Form versus substance: The effect of ownership structure and corporate governance on firm value in Thailand

Journal of Banking & Finance 36 (2012) 1722–1743

Contents lists available at SciVerse ScienceDirect

Journal of Banking & Finance journal homepage: www.elsevier.com/locate/jbf

Form versus substance: The effect of ownership structure and corporate governance on firm value in Thailand J. Thomas Connelly a, Piman Limpaphayom b,⇑, Nandu J. Nagarajan c a

Department of Banking and Finance, Faculty of Commerce and Accountancy, Phyathai Road, Chulalongkorn University, Bangkok 10330, Thailand Portland State University, Sasin Graduate Institute of Business Administration of Chulalongkorn University, USA c Joseph M. Katz Graduate School of Business and College of Business Administration, University of Pittsburgh, Mervis Hall, Pittsburgh, PA 15260, USA b

a r t i c l e

i n f o

Article history: Received 11 March 2010 Accepted 25 January 2012 Available online 4 February 2012 JEL classification: G32 G34 Keywords: Corporate governance Family ownership Firm value Thailand

a b s t r a c t We examine the relation between the quality of corporate governance practices and firm value for Thai firms, which often have complex ownership structures. We develop a comprehensive measure of corporate governance and show that, in contrast to conventional measures of corporate governance, our measurement, on average, is positively associated with Tobin’s q. Furthermore, we find that q values are lower for firms that exhibit deviations between cash flow rights and voting rights. We also find that the value benefits of complying with ‘‘good’’ corporate governance practices are nullified in the presence of pyramidal ownership structures, raising doubts on the effectiveness of governance measures when ownership structures are not transparent. We conclude that family control of firms through pyramidal ownership structures can allow firms to seemingly comply with preferred governance practices but also use the control to their advantage. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction In July 1997, the Bank of Thailand discontinued the fixed exchange rate regime which had been in place for decades. Thailand, an emerging nation, with a promising economic future, tumbled into the worst financial crisis in its history. In the process, Thailand also dragged many neighboring economies along with it into a region-wide economic downturn, the likes of which no country in the region had ever experienced before. Following this crisis, and its resolution, corporate governance has received considerable attention from regulators and practitioners in all the Asia–Pacific countries. The main impetus for this heightened attention stems from evidence emerging from the financial crisis that the aggressive financing practices and poor investment decisions, associated with the financial downturn, were a result of poor corporate governance practices among large public corporations and financial institutions in the afflicted economies. Consequently, the central governments of most Asia–Pacific nations, along with international organizations such as the OECD, implemented corporate governance reforms ⇑ Corresponding author. Tel.: +1 503 725 9991; fax: +1 503 725 5850. E-mail addresses: [email protected] (J.T. Connelly), [email protected] (P. Limpaphayom), [email protected] (N.J. Nagarajan). 0378-4266/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.jbankfin.2012.01.017

throughout the region.1 The effectiveness of these governance reforms is an empirical question because business environments in this region are heterogeneous and Asian companies have many unique features, the most notable among them being the concentrated ownership and direct and indirect control exercised by the firms’ founding families. In this paper, we provide empirical evidence on the relation between control, ownership structure and firm value for all industrial companies that were publicly traded on the Thai stock exchange in 2005. We find that Thai family companies developed pyramidal ownership structures after the financial crisis of 1997, probably in response to reductions in their ownership holdings. In particular, we find that the governance measures mandated in Thailand, and subsequently adopted by Thai family firms, are not as effective in mitigating agency conflicts in this new opaque environment as they have been shown to be in the US. We also develop a corporate governance index that we show is a more effective measure of

1 There is recent empirical evidence of an association between adoption of internationally accepted corporate governance practices and firm valuation in these Asia–Pacific economies. See, for example Cheung et al. (2010) who examine large Chinese firms, Black et al. (2006a, 2009) for example Korean firms, and Cheung et al. (2007, 2011) for Hong Kong firms.

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

compliance with good governance practices in the Thai business context than other, conventional measures of effective governance. However, even this governance index appears to be associated with value only when companies do not have a pyramidal ownership structure, suggesting that Thai families are able to manipulate governance measures when they have high voting control over their firms. Recent anecdotal and research evidence document that family firms in the US outperform non-family firms.2 Founding families have considerable wealth invested in their firms and, thus, have an incentive to manage the firm in ways that improve value. On the other hand, families and founders may use their greater control over decision rights to expropriate wealth from minority shareholders, resulting in a decrease in value.3 Family firms in Thailand, however, operate in a very different environment from that in the US. Recent empirical research (La Porta et al., 2000b, 2002) provides evidence that companies with controlling shareholders have lower valuations in civil law countries, like Thailand, where minority shareholders are less well protected from expropriation by a controlling shareholder, compared to valuations in common law countries, like the US. Mitton (2002) argues that in an environment where legal protection for outside shareholders may be insufficient, the firms themselves can pre-commit to not expropriating wealth from minority shareholders by implementing appropriate corporate governance measures. For instance, such family firms could increase the number of independent directors on the board. The foregoing argument suggests that implementing appropriate governance measures may be particularly important in East Asian countries, where families commonly tend to own controlling interests in firms and legal protections for minority shareholder rights are limited. The issues that emerge are twofold. First, what measures of governance are appropriate benchmarks in an environment where implementation of good governance practices is hard to measure and monitor? Second, is compliance with prescribed measures of governance effectiveness sufficient to mitigate agency problems between majority and minority shareholders in these East Asian family firms or do pyramidal ownership structures and extra-contractual arrangements between family members and directors allow subversion and manipulation of these governance measures, rendering them ineffective proxies for value creation?4 In other words, is the form of governance measures adopted by these firms consistent with the substance of what such measures are intended to accomplish? Previous studies on governance arrangements in Thai family firms have generally focused on the period before the Asian financial crisis in 1997 (Bertrand et al., 2008; Wiwattanakantang, 2001). Before 1997, founding families held significant proportions of the outstanding shares and had full control over their businesses. Consequently, these families may have had no need to employ

2 Anderson and Reeb (2003) document superior performance of family-owned firms among S&P 500 firms, across seven years (1992–1999). The improved performance is further increased when a family member holds the CEO position rather than an outside manager. An article in Business Week (November 10, 2003, p. 100) reports that one-third of the S&P 500 firms have founding families involved in management and these are the best performers. Villalonga and Amit (2006) and Perez-Gonzalez (2006) find that founder-controlled firms are associated with higher values which decline when these firms are managed by heirs. 3 Demsetz and Lehn (1985) argue that founder or family control may be exercised because of ‘‘amenity potential’’, the non-pecuniary benefits that family members gain from control. However, because this leads to entrenchment, family control may also be associated with weaker management than that provided by professional managers. Johnson et al. (1985) find that positive unexpected returns are associated with the sudden demise of founder-CEOs and Morck et al. (1988) find that older firms with founders present among the management have reduced q values. 4 For instance, Hwang and Kim (2009) provide evidence that social ties between managers and directors can weaken managerial pay-performance and turnoverperformance sensitivity.

1723

complicated ownership structures, such as pyramids. In corroboration, Claessens et al. (2000) document that Thai companies rarely employed a pyramidal ownership structure prior to 1997. However, recent changes in the characteristics of Thai family firms have affected the nature and extent of corporate family ownership and control. Specifically, many firms in Thailand experienced financial difficulties and some went bankrupt following the Asian financial crisis (Zhuang et al., 2000). As a result, some of these firms, including family firms, had to raise additional capital and restructure themselves after 1997, which, in turn, led to reductions in family ownership.5 More importantly, many publicly traded family firms have also implemented pyramidal ownership structures, probably because they wished to counter the dilution in control arising from reductions in their shareholdings. Based on these considerations, we expect that family firms in Thailand are no longer the homogeneous group of firms with similar characteristics that they were prior to 1997. In particular, the nature and extent of agency problems, and corresponding value consequences, can vary across family firms. Because of these new institutional settings, we expect that our post-1997 analysis of these family firms can add additional insights and make a significant contribution to the literature. In our analysis, we classify industrial companies that were publicly traded on the Thai stock exchange in 2005 into four groups, based on the extent of family ownership and the presence of pyramidal ownership structures.6 Of the total sample of 216 firms, we find that firms with high family ownership are associated with lower values of Tobin’s q. In particular, these high family ownership firms have an average q value that is lower than the mean q for low family ownership firms.7 This difference is not only statistically significant, but also economically significant. Past research conducted on US firms (e.g., Yermack, 1996) has revealed a positive relation between conventional governance variables, such as board size and firm value. However, we find no such relation between conventional governance variables and q for Thai firms. We conjecture that this lack of association between broad governance measures and q for our sample firms arises because Thai family firms only maintain the external trappings of good governance practices, such as having several directors labeled as independent, while 5 As part of restructuring efforts, many families were forced to sell significant stakes in their firms. Others sought additional capital through strategic partnerships, often with foreign companies. The shareholdings of two of the largest commercial banks in Thailand were drastically restructured as the founding families gave up significant portions of their ownership stakes. Non-financial companies were also hard-hit. For example, at SHIN, a large stake in the nation’s largest telecommunication company, owned by the former Prime Minister of Thailand, was sold to a consortium of Singaporean investors. BIG C, a large family-owned discount retailer, forged a business alliance with a foreign retailer by issuing shares to the new partner. NTS, a family-owned steel maker, merged its steel business into a new company jointly owned by another conglomerate. The founder of QH, a large real estate development company, sold a stake to the Government of Singapore Investment Corporation. 6 Bertrand et al. (2008) study a sample of public and private firms using 1996 data, drawn from 93 Thai family groups. They show that family structure is important to value creation in Thai companies. Firms with more male heirs end up with sons having greater control and such firms are associated with lower levels of firm performance, especially if the founder is dead. Families with more sons show a greater gap between control and ownership rights. Our study differs in important ways from theirs because we focus only on publicly traded firms drawn from a period after the financial crisis of 1997 when family firms’ ownership structures were drastically different. The study by Bertrand et al. (2008) consists of 528 firms, of which only 94 are publicly traded companies. More importantly, our study is able to assess the relative importance of control and management on agency costs while examining the relation between market value and a quantified and detailed governance index. 7 As anecdotal evidence, Studwell (2007, p. 24) reports, ‘‘Despite now bullish stock markets in the region, the billionaires-with their lousy corporate governance and manipulation of local banks to provide cheap and easy alternative sources of creditalso have contributed to the worst long-term emerging-market-equity performance in the world. From 1993-when the first significant international portfolio investments came into Southeast Asian bourses-to the end of 2006, total dollar returns with dividends reinvested in Thailand and the Philippines were actually negative’’.

