Introduction
Research on corporate information disclosure has grown rapidly during the last 15 years. A timely and detailed disclosure of corporate information enables capital market investors to obtain sufficient and reliable knowledge about companies in order to make informed economic decisions and reduces the information asymmetry between companies and their investors. The content of disclosure not only makes investors more fully aware of a firm’s financial and operational condition, but it also reflects firm managers’ incentives and preferences to disclose relevant information. Significant reforms for greater transparency and disclosure by companies have been initiated around the world. These changes in institutional regulation clearly impact firm management’s motivation to disclose information. Most previous studies are based on agency theory perspectives to investigate the association between corporate governance and information disclosure. However, most of empirical research on information disclosure has been conducted in developed markets such as the United States and European countries (Chen, Chen, & Cheng, Reference Chen, Chen and Cheng2008; Vander Bauwhede & Willekens, Reference Vander Bauwhede and Willekens2008; Garcia-Meca & Sanchez-Ballesta, Reference Garcia-Meca and Sanchez-Ballesta2010); only a few studies have examined a sample of companies in Asian countries (e.g., Chau & Gray, Reference Chau and Gray2002, Reference Chau and Gary2010; Cheng & Courtenay, Reference Cheng and Courtenay2006). Nonetheless, cultural dissimilarities and institutional divergence among nations may lead to different relationships between corporate governance structure and information disclosure. Therefore, the relationship between corporate governance and disclosure quality in East Asian countries is worthy of further research.
The family-governance model continues to be relevant as Taiwan’s economy shifts from entrepreneurial capitalism to managerial capitalism, and the family-governance model has been treated as possessing both a normative belief structure and a cognitive framework about legitimate practices in Greater China societies. To the extent that family wealth is closely linked to firm wealth, family businesses have substantial incentives to maintain firm survival (Anderson & Reeb, Reference Anderson, Mansi and Reeb2003). We therefore incorporate agency arguments with an institutional perspective to analyze the influence of family ownership on the relationship between governance practices and disclosure quality.
This study makes three contributions to the corporate governance and family business literature. First, this study investigates the moderating effect of family ownership on governance practices and disclosure quality, using the perspectives of institutional conformity and socioemotional wealth (SEW) theories. Ownership type can influence corporate disclosure decisions because different types of owners often have different incentives and preferences. Agency theory proposes that firms with concentrated ownership will disclose less information because the dominant shareholders normally have access to corporate information (Jensen & Meckling, Reference Jensen and Meckling1976; Cormier, Magnan, & Velthoven, Reference Cormier, Magnan and Van Velthoven2005). When families own a large percentage of shares in a firm, however, family members may exert their power to force the firm to promote family interests (Anderson, Mansi, & Reeb, Reference Anderson and Reeb2003; Brundin, Samuelsson, & Melin, Reference Brundin, Samuelsson and Melin2014) including the protection of SEW (Gomez-Mejia, Haynes, Nuñez-Nickel, Jacobson, & Moyano-Fuentes, Reference Gomez-Mejia, Makri and Larraza-Kintana2007). In order to maintain a good reputation and project a positive family image, a family-owned firm is more likely to respond to institutional pressures in a more substantive manner than is its nonfamily counterpart. Some scholars report that family firms convey financial information of higher quality than nonfamily firms (e.g., Cascino, Pugliese, Mussolino, & Sansone, Reference Cascino, Pugliese, Mussolino and Sansone2010). Corporate information disclosures are complex phenomena that cannot be explained solely by agency-based economic theory. The findings thus contribute to our understanding how family effect interacts with board structure and the process of disclosure quality.
Second, the study uses an aggregate measurement of information disclosure – that is, a transparency raking – to examine the association between attributes of corporate governance and disclosure quality. Most prior research, particularly in the accounting and finance arena, has focused on voluntary disclosure (Ho & Wong, Reference Ho and Wong2001; Eng & Mak, Reference Eng and Mak2003; Lim, Matolcsy, & Chow, Reference Lim, Matolcsy and Chow2007; Donnelly & Mulcahy, Reference Donnelly and Mulcahy2008); in contrast, few researchers examine the variation in the level of mandatory disclosure (Gao & Kling, Reference Gao and Kling2012). The amount of detail contained in filings by firms may vary prominently even though all firms are in compliance with regulatory requirements. Disclosure quality can be evaluated by the credibility and usefulness of the information provided, which includes examination of the timeliness, precision, completeness, and overall compliance. Instead of investigating voluntary or mandatory disclosure individually, our study addresses broader information disclosure characteristics including mandatory and unregulated disclosure. Several global rating agencies, such as Standard & Poor’s and Financial Analysts Federation, have launched disclosure evaluations and rankings; however, the evaluation criteria selected by those rankings do not provide an overall evaluation of disclosure practices in Taiwanese local markets. Therefore, we adopt the ranking results from the Taiwan Information Disclosure and Transparency Rankings System (IDTRS)Footnote 1 , as the proxy for the quality of information disclosure. The IDTRS gauges the level of corporate information disclosure by considering five categories, including compliance with mandatory disclosures, timeliness of reporting, disclosure of financial forecasts, disclosure in annual reports, and corporate website disclosure.
