and Accounting 7: 17–35.