1724

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

subverting the substance of such practices through the influence of family ownership and pyramidal control arrangements.8 To test this conjecture, we design an in-depth assessment of publicly reported details of governance practices and disclosure, which we term the Corporate Governance Index (CGI).9 In regression analyses using all firms, we show that, unlike conventional governance variables, the CGI is significantly associated with q. Further, in regression analyses with subgroups, we show that the positive association between CGI and q is driven by family firms without pyramidal ownership structures. Our results remain robust after controlling for past performance and the potentially endogenous relationship between CGI and q through the use of a two-stage least-squares regression approach. Overall, our study makes several contributions to the literature. We document that the nature of family ownership in Thai family firms changed after the 1997 crisis, facilitating more opaque ways for families to establish control. Furthermore, we construct and use a detailed governance index which allows us to establish the relation among stock ownership, governance and value for Thai firms. We also provide evidence that, contrary to the findings for US companies, conventional governance variables such as board size and board independence are not associated with value for Thai family firms. On the other hand, the Corporate Governance Index (CGI) we construct, to assess the effectiveness of the implementation of governance practices, is significantly associated with value for family firms without pyramidal ownership structures. That is, our empirical results show that (1) governance measures that effectively capture the relation with value for Thai family firms have to be more detailed than conventional measures such as board independence and (2) the relation between these more detailed measures of good corporate governance and firm value is positive and statistically significant only for family firms without a pyramidal ownership structure, suggesting that such complex ownership structures can subvert and render governance measures ineffective. In support of this argument, we also find that the use of a pyramidal structure negatively affects firm value after controlling for relevant factors. Our results have implications for governance regulation in Thailand, namely that the adoption of so called ‘‘good’’ corporate governance practices alone is not a guarantee of firm performance. We organize the remainder of the paper as follows. Section 2 discusses the theory and institutional background for our study

8 According to Stock Exchange of Thailand regulations, Thai companies are required to have at least five members on the board of directors, three of whom must be independent directors. The CEO or top operating executive is permitted to hold the board chairmanship (Sersansia and Nimmansomboon, 1996). However, Thai boards often have directors among their members, who are connected, with an affiliation to a related organization (Khantavit et al., 2004; Nam and Nam, 2004). For example, a director may be an employee of a related company or a creditor financial institution. These affiliations mean true director independence can be difficult to achieve. Contributing to this mix is the fact that personal relationships, often spanning decades, play a strong role in Thai culture. Directors are quite likely to have had previous business, social, or even high school connections with the top executive. Directors may also feel beholden to the executive who appointed them to the board and thus may be hesitant about objecting openly to his policies, which would violate cultural norms (Nam and Nam, 2004). 9 The composition and construction of the Corporate Governance Index (CGI) are described on pages 6–9. The full set of measurements is shown in Appendix. By corporate governance activities in closely held firms (where Type II agency problems exist, that is agency conflicts between majority shareholders and outside/minority shareholders), we mean actions and disclosures by the firm that can affect the welfare of minority shareholders. Such actions could include more transparency regarding board activities, such as minutes of board meetings, public disclosure of compensation arrangements for directors, shareholders being allowed to ask questions during annual general meetings, disclosure of crossholdings and pyramidal holdings, and prevention of insider trading by controlling shareholders. Similarly governance activities in widely held firms, in which Type I agency problems, that is agency conflicts between managers and shareholders, are found, would be compensation arrangements for managers that tie their interests to the shareholders, monitoring of managerial actions by the directors, and other practices.

of governance arrangements. The data and empirical methods are presented in Section 3. Section 4 provides the empirical results and Section 5 concludes the paper. 2. Theory and institutional background The Stock Exchange of Thailand (SET) can be characterized as a retail-driven market, with domestic retail investors conducting the lion’s share of trading. Institutional investors make up a smaller, though growing segment. Overall, Thai companies are mostly owned by insiders while trading activities are mostly initiated by small minority investors. Moreover, legal protection for investors is relatively weak and ownership structures are highly concentrated. Taken together, the Thai market provides an interesting setting in which to examine the relation among family ownership, corporate governance and firm performance. 2.1. Family ownership and firm valuation Financial economists have long recognized that ownership structure has an important influence on value creation and firm performance (Shleifer and Vishny, 1997). For example, Morck et al. (1988) find that, at low levels, managerial ownership is value-enhancing through the alignment of interests between insiders and outside shareholders, while at high levels, managerial ownership becomes value destroying through the entrenchment effect. This type of agency conflict is referred to as Type I (between managers and shareholders). For family firms, the agency conflicts between majority shareholders and outside/minority shareholders (referred to as Type II agency conflicts) are more prevalent. Recent empirical evidence for US firms shows that family firms perform better than non-family firms (Anderson and Reeb, 2003). US family firm performance is even stronger when founders manage the firm (Villalonga and Amit, 2006; Perez-Gonzalez, 2006). Demsetz and Lehn (1985) argue that, because of their concentrated and undiversified holdings, families have more incentives to minimize agency conflicts and maximize firm value. There is also evidence that family ownership leads to long investment horizons and efficient investment decisions (Stein, 1989; James, 1999). However, the relation between family ownership and firm performance can be different in a developing economy with relatively weak legal rights and weak investor protection. This is because, in such an environment and by virtue of their larger ownership stakes, families have more opportunities and the power to take actions that benefit themselves at the expense of minority shareholders (Fama and Jensen, 1983). An examination of a sample of companies in East Asia reveals that the majority of top managers in these companies come from controlling families (Claessens et al., 2000). Lins (2003) finds that, among East Asian firms, firm value is lower when management control is excessive. Following this evidence, we expect that Type II agency conflicts for many Thai companies could be quite severe. 2.2. Ownership structure and firm valuation Recently, corporate finance theorists and corporate governance researchers have turned their collective attention to the relation between ownership structure and corporate governance in East Asian firms. La Porta et al. (2006) argue that securities laws and enforcement are critical to financial market development. In fact, high ownership concentration by insiders is a response to the lack of legal protection for shareholders (La Porta et al., 1998; hereafter LLSV). Several prior studies have classified Thailand as a country with weak legal enforcement and shareholder protection. For example, LLSV (1998) rate Thailand fairly low in terms of shareholder

1725

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

1996

Anan Asawaphokin 15.77%

Land and House PLC 6.99%

2005

Chai Srivikorn 10.11%

Financial Institutions 15.02%

QH PLC

Land and House PLC 23.14%

Asawaphokin Family 7.17%

Private Company 8.86%

Government Government of ofSingapore Singapore 13.23% 13.23%

QH PLC

Fig. 1. An illustration of the ownership structure changes experienced by publicly traded Thai Family Companies After 1997. The figure shows the transformation of the ownership structure of QH, a large family-owned real estate company, between 1996 and 2005. The thin arrows indicate the direct shareholdings in QH by the various large shareholders. The wide arrows show the indirect control by the family over other entities.