Third, despite of the widespread diffusion of common governance guidelines, governance practices still can vary significantly around the world (Young, Ahlstrom, & Bruton, Reference Young, Ahlstrom and Bruton2004; Liu, Wang, Zhao, & Ahlstrom, Reference Liu, Wang, Zhao and Ahlstrom2013). Our findings provide insights into the determinants of disclosure quality of listed companies in Taiwan that are representative of firms in the newly industrialized economies, which extend studies of corporate governance beyond Western countries. Taiwanese companies have been characterized as an ‘important research laboratory’ for developing corporate research (Filatotchev, Lien, & Piesse, Reference Filatotchev, Lien and Piesse2005, p. 258).
Theory and Hypotheses
Previous research has shown that the quality of information disclosure is often dependent on both the corporate governance mechanisms in place at the focal firm and the extent to which such governance controls operate to monitor management. Moreover, ownership concentration can affect governance effectiveness either positively or negatively. Thus, we posit that governance structures, board activity, and family ownership will influence the level of disclosure quality among publicly traded firms.
Governance structure and disclosure quality
In order to strengthen corporate governance mechanisms, regulators in Taiwan have enacted the Corporate Governance Best-Practice Principles, which are a set of recommendations regarding the behavior and structure of the board of directors of a firm and are designed to encourage listed companies to adopt codes of good governance. Under the Corporate Governance Best-Practice Principles, two corporate governance practices are extensively encouraged: independent board composition and non-CEO duality.
Board independence
Boards of directors of listed companies provide a governance safeguard to both equity capital and managerial employment contracts. As Fama (Reference Fama1980) and Mizruchi (Reference Mizruchi1983) noted, the board’s most important role is to scrutinize the highest decision makers, and the board is the ultimate center of control in a publicly held organization. Researchers have typically evaluated the effects of board monitoring through the use of proxy variables, most commonly, the proportion of outside directors on a board (e.g., Kosnik, Reference Kosnik1990; Boyd, Reference Boyd1994; Zajac & Westphal, Reference Zajac and Westphal1994). According to agency theory, independent boards are likely to enhance the oversight function, alleviate the agency problem, and ensure that managers act in the interests of shareholders (Jensen & Meckling, Reference Jensen and Meckling1976). According to Mace (Reference Mace1986) and other agency theorists, the board is an important element of corporate governance and, therefore, outside directors can monitor management more effectively because they are independent from the company’s managers and because they have experience from their positions as managers of other firms. Zajac and Westphal noted that ‘insider-dominated boards imply problematic self-monitoring and particularly weak monitoring of the CEO, since the CEO is likely to be in a position to influence the insider directors’ career advancement within the firm’ (Reference Zajac and Westphal1994, p. 125). Conversely, outside directors bring a sense of impartiality when evaluating decisions made by a firm’s management (Baysinger & Hoskisson, Reference Baysinger and Hoskisson1990). Further, while executive directors may be directly impacted by the outcomes of their decisions, outsiders can remain neutral and thus can make more objective solutions (Rechner, Sundaramurthy, & Dalton, Reference Rechner, Sundaramurthy and Dalton1993). Based on an agency perspective, studies have used the presence of outsiders as a measure of board vigilance (Finkelstein & D’Aveni, Reference Finkelstein and D’Aveni1994), the ability to monitor managerial activity (Tosi & Gomez-Mejia, Reference Tosi and Gomez-Mejia1994), the power of the board over the CEO (Westphal, Reference Westphal1998), and committee effectiveness (Conyon & Peck, Reference Conyon and Peck1998). Similarly, the number of board seats held by outside members has been used as a proxy for expertise (Subrahmanyam, Rangan, & Rosenstein, Reference Scott1997).
Using agency theory arguments, the vast majority of research investigating the relationship between board independence and information disclosure has suggested that a higher proportion of independent directors will lead to higher levels of voluntary disclosure (Filatotchev, Zhang, & Piesse, Reference Filatotchev, Zhang and Piesse2011). Independent directors provide the compulsory checks of corporate financial and operational information and may thus have greater incentives than inside directors to encourage companies to disclose more information to outside stakeholders (Fama & Jensen, Reference Fama and Jensen1983). Empirical evidence on the relationship, however, is mixed. Some studies report a positive relationship between the proportion of independent directors and the extent of voluntary disclosure (e.g., Leung & Horwitz, Reference Leung and Horwitz2004; Cheng & Courtenay, Reference Cheng and Courtenay2006). Conversely, other studies have yielded negative correlations (e.g., Eng & Mark, Reference Eng and Mak2003; Gul & Leung, Reference Gul and Leung2004), or no significant relationships (e.g., Ho & Wong, Reference Ho and Wong2001). These inconsistencies lead us to examine the relationship between independent board of directors and disclosure quality using a comprehensive measure, which not only includes voluntary disclosure but also variations within mandatory disclosure requirements and the timeliness of such disclosures.