rights and the rule of law. The same authors (LLSV, 2000a) also classify Thailand as a low-investor protection country in their 33-country survey of dividend policies around the world. Given that low protection for minority shareholders is associated with higher ownership concentrations, it is not surprising that LLSV (2000b) and Claessens et al. (2000) find that East Asian firms have highly concentrated ownership structures, which, in turn, allows insiders to exercise effective control over their firms. Although Thai companies have very high ownership concentrations, Claessens et al. (2000) provide evidence that the ratio between cash flow rights and control rights for Thai firms is close to one. However, the evidence in Claessens et al. (2000) describes Thai ownership structures prior to the Asian financial crisis of 1997. In other words, Thai firms seldom used pyramidal structures to enhance control before 1997 because family ownership levels were high and there was no need to do so. Subsequent to the crisis, many Thai family firms had to dilute their ownership stakes in order to raise additional outside investment. To illustrate this phenomenon, we examine the transformation of a large real estate company, Quality House, PLC (QH). QH was founded by Anan Asawaphokin as a construction company in 1983. In 1990, QH began to engage in both residential and commercial real estate development. In 1996, QH would not have been considered to have a pyramidal ownership structure (see Fig. 1). The founder, Anan Asawaphokin, owned 15.77% directly plus 6.99% through Land and House PLC, another large real estate company controlled by the Asawaphokin family. Another individual from another well-known family, Chai Srivikorn, owned 10.11% of the outstanding shares. Unaffiliated foreign financial institutions owned a total of 15.02%. Consistent with Claessens et al. (2000), there is little evidence of any deviation between cash flow rights and voting rights for QH and many other Thai companies during this pre-1997 period. In 1997, QH experienced financial and operating problems. After the Asian financial crisis, the Government of Singapore Investment Corporation became a major shareholder in 2001 with a 20% holding. By 2005, the ownership structure of QH had become very complicated. The direct shareholdings in QH by members of the Asawaphokin family had declined to 7.17%, and the Government of Singapore Investment Corporation held 13.23% of the outstanding shares. However, Anant Asawaphokin’s shareholdings in QH, via control of Land and House, increased from 6.99% in 1996 to 23.14%. Two other companies, private firms affiliated with the Asawaphokin family, held a combined 8.86% ownership stake in QH. The pyramidal structure, now apparent, pushed the ratio between cash flow rights and voting rights lower than one. The

transition of the control structure of QH is graphically presented in Fig. 1. Overall, it appears that in order to maintain control, families and founders employ more opaque (pyramidal) ownership structures that allow the owners to exert significant control even with lower direct ownership holdings. One potential consequence of excessively high control is that controlling shareholders are able to expropriate wealth from minority shareholders. Following the arguments above, we predict that, for our sample of Thai family firms, the use of more control through a pyramidal ownership structure should be detrimental to firm value. In other words, the presence of deviations between control rights and cash flow rights should be negatively associated with q. 2.3. Corporate governance and firm performance Recently, corporate governance researchers have begun to use composite indexes to assess governance practices, recognizing that corporate governance mechanisms may serve as complements or as substitutes for one another. Four studies are notable in this regard. Gompers et al. (2003), Bebchuk et al. (2009), and Brown and Caylor (2006) create separate indexes for US firms while Drobetz et al. (2003) do likewise for German companies. These authors demonstrate the relation between good corporate governance and firm value.10 There are also a number of studies examining the governance to performance link for firms in emerging markets. Generally, authors of these studies also find that better-governed firms are associated with better performance.11 Mitton (2002) suggests that in an environment where legal protection for outside shareholders may be insufficient, firms themselves can preclude expropriation of minority shareholders’ resources through the use of appropriate corporate governance measures. Therefore, adoption of appropriate or effective corporate governance practices is particularly important among closely held firms in East Asian countries. A further implication of this argument 10 These types of governance indexes may not be relevant for Thai firms because they are largely structured to assess the extent of takeover defenses. Hostile takeovers are infrequent in Thailand and in emerging markets throughout the Asia–Pacific region. 11 Using an internationally accepted benchmark (OECD, 2004), Cheung et al. (2010) document a positive relation between market valuation and corporate governance practices among large Chinese firms. Other studies on the relation between a portfolio of governance measures or index and value include Black et al. (2006a, 2009) for Korean firms; Doidge et al. (2007), Mitton (2004), Durnev and Kim (2005), and Klapper and Love (2003) which use the index created by Credit Lyonnaise Securities Asia (CLSA, 2002); Cheung et al. (2007, 2011) for Hong Kong firms; and Black et al. (2006b) for Russian firms.

1726

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

is that if appropriate governance measures, such as internationally accepted corporate governance practices, were to be adopted by Thai family firms, they could potentially mitigate any adverse impact of family ownership. In addition to the changes in the ownership structures of Thai family firms after 1997, the institutional environment in which these firms operate has also changed drastically. In particular, many corporate governance reforms have been implemented in the post-crisis years through the concerted efforts of many organizations (Limpaphayom and Connelly, 2004). For example, the Stock Exchange of Thailand (SET) required all firms to have audit committees comprised entirely of independent directors by 1999. The SET also established several other guidelines for improving the quality of corporate governance practices and for conducting shareholders’ meetings. In addition, new and updated rules from the SET, new and revised laws, and increased regulatory oversight have been at the forefront of the push for improved corporate governance. For example, in 1999, the Thai government passed laws requiring disclosure of related party transactions, asset divestitures, and disclosure of accounting practices that deviated from generally accepted principles. In 2002, the government passed a new law which shortened the financial statement reporting time from 60 to 45 days after the end of a fiscal year. In addition to regulatory reforms, other organizations like the Thai Institute of Directors Association (Thai IOD), the Thai Investors Association (TIA), and professional associations for accountants, auditors, and internal auditors have pushed forward initiatives to improve the quality of corporate governance practices. Consequently, Thai companies operate in an environment that is very different from the time before the Asian financial crisis. Although conventional wisdom would suggest that firms conforming to mandated governance requirements (e.g., board independence) are less likely to experience impairments in value, there are often extra-contractual relationships between the top managers and directors that can complicate the situation (Romano, 2005). The number and extent of these relationships in Thai companies can be considerable, stemming from long associations and personal or family contacts. These extra-contractual relationships can result in boards, with ostensibly independent directors, being unable to effectively monitor and control management. One implication of the foregoing observation is that traditional measures of board effectiveness, such as board size or board independence, while effective as measures of corporate governance in the US, may be inadequate to capture the actual quality of corporate governance practices among Thai family firms. Therefore, we predict that these conventional proxies for effective corporate governance should exhibit no significant relation with firm value for Thai firms. In addition, we note that other governance measures commonly used to evaluate the level of entrenchment of insiders, such as the GIM index (Gompers et al., 2003) or the E-Index (Bebchuk et al., 2009) are not applicable to Thai family firms because takeovers are not common in Thailand and restrictions in shareholder rights such as staggered boards or supermajority amendments to the charter are not permitted (Limpaphayom and Connelly, 2004). However, we expect that more comprehensive and detailed measures of the quality of corporate governance practices, that go beyond superficial conformity to prescribed standards, by capturing how governance practices are actually implemented, should exhibit a positive relation with value. Furthermore, we predict that the use of a complex ownership structure, such as a pyramidal structure, can moderate the relation between the quality of corporate governance practices and firm value. In particular, we expect that implementation of prescribed governance standards will have little effect on corporate value if owners maintain excessive control over the firm through the use of a pyramidal structure, which allows

them to effectively subvert these governance measures through indirect means. Overall, we address the following research questions: First, how does the quality of corporate governance practices relate to firm value in Thailand? Second, what is the relation between family influence, in the form of ownership or control structures, and firm value in Thailand? Finally, does family ownership or an enhanced control structure moderate the relation between the quality of corporate governance practices and firm value? The next section presents the empirical approach we use to address these questions. 3. Data and methodology The sample used in this study comprises all industrial companies traded on the Stock Exchange of Thailand (SET) in 2005. Firms in the financial services industry (banks, insurance companies, finance and securities companies, listed mutual fund companies, and property investment funds) are excluded from the sample. Firms that do not have complete financial data available for fiscal year 2005, newly listed firms, delisted firms, inactive firms, or firms undergoing financial rehabilitation or restructuring are also excluded from the sample. We next eliminate firms where the majority shareholder is the government or a widely-held domestic or foreign financial institution. Lastly, we eliminate firms that are subsidiaries of foreign corporations.12 This procedure results in a sample size of 216 firms. Financial data are obtained from Datastream, published by Thomson Financial, with additional financial information supplied by the Stock Exchange of Thailand through the SETSMART data service. 3.1. Identification of family-owned firms We determine whether or not a sample firm is considered to be family-owned based on the identity of the shareholders. We use the SETSMART database, provided by the Stock Exchange of Thailand, which lists the top ten shareholders of each firm, to identify share ownership. In addition to shareholding records, we use annual reports and other outside sources to trace share ownership and, thus, identify family ownership. For direct shareholdings, we count shareholders with the same surname as the family as well as shareholders with known familial relationships (relatives, spouse, children, etc.), even if the last names are different. Many firms in the sample exhibit indirect shareholdings; that is, shares owned by investors in a sample firm through another public or privately-held company. For each firm in the sample, we trace any indirect ownership of shares upwards through networks of public and/or private companies. We also add together shares owned by individual family members or owned by family-affiliated firms under family control to find total family ownership. We classify any company in the sample, which is part of a family-controlled network, as a ‘family’ firm. We begin by classifying firms as ‘high family ownership’ if the total family ownership exceeds the median ownership for all firms (41.2%).13 The objective of this initial classification of all sample firms into two groups based on the level of family ownership is to 12 We are interested in the effects of family versus non-family ownership. Thus government-owned firms are eliminated because a government may pursue public policy-related objectives through state-owned firms rather than policies designed to maximize shareholder wealth. For example, Gugler (2003) finds that state-owned Austrian firms smooth dividends and are much less likely to cut payouts when reductions are warranted compared with firms owned by families. In a similar vein, for a local (Thai) subsidiary of a foreign corporation, countervailing issues of tax management, profit repatriation, or transfer pricing may shift management priorities away from shareholder wealth maximization. 13 We also confirm that a more restrictive definition of high ownership (50%) yields qualitatively similar results.

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

compare the characteristics of firms that have high family ownership to firms with low family ownership. This initial analysis provides us with some insights about family firm behavior that we use to guide our subsequent empirical analyses. For the subsequent regression analyses, we maintain the division of high and low family ownership.14 This classification scheme results in two distinct groups: (i) high family ownership firms (family ownership above the median for all family firms); and (ii) low family ownership firms (family ownership below the median for all family firms). 3.2. Identification of pyramidal structures (ratio of cash flow rights to voting rights) Though Thai law requires one share, one vote, it is possible to have differences between voting or control rights and cash flow rights through use of indirect shareholdings or pyramidal ownership. The deviation between voting rights and cash flow rights for firms in our sample indicates the extent of control-enhancing mechanisms in place, such as pyramidal ownership. We follow prior research in the approach we use to identify cash flow rights and voting rights (La Porta et al., 1999).15 The ratio of cash flow rights to voting rights, or the ‘‘wedge’’, is the key measure we use to determine the ultimate owners of a firm through the chain of control. For companies that only have direct shareholdings, (i.e., no chain of control or holdings by affiliated companies), the cash flow rights are exactly equal to the voting rights and thus the ratio of the cash flow rights to the voting rights or the ‘‘wedge’’ is equal to one. A wedge value lower than one indicates that, through indirect or pyramidal ownership, firms have voting rights that exceed cash flow rights. For example, a sample firm may be majority-owned (50%) by a second public company which is, in turn, majority-owned (60%) by a group of family members. This firm is not widely held and clearly has a chain of control. Examining the chain of control, the family is the ultimate owner of the sample firm via its indirect holdings through the second firm. Thus, the sample company is considered a family firm. As in other studies, we determine the voting rights as corresponding to the smallest shareholding or ‘‘weakest link’’ in a chain of control. In this example, the voting rights would be 50%. However, the cash flow rights are 50%  60% or 30%, determined by multiplying the successive shareholdings along the chain of control. Thus, the deviation from cash flow rights and voting rights or ‘‘wedge’’ would be equal to the cash flow rights of 30% divided by the voting rights of 50%, yielding a wedge value of 0.60. We use the procedure described above to calculate the cash flow rights, voting rights, and wedge values for each firm in our sample. 3.3. Corporate governance measurement Beyond measurement of the usual variables, such as board size and independence, we construct a corporate governance index (CGI) in order to assess the quality of corporate governance practices among listed Thai firms. Calculated from a total of 117 separate criteria, the CGI quantifies the overall quality of corporate governance practices. We develop the criteria from the OECD’s (Organization for Economic Cooperation and Development) five corporate governance principles (OECD, 1999; 2004) and then 14 We identify 17 firms that have no family ownership. These firms are included in our low family ownership group. Our results remain qualitatively unaffected if we repeat the subsequent analyses after dropping these firms. 15 La Porta et al. (1999) use six classifications of ownership: widely-held (no dominant owner); family (members of the same family with the same last name); state (government ownership); widely-held financial institutions (financial institutions that do not have a single controlling large shareholder); widely-held corporations (corporations that do not have a single controlling large shareholder; and other. Essentially the same classification scheme is used in subsequent studies such as Claessens et al. (2000, 2002).

1727

adjust the criteria to take into account the subtleties of Thai laws and regulations.16 We construct this corporate governance index, rather than using an existing index (e.g., Gompers et al., 2003; Bebchuk et al., 2009; Brown and Caylor, 2006) because these other measures of governance may not be fully applicable to Asian markets, in general, and Thai firms, in particular. The other indexes are built primarily from provisions relating to takeover defenses and other shareholder rights. Hostile takeovers are rare in Asian markets largely because of concentrated and complex ownership structures, and unique institutional settings. The strengths of the corporate governance measure (CGI) used in this study are twofold. First, the CGI index measures the actual quality of corporate governance, i.e., the corporate governance-related activities or firm disclosures which, thus, reveal the practices actually implemented by the firms. The CGI also acts as a gauge of quality, showing whether the observed practices are missing (poor), match the level required by law (good), or reach the highest level of quality equivalent to international best practices (best). The second strength of this measure concerns the foundations of the survey measures themselves. The measures are rooted in economic and financial research findings and theories, aggregating many of the conclusions that have received empirical support from prior research. These theories and findings form the basis for the OECD’s corporate governance principles. The complete scorecard is shown in Appendix A. The scorecard criteria span five sections, matching the five OECD corporate governance principles: the rights of shareholders, equitable treatment of (minority) shareholders, role of stakeholders, disclosure and transparency, and board responsibilities. 3.3.1. Shareholders’ rights Shareholders’ rights should be protected by the corporate governance structure.17 In addition, the corporate governance structure should also make it easy for shareholders to exercise their rights. The first section of the measurement assesses the provision of basic shareholder rights, including the right to: (i) secure methods of ownership registration; (ii) convey or transfer shares; (iii) obtain relevant and material information on the corporation on a timely and regular basis; (iv) participate and vote in general shareholder meetings; (v) elect and remove members of the board; and (vi) share in the profits of the corporation. Twenty-two measures capture the actual rights of shareholders. To evaluate the quality of protection of shareholders’ rights, we examine in detail the relevant documents provided by 16 The measurement itself was initially developed as part of an annual project by the Thai Institute of Directors Association (Thai IOD) called ‘‘Corporate Governance Baselining in Thailand’’, which commenced in 2000. The late Mr. Charnchai Charuvastr, the former President of the Thai IOD, was instrumental in working with the Thai government to include this annual project as a part of the Thai National Governance Committee. To ensure adherence to international standards and comparability, the original questionnaire was designed with technical assistance from McKinsey and Company and the World Bank. The steering committee supervising the development and scoring process consists of representatives from the Stock Exchange of Thailand, the Securities and Exchange Commission, and related parties (e.g., the Institute of Certified Accountants and Auditors of Thailand, the Thai Investors’ Association, and the Government Pension Fund). The Steering Committee also determines the composition and weighting scheme of the CGI. In the first year, the World Bank generously provided financial support in the early stage of the project in Thailand. 17 Protection of shareholders’ rights is critical to economic and capital market development (LLSV, 1997). Previous studies have shown evidence that firm value declines when the control rights of the largest shareholders exceed their ownership rights (Claessens et al., 2002). An example of a shareholders’ right that should be protected is the ability to review the compensation for board members (Bushman et al., 2004; Murphy, 1999). Shareholders need to receive company information that is relevant and material (Gillian and Starks, 2000; Karpoff et al., 1996). Owners should also be able to participate and vote in general shareholder meetings (Bhagat and Brickley, 1984; Gordon and Pound, 1993; Ferris et al., 2003; Fich and Shivdasani, 2005), and elect and remove members of the board (LLSV, 1998; Fama and Jensen, 1983).

1728

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

the companies and the regulators. The resultant score for this subsection is scaled to comprise 25% of CGI, the governance index.18 3.3.2. Equitable treatment of shareholders All shareholders should be treated similarly whether they are dominant owners, minority owners, or foreign shareholders. In East Asia, the treatment of shareholders is critical because the companies operate in an environment in which inside, majority shareholders have advantages over outside, minority shareholders (Claessens et al., 2002). For example, the second section of the CGI measurement assesses whether processes and procedures for general shareholder meetings allow for equitable treatment of all shareholders and whether company procedures make it unduly difficult or expensive for shareholders to cast votes.19 There are 13 items to evaluate the equitable treatment of shareholders reported in the official documents provided by the firms and the regulators. The score for this sub-section makes up approximately 15% of CGI. 3.3.3. Role of stakeholders The corporate governance framework should recognize the rights of stakeholders established by law or through mutual agreements and encourage active cooperation between corporations and stakeholders in creating financially sound enterprises. The third part of the scorecard covers the role of stakeholders, specifically the interactions of the firm with stakeholder groups, such as employees, creditors, suppliers, shareholders, and the environment.20 This section of the CGI measurement assesses whether stakeholders can participate in the corporate governance process or have access to relevant and reliable information on a timely and regular basis. There are nine survey items in this sub-section, making up approximately 10% of CGI. 3.3.4. Disclosure and transparency Corporate disclosure and transparency are two cornerstones of good governance.21 Therefore, the corporate governance framework 18 The Steering Committee (See Footnote 16) assigned weights to each component of CGI. However, our results are robust to alternative weighting schemes including assigning equal weights to each component. 19 There are several aspects of treatment of shareholders included in the measurement. For example, Thai law prescribes one share, one vote, which has been shown to benefit shareholders (Grossman and Hart, 1988; Harris and Raviv, 1988). Additionally, shareholders should have the opportunity to obtain effective redress should their rights be violated. One example would be the abuse of minority shareholders at the hands of controlling shareholders. Abusive related-party transactions (tunneling or propping) have been documented as a serious violation of shareholder rights in emerging markets (Cheung et al., 2006; Friedman et al., 2003; Johnson et al., 2000). In addition, internal control systems need to be established to prevent the use of material inside information (Givoly and Palmon, 1985), for example to prevent insider trading. 20 Jensen (2002) contends that a firm cannot maximize value if it ignores the interest of its stakeholders. This view is supported by Allen et al. (2009) who note that firms may voluntarily choose to be stakeholder-oriented because this increases their value in certain circumstances. Connelly and Limpaphayom (2004) find that an optimally designed environmental policy can maximize market valuation in an emerging market. 21 Prior research (e.g., Ball, 2001; Bushman et al., 2004; Khanna et al., 2004) demonstrates that both the quantity and the quality of corporate disclosure are integral parts of effective governance practices. First, the ownership stakes of a firm should be transparent (LLSV, 1998; Bushman et al., 2004; Claessens et al., 2002; Mallette and Fowler, 1992; Himmelberg et al., 1999). The quality of information provided to investors is another important dimension(Meek et al., 1995; Singhvi and Desai, 1971). The quality of disclosure can also be a useful indicator of the levels of agency conflicts and information asymmetries within the firm (Healy and Palepu, 2001; Meek et al., 1995). Further, information such as related-party transactions, external audit results, and insider transactions would be important to investors (Cheung et al., 2006; Johnson et al., 2000; Fan and Wong, 2005). Also, prior research has identified the benefits of using multiple channels to communicate with investors (Ashbaugh et al., 1999; Lang and Lundholm, 1993, 1996; Farragher et al., 1994). Anderson et al. (2009) find that family firms in the US lack transparency and that founder- and heir-controlled firms are associated with weaker performance when they are opaque.