Following governance practices espoused for Western corporations, reforms in the greater China society foster independence in board membership (Tian & Lau, Reference Tian and Lau2001). Under the Company Law and Securities and Exchange Law in Taiwan, listed companies are encouraged but not currently forced to appoint independent directorsFootnote 2 as they are required in Anglo-American countries. A firm that voluntary invites outsiders as its independent directors demonstrates its willingness to implement better corporate governance practices and reduces incentives to withhold information. However, not all Taiwanese firms were prepared, culturally or organizationally, to embrace Western governance practices, including expansion of outside representation on the board, causing variation in the extent of information disclosure. Accordingly, we expect that the proportion of independent directors on a corporate board would be positively associated with disclosure quality.
Hypothesis 1: The proportion of independent directors is positively associated with a firm’s information disclosure quality.
CEO duality
Holding the position of CEO is an indicator of one’s power, while holding multiple titles, such as a CEO who jointly serves as the board chairperson, may enhance the CEO’s power. Several researchers have suggested that CEOs who also hold the chairperson title have more structural power than those who hold only the CEO title (Harrison, Torres, & Kukalis, Reference Harrison, Torres and Kukalis1988; Ocasio, Reference Ocasio1994). Dalton and Dalton described CEO duality as one of two ‘contemporary and intensely contentious issues related to the governance of publicly traded companies’ (2011, p. 405).
According to agency theory, the dual structure provides the potential for CEO domination of the board of directors and promotes CEO entrenchment by reducing board monitoring effectiveness (Finkelstein & D’Aveni, Reference Finkelstein and D’Aveni1994). The unification of CEO and board chairperson represents the ultimate exercise of executive power. Without an independent chair, a board may find its monitoring role to be particularly difficult, because a CEO who is also the chair of the board can control the agenda of board meetings, determine what information directors receive in advance of meetings, and dominate board meeting discussions. Previous empirical studies offer some evidence that companies with CEO duality affect information disclosure (Carcello & Nagy, Reference Carcello and Nagy2004; Gul & Leung, Reference Gul and Leung2004; Lakhal, Reference Lakhal2005). For example, Davidson, Jiraporn, Kim, and Nemec (Reference Davidson, Jiraporn, Kim and Nemec2004) showed that CEO duality is associated with greater earnings management.
In Taiwan, the positions of CEO and chairman of the board are often held by the same person or two persons from the same family. Effective oversight of the board is seriously compromised when the CEO formally dominates the board as the chairperson. Under these circumstances, directors may feel unable to ask difficult questions, raise critical issues, or make correct judgments. The lack of independent leadership in a firm with a single CEO-Chairman or a kinship relation between the chairman and CEO will reduce monitoring by the board and hence increase the tendency to withhold information from outside stakeholders. We therefore hypothesize the following:
Hypothesis 2: The dual position of CEO and chairman is negatively associated with a firm’s information disclosure quality.
Active boards and disclosure quality
Boards of directors need to be active to carry out their corporate governance commitments, particularly in ensuring high quality, transparent disclosure in annual reports. Some corporate boards may be more active and vigilant than others. Meetings provide board directors with the chance to come together, to advise managers on the firm’s strategic orientation, and to perform their duties consistent with shareholders’ interests. Board meetings also cause the company and its directors to incur costs, including managerial time, travel expenses, and directors’ meeting fees. Thus, board meetings can be viewed as a proactive measure for improved governance leading to a better disclosure quality. We use two proxies to measure board activity. Our first proxy is the frequency of board meetings (Vafeas, Reference Vafeas1999). Prior studies have demonstrated that boards who meet frequently are more likely to perform their duties more effectively and result in improved financial performance (Conger, Finegold, & Lawler, Reference Conger, Finegold and Lawler1998; Vafeas, Reference Vafeas1999). For example, Xie, Davidson, and DaDalt (Reference Xie, Davidson and DaDalt2003) found an association between the meeting frequency of boards and lower levels of managerial earnings management. On the contrary, boards that meet infrequently may play only rubber-stamp roles and may not ensure firms’ compliance with regulations and institutional expectations. We therefore predict a firm with more frequent board meetings tends to have a higher level of disclosure quality.
Our second proxy uses the attendance rate at board meetings. Attending meetings is the primary opportunity for directors to acquire company information, interact with other board members, and monitor management decisions. Busy directors, who hold multiple board seats, may face tight time constraints and limited attention capacities (Jiraporna, Davidson, DaDalt, & Ning, Reference Jiraporn, Davidson, DaDalt and Ning2009; Lin, Yeh, & Yang, Reference Lin, Yeh and Yang2014). Failure to attend meetings may limit the directors’ ability to do their jobs effectively and may lessen their contribution to improve the firm’s governance. As the complexity of a firm’s operations and finances requires directors to spend time spent for review, busy directors are less likely to question the information in operational and financial reports provided by managers and are therefore less effective monitors. Accordingly, a higher attendance rate indicates directors’ strong commitment to perform their duties and promotes more effective functioning of the board. We expect to find a positive relationship between board attendance rate and the level of disclosure quality. Based on the discussion above, we develop the following two hypotheses:
Hypothesis 3: The number of board meetings is positively associated with a firm’s information disclosure quality.