should ensure timely and accurate disclosures are made on all material matters regarding the corporation, including the financial situation, performance, ownership, and governance of the company. This section of the CGI measurement assesses whether the information was prepared and disclosed in accordance with acceptable corporate standards for accounting and financial and non-financial disclosures. Further, it assesses whether the channels for disseminating information provide for all users equal, timely, and cost-efficient access to the relevant information. There are 32 items in the survey that examine disclosure practices. These measures account for approximately 25% of the CGI score. 3.3.5. Board responsibilities The final section of the scorecard addresses board responsibilities, an area that has generated an extensive amount of research interest. The OECD principles reaffirm that the corporate governance framework should ensure the strategic guidance of the company, the effective monitoring of management by the board, and the board’s accountability to the company and the shareholders.22 Much of the prior research focuses on governance variables, identifying broad board characteristics such as independence and board size. Often, however, in Thailand, there are deep personal connections between the CEO and board members. Extra-contractual mechanisms are usually available to the CEO or family for disciplining and controlling directors, who are often family members or otherwise connected to the family through business or personal dealings. We believe our in-depth assessment of board practices and policies provides a more accurate gauge of a board’s ability to implement deliberate and meaningful practices designed to enhance corporate governance. In total, there are 41 survey items to assess board responsibilities. The items in this section make up approximately 30% of CGI. We draw the data used to evaluate the governance practices for each firm from a wide variety of publicly available information sources, such as annual reports, Securities and Exchange Commission and Stock Exchange of Thailand filings, minutes from annual shareholders’ meetings, articles of association, company by-laws, and company websites. To reinforce the emphasis on minority shareholders, this study assumes the viewpoint of an outside investor. Specifically, only publicly available official documents serve as source documents since this information would be readily available to outside investors. We score each company on all applicable areas of CGI. As assessing the level of corporate governance for an individual company can be subjective, we design the scoring scheme to minimize this problem. In addition to cross-checking and auditing by different raters, we quantify nearly every governance measure in our study. This is also a unique feature of this study, as previous research has only checked for the presence or absence of a specific corporate governance measure. With our assessment procedure, companies that omit or do not comply with a specific scoring 22 An effective board can reduce agency conflicts by exercising its power to monitor and control management (Fama and Jensen, 1983). Frequent and well-attended board meetings can have a positive effect on firm performance (Vafeas, 1999; Ferris et al., 2003; Fich and Shivdasani, 2005). The auditing role of the board is one area where substance, in this case concrete actions that enhance governance effectiveness, can be separated from form (Adams, 1994; Scarbrough et al., 1998; Carcello et al., 2002; Turpin and DeZoort, 1998). Directors must also exercise their role in oversight, including evaluation of the CEO (Boyd, 1994) and themselves (Ingley and van der Walt, 2002). One area where substance would clearly supersede form is the important area of board committees (Klein, 1998). A number of researchers have investigated the roles of board committees in propagating principles of good governance, e.g., on the importance of audit committees (Carcello and Neal, 2000; Klein, 2002; Krishnan, 2005). Well-functioning audit, compensation, and director nomination committees, each with a clear mission and each effectively discharging its duties, play a role in protecting shareholder interests and increasing firm value (Brick et al., 2006; Daily et al., 1998).

1729

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743 Table 1 Descriptive statistics of corporate governance index by survey year. Year

CGI

Section A

Section B

Section C

Section D

Section E

Number of firms in survey

2002 2004 2005

57.61 61.30 61.79

72.45 63.68 64.40

69.32 72.52 74.85

43.06 61.80 64.04

55.54 74.05 73.09

49.26 50.09 50.84

294 327 364

This table presents the average of the Corporate Governance Index (CGI) and the five CGI sub-indexes based on the OECD corporate governance principles (1999) for three CGI surveys. The surveys were completed during 2002, 2004, and 2005. The sample is drawn from publicly-traded firms in Thailand. The CGI and each sub-index ranges are expressed as percentages and range from 0 to 100. The five subsections are: rights of shareholders (Section A); equitable treatment of shareholders (Section B); role of stakeholders (Section C); disclosure and transparency (Section D); and board responsibilities (Section E). Survey questions are shown in Appendix A.

80 04

02 70

Corporate Governance Index (CGI) 2002, 2004, and 2005 Surveys

04 05 60

05

04 05

02 04 05

04

05

02

02 05 02 04

50 02 40

30

20

10

0 CGI

Section A

Section B

Section C

Section D

Section E

CGI Scores by Section, based on the OECD Principles of Corporate Governance (1999) Fig. 2. Average corporate governance survey scores (CGI). The figure shows the average scores of the Corporate Governance Index (CGI) in Thailand for three survey years: 2002, 2004, and 2005. The five subsections are: rights of shareholders (Section A); equitable treatment of shareholders (Section B); role of stakeholders (Section C); disclosure and transparency (Section D); and board responsibilities (Section E).

criterion receive a ‘poor’ score. Meeting the legal compliance standard earns a firm a score of ‘good’, while firms which exceed the regulatory requirements and/or meet international standards receive the highest score. Once the assessment is complete, we scale the CGI score, aggregated across the five sub-sections, from zero to 100%. The original version of the survey was designed and piloted on a select group of 133 companies in 2000. The Thai Institute of Directors Association conducted the second and third studies in 2002 and 2004. After the initial survey and during the intervening years, a number of questions in the original survey were revised while new questions were added. The original version contained 57 questions whereas the latest version contains 117 questions (see Appendix A). While the number of questions and the format of the survey have changed over the years, the structure of the survey retains the same mapping of the OECD Principles of Corporate Governance (1999). The survey was then executed again in 2005 with the most comprehensive coverage: nearly all firms listed on the Stock Exchange of Thailand were included in the study. The time-series results of the survey from 2002 to 2005 are shown in Table 1 and, graphically, in Fig. 2. The results from the overall CGI scores show that the quality of corporate governance practices of Thai companies has gradually improved. Specifically, the average CGI score improved from 57.61 (out of 100) in 2002 to 61.79 in 2005. This is attributable to the reform efforts initiated by the government and related organizations. The results clearly show that Thai companies are going through a major transformation

and, therefore, no longer have the same characteristics as Thai companies examined in previous studies using samples before the Asian financial crisis (e.g., Claessens et al., 2000; Wiwattanakantang, 2001). From the descriptive statistics in Table 1, the increase in the overall quality of corporate governance practices comes largely from two major areas, disclosure and transparency (Section D) and role of stakeholders (Section C). The improvement in the area of disclosure and transparency is shown by the increase in the average score from 55.54 in 2002 to 73.09 in 2005. This rise is most likely a result of regulatory reforms and the implementation of new rules and regulations by the Stock Exchange of Thailand (SET) and the Securities and Exchange Commission. During this period, the SET and the Thai Institute of Directors Association also launched programs to increase the awareness of the role of stakeholders among listed firms which, in turn, led to a sizable increase in the score (43.06 in 2002–64.04 in 2005) in Section C of the survey. The equitable treatment of shareholders (Section B) also shows some improvement (69.32 in 2002–74.85 in 2005) whereas the scores for the board responsibilities segment (Section E) show virtually no improvement. The most striking result is a substantial decline in the scores for the rights of shareholders (Section A), going from 72.45 in 2002 to 64.40 in 2005. This is largely due to an increase in cross-holdings and the use of pyramidal structures by Thai family firms. In a pre-crisis study by Claessens et al. (2000), the average ratio between cash flow rights and voting rights among Thai firms is close to 1.0 (i.e.,

1730

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

no pyramidal structures are observed). In 2005, the average ratio between cash flow rights and voting or ownership rights is 0.833, indicating that Thai firms now have substantially increased the use of this type of control-enhancing structure. Overall, it appears that while Thai firms have improved some governance measures, these firms have also experienced an effective decline in shareholder rights. 3.4. Empirical methods In our initial analyses, we examine the influence of family control and family ownership on firm performance using OLS regression analyses.23 Tobin’s q, a measure of firm market valuation, is the dependent variable.24 We include variables for family ownership and various board characteristics in the regressions, as well as other firm characteristics, which serve as control variables. We evaluate the first model, shown below as Eq. (1), using the full sample.

qi ¼ bo þ b1 BOD SIZEi þ b2 BOD INDi þ b3 FAM OWNi þ b4 FAM OWN2i þ b5 SIZEi þ b6 FIRM AGEi þ b7 ROAi þ b8 CAPITAL EXPENDITURESi þ b9 LEVERi þ e1i

ð1Þ

For the second regression, we add a dummy variable (Wedge dummy), indicating the presence of a control-enhancing pyramidal ownership structure, as an additional explanatory variable to the first model.

qi ¼ bo þ b1 WEDGE DUMMYi þ b2 BOD SIZEi þ b3 BOD INDi þ b4 FAM OWNi þ b5 FAM OWN2i þ b6 SIZEi þ b7 FIRM AGEi þ b8 ROAi þ b9 CAPITAL EXPENDITURESi þ b10 LEVERi þ e1i

ð2Þ

For the third regression, we add CGI, the corporate governance score, as an additional explanatory variable to Model 1.

qi ¼ bo þ b1 CGIi þ b2 BOD SIZEi þ b3 BOD INDi þ b4 FAM OWNi þ b5 FAM OWN2i þ b6 SIZEi þ b7 FIRM AGEi þ b8 ROAi þ b9 CAPITAL EXPENDITURESi þ b10 LEVERi þ e1i

ð3Þ

Lastly, we include both the Wedge dummy variable and CGI as additional explanatory variables.

qi ¼ bo þ b1 CGIi þ b2 WEDGE DUMMYi þ b3 BOD SIZEi þ b4 BOD INDi þ b5 FAM OWNi þ b6 FAM OWN2i þ b7 SIZEi þ b8 FIRM AGEi þ b9 ROAi þ b10 CAPITAL EXPENDITURESi þ b11 LEVERi þ e1i

ð4Þ

Tobin’s q, as a market-based measure of firm performance, is defined as the sum of the book value of long-term debt and the market value of equity divided by the book value of total assets. CGI is the percentage score from the Corporate Governance Index based on the OECD Principles of Corporate Governance (2004). The CGI score is comprised of scores on five sub-components of CGI: rights of shareholders, treatment of shareholders, role of stakeholders, disclosure and transparency, and board responsibilities. Both CGI and the sub-component scores are expressed as percentages. Family 23 In subsequent analyses, we use a simultaneous equations approach to control for a potentially endogenous relation between CGI and q. 24 To make the results comparable to Bebchuk et al. (2009), the empirical analyses are repeated using the natural logarithm of q as the dependent variable. The overall results are qualitatively similar to the reported results.

ownership is the proportion of outstanding shares held by the founding family and affiliated members. We include several board-related measures that have been used in other studies as control variables. These elements, i.e., board size and board composition, and their relation to firm performance, have been investigated in the literature but show mixed results. In this study, board size is the number of directors on a firm’s board while board independence (IND) is the proportion of directors who are independent/outside directors. For robustness, we also include several other control variables. We calculate profitability by taking the ratio of net income after taxes divided by total assets. Firm size is the natural log of total assets. Firm age is the natural logarithm of the years since the firm’s founding. We define financial leverage as the ratio of long-term debt divided by total assets. 4. Results Table 2 contains the descriptive statistics for the 216 firms in the sample. We classify companies into ‘‘High Family Ownership’’ and ‘‘Low Family Ownership’’ groups depending on whether the proportion of outstanding shares held by family or related members is above or below the median for all firms (41.2%). Table 2 shows three sets of statistics: (i) for the full sample of 216 firms, (ii) for 108 firms with high family ownership, and (iii) 108 firms with low family ownership (including firms with zero family ownership). When comparing the high family ownership firms to low family ownership firms, the descriptive statistics show some striking differences. The average share ownership of family and affiliated members is 39.29% for the full sample. Low family ownership firms have a mean family ownership percentage of 20.21% whereas the mean percentage is 58.37% for high family ownership firms. The difference is statistically significant at conventional levels (t = 20.76). The difference in firm performance, as measured by Tobin’s q, is also statistically significant and lower for high family ownership firms. The average Tobin’s q for the full sample is 0.82. Low family ownership firms have an average q of 0.88, which is higher than the average q for high family ownership firms (0.76). The difference is statistically significant at the 10% level (t = 1.77). The average return on assets for high family ownership firms is also lower than that for low family ownership firms but the difference is not statistically significant (t = 1.18). Firms with high family ownership are, on average, slightly smaller (mean firm size of 14.74, measured by the natural logarithm of total assets) than low family ownership firms (15.10; t = 2.07) and have lower levels of financial leverage, measured by the ratio of longterm debt to total assets (0.09 versus 0.14; t = 2.48), as judged by the ratio of long-term debt to total assets, suggesting that high family ownership is associated with more conservative financing strategies.25 With respect to general governance characteristics, the two types of firms have similar percentages for the average proportion of independent directors on the board. The mean is 35% (low ownership) versus 33% (high ownership) for the two groups of firms (t = 1.37). Similarly, the two types of firms have average board sizes (11.1 directors versus 11.3) that are nearly identical. While the overall CGI scores for the two groups are similar, with a mean score of 69.17 for high family ownership firms versus a mean of 70.25 for low family ownership firms (t = 0.84), some CGI subsection scores are lower for high family ownership firms than low family ownership companies. Rights of Shareholders and Disclosure 25 Family-controlled firms also appear to be associated with lower levels of debt in the US. Agrawal and Nagarajan (1990) report that all-equity firms have higher cash holdings and significantly higher family control compared to a matched control sample of levered firms.

1731

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743 Table 2 Descriptive statistics.

Family ownership (%) Wedge Tobin’s q Profitability (ROA) Firm size Firm age Capital expenditures Financial leverage Board size Board independence Corporate governance index Rights of shareholders Treatment of shareholders Role of stakeholders Disclosure and transparency Board responsibilities N

All firms

High family ownership (1)

Low family ownership (2)

tStatistic (1–2)

No pyramidal structure observed (3)

39.29 (23.39) 0.83 (0.30) 0.82 (0.50) 0.05 (0.07) 14.92 (1.31) 3.18 (0.47) 0.08 (0.08) 0.11 (0.13) 11.20 (3.02) 0.34 (0.10) 69.71

58.37 (11.46) 0.82 (0.31) 0.76 (0.47) 0.05 (0.07) 14.74 (1.21) 3.24 (0.41) 0.08 (0.07) 0.09 (0.12) 11.26 (3.24) 0.33 (0.10) 69.17

20.21 (15.28) 0.85 (0.29) 0.88 (0.52) 0.06 (0.08) 15.10 (1.37) 3.13 (0.51) 0.08 (0.08) 0.14 (0.14) 11.14 (2.81) 0.35 (0.11) 70.25

20.76*** (23.47) 0.87 (0.00) 1.77* (0.49) 1.18 (0.08) 2.07** (1.31) 1.75* (0.44) 0.20 (0.07) 2.48** (0.12) 0.29 (2.98) 1.37 (0.10) 0.84

37.56 (22.72) 1.00 (0.22) 0.78 (0.49) 0.05 (0.05) 14.88 (1.30) 3.19 (0.53) 0.08 (0.09) 0.11 (0.16) 10.91 (3.04) 0.34 (0.11) 68.40

(9.47) 70.93 (13.89) 74.06

(9.40) 69.27 (12.78) 73.72

(9.55) 72.59 (14.78) 74.40

(9.49) 1.76* (14.11) 0.70

(8.51) 69.91 (12.96) 73.81

(7.15) 70.49 (21.18) 81.74

(6.59) 71.05 (20.09) 80.99

(7.68) 69.94 (22.30) 82.48

(7.44) 0.38 (22.24) 1.22

(6.29) 68.64 (17.15) 80.62

(8.99) 53.93 (14.08) 216

(8.96) 53.70 (14.50) 108

(8.98) 54.15 (13.72) 108

(9.45) 0.23

(6.77) 51.82 (13.60) 158

Pyramidal structure observed (4) 44.00

tStatistic (3–4) 1.80*

0.38

35.13***

0.93

2.03**

0.06

0.95

15.03

0.74

3.17

0.28

0.09

0.94

0.11

0.01

12.00

2.38**

0.32

1.21

73.28

3.44***

73.70

1.79*

74.74

0.85

75.54

2.14**

84.77

3.07***

59.67 (13.88) 58

3.74***

This table presents summary statistics of variables used in the study. The sample consists of firms listed on the Stock Exchange of Thailand in 2005. The sample is split based on the median value of family ownership. Family ownership is the number of outstanding shares held by the founding family and affiliated members divided by the total number of shares outstanding. Wedge is the ratio of the cash flow rights divided by the voting rights. Tobin’s q is the book value of long-term debt plus the market value of equity divided by the book value of total assets. Profitability (ROA) is the ratio of net income after taxes divided by total assets. Firm size is the natural log of total assets. Firm age is the natural log of the years since the firm’s founding. Capital expenditures are the ratio of capital expenditures divided by total assets. Financial leverage is the ratio of long-term debt divided by total assets. Board size indicates the number of directors on the board of directors. Board independence is the number of directors who are independent/outside directors divided by board size. CGI is the percentage score from the Corporate Governance Index based on the OECD Principles of Corporate Governance. The five subcomponents of CGI are: rights of shareholders, equitable treatment of shareholders, role of stakeholders, disclosure and transparency, and board responsibilities, expressed as percentages. Standard deviations are shown in parentheses. t-Statistics are calculated for the differences between family-owned firms and other firms. * statistical significant differences at the 10% level (two-tailed). ** statistical significant differences at the 5% level (two-tailed). *** statistical significant differences at the 1% level (two-tailed).