Hypothesis 4: The attendance rate of directors at board meetings is positively associated with a firm’s information disclosure quality.
The moderating effect of family ownership on disclosure quality
Information disclosure can be one way for listed companies to signal that they act in the best interest of the owners. Previous studies, following the logic of agency theory, have examined the impact of controlling owners on corporate disclosures. Agency theorists assert that when ownership becomes concentrated, firms will be less likely to disclose information because controlling owners have greater access to internal information and need to rely less on public disclosure to monitor their investments; furthermore, they have an incentive to avoid disclosing detailed information that could attract close monitoring by outside stakeholders (Jensen & Meckling, Reference Jensen and Meckling1976; Chau & Gray, Reference Chau and Gray2002; Mohamed & Sulong, Reference Mohamed and Sulong2010). However, when ownership is controlled by a family, agency-based economic explanations seem insufficient to explain some empirical findings (Gomez-Mejia, Cruz, Berrone, & Castro, Reference Gómez-Mejía, Cruz, Berrone and De Castro2011). For example, Wang (Reference Wang2006), using data from the Standard & Poor’s 500 companies for the period 1994–2002, suggested that founding family ownership enhances the communication between insiders and users of financial statements through high-quality financial information. Some scholars also find that listed family firms are less likely to manage earnings and more likely to convey financial information of higher quality compared with nonfamily firms (Jiraporn & DaDalt, Reference Jiraporn and DaDalt2009; Cascino et al., Reference Cascino, Pugliese, Mussolino and Sansone2010).
We suggest two theoretical arguments that explain why family firms may be more motivated to disclose otherwise confidential information. According to institutional theory, organizations are strongly influenced by their institutional environments and, as a result, adopt structures that are believed to further their legitimacy within the environment (Scott, 1992; Deephouse & Suchman, 2008). Firms conforming to environmental norms will secure greater legitimacy and hence attract resources from external stakeholders. Meyer and Rowan (Reference Meyer and Rowan1977) first suggested that to achieve legitimacy among their constituents, organizations adopted symbolic processes and structures that corresponded to socially prescribed rules regarding organizational behavior. DiMaggio and Powell (Reference DiMaggio and Powell1983) further developed this theme, tying it more explicitly to organizational and sociological theory. Noting the remarkable similarity of organizations within industries, they argued that this similarity arose not because of competition or an objective requirement of efficiency, but rather as a result of organizations’ quests to attain legitimacy within their larger environments. As accepted practices within an organizational field become more widespread, firms within the field adopt similar practices to conform to these norms. In the area of corporate governance, evidence of institutional pressures to shape organizational structures has been found to exist with respect to board of director composition (Jones & Goldberg, Reference Jones, Makri and Gómez-Mejía1982; Luoma & Goodstein, Reference Luoma and Goodstein1999) and subcommittee formation (Kalbers & Fogarty, Reference Kalbers and Fogarty1998; Newman & Mozes, Reference Newman and Mozes1999).
Family firms may more inclined to pursue those activities that conform to industrial norms because of their concern with social capital and reputation. Institutional theory suggests that family firms will mimic the actions of large publically traded companies to enhance legitimacy and reputation. As more family firms adopt such well-regarded behaviors, other similarly sized family-run companies will feel compelled to act in an analogous fashion, including the provision of enhanced information disclosure to the public.
A second more recent theory suggests that family firms are unique organizational forms because they are focused on the attainment of both economic goals and family-centered goals, which leads to the creation of SEW (Berrone, Cruz, & Gomez-Mejia, Reference Berrone, Cruz and Gómez-Mejía2012). The concept of SEW suggests the family gains affective value simply from family members’ association with the family firm (Berrone, Cruz, Gómez-Mejía, & Larraza-Kintana, Reference Berrone, Cruz, Gómez-Mejía and Larraza-Kintana2010). In addition to pursuing purely economic gains, family firms are also interested in preserving family values and maintaining harmony among family members (Gomez-Mejia et al., Reference Gomez-Mejia, Makri and Larraza-Kintana2007; Chrisman & Patel, Reference Chrisman and Patel2012). Decision making in family firms is often dominated by their desire to preserve their SEW such as maintaining family control and management of the business (Bertrand & Schoar Reference Bertrand and Schoar2006; Gómez-Mejía, Cruz, Berrone, & Castro, Reference Gómez-Mejía, Cruz, Berrone and De Castro2011), extending family values and reputation (Westhead, Cowling, & Howorth, Reference Westhead, Cowling and Howarth2001; Sharma & Manikutty, Reference Subrahmanyam, Rangan and Rosenstein2005), and securing succession and control for later generations (Chrisman, Chua, & Sharma, Reference Chrisman, Chua and Sharma2005). Maintaining SEW is also essential for fulfilling family members’ needs for belonging and intimacy and discharging family obligations (Zellweger, Kellermanns, Chrisman, & Chua, Reference Zellweger, Kellermanns, Chrisman and Chua2012).