and Transparency are both lower for high family ownership companies. However, the difference is statistically significant only for Rights of Shareholders. The scores for the Treatment of Shareholders, Roles of Stakeholders, and Board Responsibilities sections are statistically indistinguishable between the two types of firms. We classify our sample companies into two groups based on the ratio between cash flow rights and voting rights. Companies with a ‘‘wedge’’ or pyramidal structure are defined as those that have a ratio of cash flow rights to voting rights lower than one. The comparison between sample firms that employ a pyramidal ownership structure and other firms, shown in columns 4 and 3, respectively of Table 2, reveals some interesting findings. From the descriptive statistics in Table 2, both groups of companies are quite similar with respect to financial characteristics. There is no statistical difference in firm age, firm size, profitability, capital expenditures and leverage between the two groups. However, companies with a pyramidal structure have higher average family ownership than those with no wedge (t = 1.80). This is consistent with the notion that founding families attempt to use the pyramidal structure to maintain control

of their firms. Interestingly, firms with pyramidal structures have a higher mean valuation than those that do not employ the pyramidal structure (t = 2.03). It also appears that companies with pyramidal structures exhibit better corporate governance, as measured by CGI. On the surface, it appears that pyramidal structures are valueenhancing. However, these univariate comparisons could be misleading not only because of the complex and potentially endogenous nature of the interrelationship among ownership, control and value, but also because greater control may allow family firms to manipulate the corporate governance measures to show better scores, without actually implementing value enhancing strategies. We examine these issues further in subsequent empirical analyses. Table 3 contains a correlation matrix for the variables used in the study. Panel A contains the correlations between the overall CGI score and firm characteristics. The correlation between Tobin’s q and board independence is positive and statistically significant while the correlation with board size is not statistically significant. Most importantly, there appears to be a positive and statistically significant relation between q and the quality of corporate governance

1732

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Table 3 Correlation matrix. (1)

(2)

Panel A: Correlations between CGI and firm characteristics (1) Family 1.0 (2) Wedge 0.09 1.0 (3) Tobin’s q 0.09 0.14 (4) ROA 0.05 0.08 (5) Firm size 0.18 0.01 (6) Firm age 0.11 0.01 (7) Cap ex 0.02 0.05 (8) Leverage 0.19 0.02 (9) Board size 0.08 0.19 (10) Independence 0.13 0.14 (11) CGI 0.12 0.19

Panel B: Correlations among CGI and sub-components of CGI (1) CGI (2) Rights of shareholders (3) Treatment of shareholders (4) Role of stakeholders (5) Disclosure and transparency (6) Board responsibilites

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

1.0 0.49 0.20 0.06 0.24 0.14 0.05 0.12 0.30

1.0 0.12 0.01 0.18 0.01 0.02 0.09 0.25

1.0 0.08 0.05 0.47 0.14 0.06 0.47

1.0 0.14 0.14 0.26 0.14 0.02

1.0 0.31 0.06 0.02 0.02

1.0 0.04 0.13 0.27

1.0 0.62 0.05

1.0 0.07

1.0

(1)

(2)

(3)

(4)

(5)

(6)

1.0 0.80 0.24 0.74 0.75 0.87

1.0 0.13 0.44 0.52 0.63

1.0 0.07 0.18 0.04

1.0 0.37 0.54

1.0 0.60

1.0

This table presents correlation coefficients among variables used in the study. The sample consists of firms listed on the Stock Exchange of Thailand in 2005. In Panel A, family ownership is the number of outstanding shares held by the founding family and affiliated members divided by the total number of shares outstanding. Wedge is the ratio of the cash flow rights divided by the voting rights. Tobin’s q is the book value of long-term debt plus the market value of equity divided by the book value of total assets. Profitability (ROA) is the ratio of net income after taxes divided by total assets. Firm size is the natural log of total assets. Firm age is the natural log of the years since the firm’s founding. Capital expenditures is the ratio of capital expenditures divided by total assets. Financial leverage is the ratio of long-term debt divided by total assets. Board size indicates the number of directors on the board of directors. Board independence is the number of directors who are independent/outside directors divided by board size. CGI is the percentage score from the Corporate Governance Index survey based on the OECD Principles of Corporate Governance. Panel B presents the correlation coefficients between CGI and the five sub-components of CGI: rights of shareholders, equitable treatment of shareholders, role of stakeholders, disclosure and transparency, and board responsibilities. Correlations that are statistically significant at the 10 percent level are shown in bold.

practices, as measured by the CGI. Panel B shows the correlations between CGI and the five sub-components. The correlations among the sub-components are positive and statistically significant, showing that these five aspects of corporate governance practices are interrelated. Table 4 presents regression results using Tobin’s q as the dependent variable. In this table, we run regressions using all firms in the sample, adding, in turn, additional variables of interest such as dummy variables for the presence of pyramidal ownership and the CGI. To recap, the purpose of this series of regression analyses is twofold. First, we want to examine the relation between firm value, as measured by q, and (1) the proportion of family ownership and (2) conventional governance variables. We also wish to compare the predictive ability of the CGI variable to that of conventional governance variables. Secondly, we want to look at the elation between the use of pyramidal ownership structures and firm value, after controlling for other factors. In the first model, the coefficient for family ownership is not statistically significant at conventional levels. This indicates that after controlling for other factors, there is no significant relation between family ownership and firm value, consistent with ownership being endogenously determined (Demsetz and Lehn, 1985). In Model 2, we include a dummy variable to indicate the presence of a pyramidal structure (Wedge dummy). The coefficient for the presence of a pyramidal structure is positive and statistically significant. However, the statistical significance disappears after adding additional control variables, as shown in the subsequent models. Looking at the results from models 1 and 2, a striking finding is that there is no significant relation between q and conventional corporate governance variables, such as board characteristics. It appears that these governance variables, which have traditionally been found to be associated with firm value (q) in developed economies following common law,26 apparently do not

26 Thus, our comparison group is primarily firms in the US and UK. For instance, see Yermack (1996) on the significance of board size and Morck et al. (1988) on the relation between managerial stock ownership and value.

exert any influence on firm value for Thai family firms. This finding also provides preliminary support for our contention that many governance variables represent the ‘‘form’’ but not the ‘‘substance’’ of effective corporate governance practices in countries where the family exerts substantial control over voting rights and control mechanisms are not transparent. In order to shed further light on the relation between the quality of governance practices and value, we go beyond the standard governance variables and turn to the Corporate Governance Index (CGI) described earlier. Recall that the CGI index is a composite score, aggregating many of the mandated and voluntary provisions and disclosures made by firms. These provisions and disclosures are relevant to the quality and visibility of corporate governance practices. Thus, we expect the CGI index to exhibit an association with market valuation. Models 3 and 4 present regression analyses with the addition of CGI. We find that as before, the coefficients for conventional governance variables (e.g., board size, board independence and family ownership) are still not statistically significant. Most interestingly, the regression shows a positive and statistically significant relation between CGI and Tobin’s q. The coefficient for CGI is positive in model 3 and in model 4, which includes the dummy variable for the presence of a pyramidal ownership structure. Moreover, in model 4, the regression coefficient for the presence of a pyramidal structure is no longer statistically significant. We find that of the control variables, profitability shows a positive and statistically significant relation to Tobin’s q in all models. The coefficients for capital expenditures are also positive and significant in all four regression models. Financial leverage does not have a statistically significant coefficient in any model. It is possible that our results are affected by an endogenous relation between CGI and q. Lehn et al. (2007) find that after controlling for past performance, there is no contemporaneous association between the quality of corporate governance, as measured by the GIM Index,27 and firm valuation. As a robustness check, we 27

See Gompers et al. (2003).

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

þ b7 SIZEi þ b8 FIRM AGEi þ b9 ROAi

Table 4 Regression results for Tobin’s q. All firms (1)

þ b10 CAPITAL EXPENDITURESi þ b11 LEVERi þ e1i All firms (2)

All firms (3) 0.008*** (3.31)

Board size Board independence Family ownership Family own. squared Firm size Firm age Profitability Capital expenditures Financial leverage Constant

Adjusted R2 N

All firms (4)

0.001 (0.02) 0.122 (0.29) 0.004 (1.06) 0.001 (0.82) 0.042 (1.28) 0.008 (0.10) 3.026*** (4.56) 0.802* (1.94) 0.114 (0.49) 0.031 (0.07)

0.145** (2.35) 0.006 (0.41) 0.189 (0.44) 0.004 (1.03) 0.001 (0.68) 0.049 (1.57) 0.046 (0.59) 2.927*** (4.39) 0.883** (2.10) 0.098 (0.45) 0.241 (0.53)

0.001 (0.03) 0.194 (0.46) 0.001 (0.32) 0.001 (0.27) 0.014 (0.43) 0.058 (0.73) 2.804*** (4.26) 0.887** (2.19) 0.038 (0.16) 0.052 (0.11)

0.007** (2.44) 0.090 (1.36) 0.004 (0.29) 0.229 (0.54) 0.002 (0.45) 0.001 (0.33) 0.024 (0.74) 0.016 (0.19) 2.760*** (4.18) 0.929** (2.27) 0.053 (0.24) 0.205 (0.44)

0.255 216

0.261 216

0.271 216

0.275 216

CGI Wedge dummy

1733

This table presents OLS regression results with Tobin’s q as the dependent variable. The sample consists of firms listed on the Stock Exchange of Thailand in 2005. CGI is the percentage score from the corporate governance index based on the OECD Principles of Corporate Governance. Wedge dummy is a dummy variable with a value of one if the ratio of the cash flow rights divided by the voting rights is less than 1 and zero if the ratio is equal to 1. Tobin’s q is the book value of long-term debt plus the market value of equity divided by the book value of total assets. Board size indicates the number of directors on the board of directors. Board independence is the number of directors who are independent/outside directors divided by board size. Family ownership is the number of outstanding shares held by the founding family and affiliated members divided by the total number of shares outstanding. Firm age is the natural log of the years since the firm’s founding. Profitability (ROA) is the ratio of net income after taxes divided by total assets. Capital expenditures is the ratio of capital expenditures divided by total assets. Financial leverage is the ratio of long-term debt divided by total assets. t-Statistics are shown in parentheses. * Statistical significant differences at the 10% level (two-tailed). ** Statistical significant differences at the 5% level (two-tailed). *** Statistical significant differences at the 1% level (two-tailed).