According to Gómez-Mejía et al. (Reference Gomez-Mejia, Makri and Larraza-Kintana2007), preserving SEW is essential for the family and becomes the primary reference point for guiding managerial choices. Thus, family firms are more willing to accept lower financial performance in order to prevent a loss of SEW (DeTienne & Chirico, Reference DeTienne and Chirico2013). For example, in the study by Gomez-Mejia et al. (Reference Gomez-Mejia, Makri and Larraza-Kintana2007), the authors found that family-owned olive oil producers would rather remain independent than join a cooperative even though the cooperative offered more financial security. Maintaining family control is also evident in many studies which found that a family firm would prefer less diversification and increased risk because diversification would require appointment of nonfamily members to business units and reduce family influence (e.g., Gomez-Mejia, Makri, & Larraza-Kintana, Reference Gómez-Mejía, Haynes, Nuñez-Nickel, Jacobson and Moyano-Fuentes2010). Finally, similar to institutional theory, SEW theory suggests that reputation is critical to family firms because family members identify with their firms and thus are motivated to pursue activities which heighten their family’s reputation (Deephouse & Jaskiewicz, Reference Deephouse and Jaskiewicz2013). Empirical studies supporting this perspective found that family firms are more responsive to institutional pressures in pursuing environmental-friendly policies in order to enhance their family’s image (Berrone et al., Reference Berrone, Cruz, Gómez-Mejía and Larraza-Kintana2010; Miller, Breton-Miller, & Lester, Reference Miller, Breton-Miller and Lester2013).
Based on institutional and SEW theories, this paper posits that family firms are different from nonfamily corporations and will be more predisposed to release information to the public. Further, this tendency will be even more prevalent for family-owned firms in Taiwan. After the Asian financial crisis and corporate fraud scandals, listed family companies in Taiwan became under great institutional pressure to engage in visible behaviors to enhance firm legitimacy, such as improving information transparency. Family ownership is a well-known influence on how the firm can be managed through controlling the board. The controlling shareholders have ultimate power over the board, including power to decide board composition by virtue of owing substantial voting rights. Family owners may view the board as a tool to fulfill the family’s agenda to preserve its SEW (Jones, Makri, & Gómez-Mejía, Reference Jones and Goldberg2008). Following the above rationale, we predict that in Taiwanese-listed firms, family ownership has a positive moderating effect on the relationship between governance structure and information disclosure. Thus, we propose the following:
Hypothesis 5a : The negative relationship between CEO duality and information disclosure quality is weaker for firms with higher family ownership.
Hypothesis 5b: The positive relationship between the proportion of independent directors and information disclosure quality is stronger for firms with higher family ownership.
Hypothesis 5c: The positive relationship between board meeting frequency and information disclosure quality is stronger for firms with higher family ownership.
Hypothesis 5d: The positive relationship between board attendance rate and information disclosure quality is stronger for firms with higher family ownership.
Research Methods
Sample and data sources
The sample of firms in this study are those listed on the Taiwan Stock Exchange Corporation and the Gre Tai Securities Market, covering fiscal years from 2006 to 2010. Corporations in the finance and insurance sectors were excluded because the regulation of disclosure for those sectors differs from that of other corporations. The study used secondary data from the following sources: the Taiwan Economic Journal, the Market Observation Post System, and firm annual reports. We identified Taiwan’s listed firms that were included continuously on the Taiwan Economic Journal database during the study period and were ranked by the IDTRS. This resulted in 2,580 available cases (516 firms multiplied by 5 years). Data on board meeting frequency and attendance rates were gathered from the firms’ annual reports. Data to compute the proportion of independent directors, CEO duality, family ownership, and other control variables were collected from the Taiwan Economic Journal and the Market Observation Post System.
Measures
Dependent variable
We used the ranking results released by the IDTRS as a comprehensive measure of information disclosure quality. The IDTRS obtains information from annual reports, regulatory filings via the internet, and company websites to evaluate the level of corporate transparency. Since 2003, the IDTRS has identified 114 disclosure items as evaluation criteria, grouped into five categories: compliance with the mandatory disclosure requirements, timeliness of reporting, disclosure in annual reports, disclosure of financial forecasts, and corporate website disclosuresFootnote 3 . IDTRS then ranks listed firms according to five grades, A+, A, B, C, and C−, beginning with 2005, the 3rd evaluation year. In this study, scores from 5 to 1 were assigned to the measure of information disclosure quality corresponding to the companies’ rankings from grade A+ to C−.
Independent and moderating variables
The study employed two variables to measure board independence. Independent directors were calculated as the proportion of independent directors to total number of directors on the board. CEO duality was a binary variable, coded as 1 if a CEO was also chairperson and as 0 otherwise. In addition, we used two proxy variables to assess the intensity of board activity. Board meeting was measured as the number of board meetings held during the financial year. Board attendance was measured as the average rate of directors attending meetings during the financial year. The moderating variable of family ownership was measured as the percentage of equity ownership held by the family, including family personal shareholdings, family unlisted company shareholdings, family foundation shareholdings, and family-listed company shareholdings (Zahra, Reference Zahra2003; Villalonga & Amit, Reference Villalonga and Amit2006).
Control variables
We included five control variables that are likely to affect firms’ information disclosure: firm size, return on assets, sales growth, board size, and industry. First, large firms have a greater incentive to adhere to established disclosure practices because they are more likely to be scrutinized by financial analysts and other outside stakeholders (Lang & Lundholm, Reference Lang and Lundholm1996; Ho & Wong, Reference Ho and Wong2001). The logarithm of the firm’s assets in a given year was used as an independent control for firm size. Second, prior researchers have shown that better performing companies tend to disclose more information in their annual reports because those firms wish to send a signal to the market regarding their superior performance (Haniffa & Cooke, Reference Haniffa and Cooke2002; Holland, Reference Holland2005). We measured two variables to control for firm performance, return on sales (ROA) and the year-over-year percentage change in sales (Sales Growth). Third, we controlled for the effect of board size. A larger board may be in a better position to monitor management and decrease the probability of information asymmetry (Chen & Jaggi, Reference Chen and Jaggi2000; Zahra, Neubaum, & Huse, Reference Zahra, Neubaum and Huse2000). Finally, the industry type was controlled by using dummy variables, which were broadly classified as heavy (Heavy industry), light (Light industry), hi-tech (High Tech industry), and others; a firm received a score of 1 if it belonged to any of these industries and 0 if it did not. The industry type of others served as the reference category in the analysis.
Analytical approach
We tested the hypotheses presented in this paper by using repeated observations of the same set of cross-sectional units (i.e., panel data) (Greene, Reference Greene2000). In our sample, the Hausman test indicated that the estimation results of the fixed effects and random-effects model were consistent, and the individual effects were not correlated with the other variables in the model. Therefore, we employed the more efficient random effects generalized least squares estimation technique.
The variance inflation factor was derived to check whether multicollinearity could be a potential problem. As long as variance inflation factor is <10, multicollinearity is not a concern (Hair, Anderson, & Tatham, Reference Hair, Anderson, Tatham and Black1998). The variance inflation factor value varied from 1.01 to 3.14, well below the threshold suggested by scholars. We took an additional action to avoid multicollinearity problems by centering the variables used to test the predicted interactions (Aiken & West, Reference Aiken and West1991).
Results
The means, standard deviations, and bivariate correlations for all the variables are presented in Table 1. The table shows that the mean proportion of independent directors of Taiwanese-listed firms was 8%, which is much lower than at least one-fifth of board members recommended by Corporate Governance Best-Practice Principles for Taiwan Stock Exchange Corporation/Gre Tai Securities Market Listed Companies. CEO duality was 0.36 which indicates that for every 100 CEOs, 36 of them served as board chairpersons. This rate is close to that of dual board leadership structure in several countries (e.g., United States, Ballinger & Marcel, Reference Ballinger and Marcel2010; European countries, China, Li & Tang, Reference Li and Tang2010; Muslu, Reference Muslu2010). The average percentage of family ownership was 27%.
Descriptive statistic and correlationsFootnote a
Note. For bivariate correlations above 0.041 are significant at p<.05, two-tailed tests.
a Number of observations=2,580 (516 firms multiplied by 5 years).
Table 2 shows the results of the random-effect generalized least squares regression analyses. The base model, Model 1, contains all of the control variables. The second model included both the main and moderating variables. In the third model, the interaction variables are entered into the regression. The Wald χ2 statistic indicates the overall significance of each model, and the second χ2 change statistic provides a test for the statistical significance of the added variables in a particular model. The χ2 statistic of Model 2 for change, compared with the control model (i.e., Model 1), is significant (Δχ2=23.85, p<.001). The χ2 statistic of Model 3 for change compared with the main effect model (i.e., Model 2) is significant (Δχ2=15.01, p<.05), indicating a significant change in the amount of variance that is explained by the interaction effect of family ownership.
Results of generalized least squares random-effects regression analyses for firm information disclosure qualitya
Note. Regression parameter appears with the standard error (in parentheses) and nonstandardized coefficient.
a Number of observations=2,580 (516 firms multiplied by 5 years).
b The industry type of others serves as reference category.
c Relative to Model 1.
d Relative to Model 2.
*p<.1, **p<.05, ***p<.01.
Model 1 shows that the three control variables with significant effects on a firm’s disclosure quality are firm size (p<.001), board size (p<.1), and light industry (p<.1). The regression coefficients for the main effects are more challenging to interpret given the addition of the interaction relations in the model. If the coefficients on interaction terms are significant, the main effects should be interpreted as conditional (Edwards, Reference Edwards2008). Model 3 shows that when family ownership is 0, the proportion of independent directors has no significant effect on disclosure quality, which result did not support our Hypothesis 1. However, when family ownership is 0, CEO duality has a negative effect on a firm’s disclosure quality (β=−0.095, p<.1), thus supporting Hypothesis 2. Hypothesis 3 and Hypothesis 4 predicted that the more frequent board meetings and higher attendance rate of board directors, the more likely that firm will have a better disclosure quality. As shown in Model 3, the first-order coefficients for board meeting and board attendance are significant and positive (β=0.007, p<.1; β=0.436, p<.01). Thus, the main effect results provide general support for both Hypotheses 3 and 4.
Model 3 also shows that the moderating effects of family ownership on the relationship between board independence and a firm’s disclosure quality (CEO duality in Hypothesis 5a and independent directors in Hypothesis 5b) were supported (β=0.044, p<.1; β=0.041, p<.1). But the interaction effects of family ownership on the relationship between the intensity of board activity and its disclosure quality (Hypotheses 5c and 5d) were not statistically significant. We further conducted a slope test in accordance with past research (Baron & Kenny, Reference Baron and Kenny1986; Aiken & West, Reference Aiken and West1991) to examine the interaction of board independence and family ownership as predicted by Hypotheses 5a and 5b. As illustrated in Figure 1, the negative relationship between CEO duality and disclosure quality is weaker in firms with higher family ownership (t=−1.01, ns) than in firms with lower family ownership (t=−1.65, p<.1). A smoother negative slope clearly shows that family ownership, through interaction with CEO duality, has a positive effect on a firm’s disclosure quality. Similarly, Figure 2 demonstrates that the proportion of independent directors is positively associated with disclosure quality of firms when family ownership is high (t=2.17, p<.05), whereas that relationship failed to reach significance (t=1.23, ns) when family ownership is low.
Moderating effects of family ownership on the relationship between CEO duality and disclosure quality
Moderating effects of family ownership on the relationship between independent directors and disclosure quality
Discussion and Implications
In the last two decades, corporate governance issues have become important not only in academic literature, but also in public policy debates. With firms in Greater China receiving more attention from researchers, this article and others on governance in that region (e.g., Lien, Piesse, Strange, & Filatotchev, Reference Lien, Piesse, Strange and Filatotchev2005; Liu et al., Reference Liu, Wang, Zhao and Ahlstrom2013) contribute to the literature on corporate governance and family firms to increase our understanding of the effects of governance mechanisms in economies with much different institutional environments and cultural traditions than those presented in the more developed economies of the world. This paper uses Taiwanese-listed firms as the subject of study to examine how board independence and board activity affect a firm’s information disclosure quality. In addition, we incorporate the perspectives of institutional conformity and family SEW to examine the moderating influence of family ownership, an important factor in a firm’s information disclosure.
Using longitudinal data (2006–2010) from firms listed on the Taiwan Stock Exchange and the Over-the-Counter Market, we found that board independence (i.e., the percentage of independent directors) and the intensity of board activity (i.e., board meeting frequency and director attendance) have a significant relationship with a firm’s disclosure quality. Contrary to the meta-analysis findings of Garcia-Meca and Sanchez-Ballesta (Reference Garcia-Meca and Sanchez-Ballesta2010) that the positive association between board independence and voluntary disclosure does not occur in Asian countries, the results of this study are generally consistent with the hypotheses put forth in this article: that better governance practices would lead to more transparent disclosures.
Additionally, we found that the relationship between board independence and disclosure will increase as family ownership increases, and this result provides support for both the institutional perspective and the arguments of Miller, Breton-Miller, and Lester (Reference Miller, Breton-Miller and Lester2013) and Berrone et al. (Reference Berrone, Cruz, Gómez-Mejía and Larraza-Kintana2010) that family-owned firms in pursuing SEW are more likely to conform to environmental regulation in order to gain legitimacy. However, we did not find that family ownership fosters a positive relationship between the intensity of board activity and disclosure quality. This may be interpreted that families exercise their power and influence not through formal board meetings but by other informal mechanisms. For example, members of family-owned firms may have social dinners with board members before each formal board meeting to agree on the board agenda.
We find it interesting that our results regarding the impact of family ownership appear to contradict the thinking of scholars in the finance and accounting fields that firms with concentrated insider equity tend to disclose less information (Ajinkya, Bhojraj, & Sengupta, Reference Ajinkya, Bhojraj and Sengupta2005; Karamanou & Vafeas, Reference Karamanou and Vafeas2005). Given the power of family owners stemming from their significant ownership, one might expect that they would have an incentive to avoid close monitoring by outside shareholders by disclosing less information; but this is not what we found. It appears that the behavior preferences of members in a family-owned firm are socially determined, and their surrounding institutions will impose normative and political influences and constraints (Davis, Reference Davis2005; Scott, Reference Sharma and Manikutty2013). Taiwan is a distinctive economy entity within the Greater China societies. Current research regarding Taiwan’s family-owned businesses has found the importance of alternative family control mechanisms in the embedded institutional environments (Chung & Chan, Reference Chung and Chan2012). Furthermore, studies also indicate the possible future challenges for family businesses operating in this area.
Our findings also extend the research on SEW. While many studies theorize that SEW is most strongly associated with retention of control and trasngenerational issues, our research supports the view that family members are more concerned with reputation than investors and managers of nonfamily owned firms (Deephouse & Jaskiewicz, Reference Deephouse and Jaskiewicz2013). As a result family ownership encourages the disclosure of information beyond what is mandatorily required in order to enhance the legitimacy and favorable view of the firm by outside stakeholders, which in turn increases family members’ self-esteem and pride in their family business. Further, as family members have substantial control over what is disclosed by the firm, disclosure of poor earnings or other bad news may be deemed necessary in order to avoid potential lawsuits that could negatively affect overall family wealth and job security (Chen, Chen, & Cheng, Reference Chen, Chen and Cheng2008).
This study has two main practical implications. First, for regulators, our findings suggest that regulations fostering the enhancement of corporate governance mechanisms lead to reducing information asymmetry and promoting transparency and the quality of corporate disclosure. Policy makers, who endeavor to improve corporate governance and information disclosure within the capital markets of Taiwan and similar institutional environments, may use these results to evaluate the present regulatory requirements and, possibly, to increase the enforcement for listed corporations to comply with the prescribed practices of corporate governance.
Second, for family owners, members of Chinese family firms typically regard family businesses as their private assets and, therefore, they believe that they should be able to keep operational information confidential. However, when family-owned firms choose to having their stock publicly traded on an exchange, they become under great pressure to respond to institutional expectations. Listed family firms are unable to avoid the public scrutiny and information demands from external shareholders. Failure to adhere to the mandates of the institutional environment can lead not only to the withdrawal of outside investors but also to social and institutional sanctions resulting in a loss of family reputation. Facing the dilemma of meeting the conflicting demand for legitimacy and need for full family control, family owners may pursue a compromise strategy, which is achieved by partial compliance with institutional expectations (Oliver, Reference Oliver1991), to achieve a balance between the simultaneous pressures from internal family members and the external institutional environment. For instance, family owners may appoint competent independent directors as well as family members to form the board, which allows firms to maintain family control and involvement while at the same time meeting institutional demands. Family members sitting on the board can retain family control and interest; conversely, competent independent directors can bring in objectivity and innovative perspectives to board processes as well as gain institutional legitimacy. This board configuration can potentially facilitate a family-owned firm to provide a better quality of information disclosure to outside investors, which will result in receiving the recognition of institutional stakeholders, thus preserving family SEW. In addition, the higher quality of disclosure subsequently may lead to a lower cost of capital and higher firm valuation (Healy & Palepu, Reference Healy and Palepu2001; Francis, Khurana, & Pereira, Reference Francis, Khurana and Pereira2005).
Limitations and Future Research
The limitations of this study provide opportunities for future research. First, although our use of Taiwanese-listed firms enabled us to clarify the relationships among governance practice, family ownership, and information disclosure, our sample may have limited generalization to other contexts. For example, our findings might be more representative of newly industrialized economies where governance standards are still taking shape, where firms rely more heavily on informal governance structures (Young, Ahlstrom, Bruton, & Chan, Reference Young, Ahlstrom, Bruton and Chan2001), and may have more incentives to distinguish themselves based on their adopted governance practices. Even among Asian companies, the impact family control may produce mixed results in terms of performance depending on the level of governance required by legal and regulatory institutions (Jiang & Peng, Reference Jiang and Peng2011). Hence, the evidence provided here might not necessarily generalize to Anglo-American or even to other Asian countries. In addition, our study may not apply to private family firms that are less monitored by public stakeholders and which are, therefore, under less pressure to conform to institutional requirements. As family-owned firms are prevalent throughout Asia and Eastern Europe, future studies can be extended to companies outside of Taiwan to compare the results with those reported here.
Second, due to unavailable data, our use of the IDTRS rankings results is another limitation of this paper, which prevents us from considering the impact of governance practices on each category of information disclosure included in the IDTRS ranking. A further study collecting systematic data on mandatory and voluntary information disclosure to examine the disclosure quality of firms would be worthwhile. Third, the evaluation process conducted by IDTRS focuses only on the existence of each disclosure item, not on the accuracy of the information; thus, the possibility of misstatement cannot be entirely eliminated. Fourth, this study has concentrated on macro indicators of board structure and activity, but we have very little understanding on the process of deciding the content and extent of information disclosures by family firms. Future research might find a way to examine these processes.
Finally, subsequent researchers might extend these findings to examine economic consequences of conformity by family firms. We believe it would be a promising avenue for future research to investigate the consequences of conforming to institutional requirements in family firms by assessing whether providing higher disclosure quality effectively enhances their competitive advantage and performance in the competitive markets and the overall environment.
Acknowledgements
The authors thank JMO Editors and two anonymous reviewers for their helpful comments. This research is supported by grants from Ministry of Science and Technology (NSC 101-2410-H-224 -034 -MY3).