include the average Tobin’s q measured over 2002 and 2004 as an additional control variable in our analysis. We next run two-stage least-squares (2SLS) regression models using the full sample, as shown below in Eqs. (5a) and (5b). We then split the sample into four groups to isolate the effects of family ownership and pyramidal structures. We split the sample according to family ownership (above or below the median level of ownership) and the presence or absence of pyramidal structures. Next, we re-estimate the twostage least-squares regression model for each of the four groups. Due to econometric reasons, family ownership (in Eqs. (5a) and (5b)) is omitted from the additional regressions for the subgroups (5c) and (5d).

qi ¼ bo þ b1 WEDGE DUMMYi þ b2 CGIi þ b3 Averageqi þ b4 BOD SIZEi þ b5 BOD INDEPi þ b6 FAM OWNi þ b7 FAM OWN2i þ b8 SIZEi þ b9 FIRM AGEi þ b10 ROAi þ b11 CAPITAL EXPENDITURESi þ b12 LEVERi þ e1i CGIi ¼ bo þ b1 WEDGE DUMMYi þ b2 Averageqi þ b3 BOD SIZEi þ b4 BOD INDEPi þ b5 FAM OWNi þ b6 FAM OWN2i

ð5aÞ

ð5bÞ

qi ¼ bo þ b1 WEDGE DUMMYi þ b2 CGIi þ b3 Averageqi þ b4 BOD SIZEi þ b5 BOD INDEPi þ b6 SIZEi þ b7 FIRM AGEi þ b8 ROAi þ b9 CAPITAL EXPENDITURESi þ b10 LEVERi þ e1i

ð5cÞ

CGIi ¼ bo þ b1 WEDGE DUMMYi þ b2 Averageqi þ b3 BOD SIZEi þ b4 BOD INDEPi þ b5 SIZEi þ b6 FIRM AGEi þ b7 ROAi þ b8 CAPITAL EXPENDITURESi þ b9 LEVERi þ e1i

ð5dÞ

The main objective of the two-stage least-squares analysis is to see whether the contemporaneous relation between the quality of corporate governance and firm valuation disappears after controlling for past performance. The results from the Tobin’s q equation in Table 5 show that our previous results are robust. Although the regression coefficient for the past average q is statistically significant, the coefficient for contemporaneous CGI continues to be statistically significant in the model using the full sample. Most interestingly, the coefficient for the presence of a pyramidal structure is now negative and statistically significant in the Tobin’s q regression. Once the analysis takes into account past performance, the use of a pyramidal structure is shown to be associated with lower firm values. This finding is consistent with the notion that controlling shareholders have the ability to expropriate wealth from other shareholders. The CGI regression also shows that the wedge dummy is positive and significantly associated with CGI. This result, rather counter-intuitively, suggests that firms with a pyramidal ownership structure score higher on the governance index. However, it is also likely that firms with greater control (wedge) firms manipulate the governance measures to show higher governance scores, while also impairing value. To understand this further, we split the full sample into subsamples based on ownership and wedge values and run the q and CGI regressions for each subsample. An interesting and important finding emerges in the subgroup analyses. We find q is significantly associated with CGI only for firms that do not have a pyramidal control structure (i.e., those firms with no deviation between cash flow rights and voting rights). This empirical evidence suggests that the observed contemporaneous relation between the quality of corporate governance practices and market valuation exists only when the owners do not employ opaque ownership structures that can be exploited to the disadvantage of other shareholders. The results are consistent with the notion that controlling owners can use the pyramidal structure to undermine the positive benefits of adopting good corporate governance practices. Controlling owners can attempt to show that their firms have improved the quality of corporate governance practices but, in fact, negate the positive effect by using the pyramidal structure as a counter measure. We also find, from the CGI equations in the subsample regressions, that there is no consistently significant relation between past performance and the quality of current corporate governance practices among firms in our sample. This result is unlike the results obtained by Lehn et al. (2007). Specifically, the regression coefficients for the average Tobin’s q are not statistically significant for most of the models in Table 5. We conclude that there is no consistent empirical evidence that the contemporaneous relation between CGI and q for our sample firms is an issue of reverse causality. In addition, the coefficients for firm size are positive and statistically significant. This finding is consistent with the notion that large firms tend to have better corporate governance practices.

1734

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Table 5 Two-stage least squares regression results. All firms Tobin’s q Wedge dummy CGI Average Tobin’s q Board size Board independence Family ownership Family own. squared Size Firm age Profitability Capital expenditures leverage

Adjusted R2 N

0.34* (1.79) 0.10*** (3.54) 0.23** (2.03) 0.01 (0.15) 0.77 (1.01) 0.01 (0.36) 0.01 (0.41) 0.37*** (3.24) 0.20 (1.39) 0.10 (0.09) 1.27 (1.51) 0.12 (0.22) 0.502 216

CGI

High FO, wedge < 1

High FO, no wedge

Low FO, wedge < 1

Low FO, no wedge

Tobin’s q

CGI

Tobin’s q

CGI

Tobin’s q

CGI

Tobin’s q

0.01 (0.08) 0.22 (1.23) 0.03 (1.15) 0.64 (0.59)

1.04 (0.25) 0.05 (0.09) 36.21 (1.64)

0.03*** (3.49) 0.52*** (5.67) 0.01 (0.26) 0.90* (1.85)

3.17 (1.41) 0.02 (0.04) 15.12 (1.24)

0.01 (0.38) 0.87*** (6.73) 0.04* (1.78) 0.41 (0.98)

5.21 (0.82) 0.30 (0.26) 8.77 (0.41)

0.10** (2.32) 0.13 (0.90) 0.07 (1.34) 0.16 (0.13)

0.66 (0.54) 0.44 (1.09) 4.87 (0.46)

0.01 (0.22) 0.19 (1.27) 6.78*** (3.57) 0.02 (0.03) 1.05 (1.57)

3.54*** (4.14) 1.65 (0.48) 83.07** (2.46) 12.98 (0.67) 2.75 (0.18)

0.12*** (2.70) 0.10 (1.22) 2.02*** (3.62) 1.24* (1.84) 0.27 (0.78)

4.16*** (6.12) 0.49 (0.22) 8.69 (0.61) 20.64 (1.22) 6.08 (0.68)

0.03 (0.78) 0.04 (0.50) 2.83** (3.06) 0.98* (1.95) 0.33 (1.01)

5.17*** (4.94) 1.56 (0.38) 3.80 (0.08) 45.12** (2.23) 9.83 (0.61)

0.29** (2.02) 0.39 (1.33) 1.11 (0.65) 1.57 (1.14) 0.54 (0.58)

2.90*** (5.47) 5.07*** (2.94) 27.48** (2.63) 3.83 (0.32) 7.92 (1.04)

CGI

4.48** (3.48)

1.04 (1.03) 0.01 (0.01) 6.56 (0.96) 0.04 (0.51) 0.01 (0.62) 3.75*** (10.98) 2.49** (2.15) 21.43*** (2.68) 4.98 (0.65) 3.40 (0.69) 0.258 216

0.655 33

0.184 33

0.484 75

0.305 75

0.798 25

0.083 25

0.500 83

0.322 83

This table presents two-stage least squares (2SLS) regression results with Tobin’s q and CGI score as the dependent variables. The sample consists of firms listed on the Stock Exchange of Thailand in 2005. The sample is split by family ownership, where low family ownership firms (Low FO) firms are firms where family and affiliated members hold less than 41.2% (sample median) of the outstanding shares respectively. High-family ownership (High FO) firms are firms where family and affiliated members own 41.2% or more of the outstanding shares. CGI is the percentage score from the corporate governance index based on the OECD Principles of Corporate Governance. Wedge dummy is a dummy variable with a value of one if the ratio of the cash flow rights divided by the voting rights is less than 1 and zero if the ratio is equal to 1. Tobin’s q is the sum of the book value of long-term debt plus the market value of equity divided by the book value of total assets. The average value of Tobin’s q from 2002 and 2004 is included as an explanatory variable. Board independence is the number of directors who are independent/outside directors divided by board size. Family ownership is the number of outstanding shares held by the founding family and affiliated members divided by the total number of shares outstanding. Firm size is the natural log of total assets. Firm age is the natural log of the years since the firm’s founding. Profitability (ROA) is the ratio of net income after taxes divided by total assets. Capital expenditures are the ratio of capital expenditures divided by total assets. Financial leverage is the ratio of long-term debt divided by total assets. t-Statistics are shown in parentheses. * Statistical significant differences at the 10% level (two-tailed) respectively. ** Statistical significant differences at the 5% level (two-tailed) respectively. *** Statistical significant differences at the 1% level (two-tailed) respectively.

5. Conclusion Thailand offers an interesting contrast to the US as an environment in which to study the interrelation among ownership structure, corporate governance and firm value. Most notably, firms in many emerging markets are characterized as operating in environments with weaker protection of minority shareholder rights and more concentrated ownership structures. In this type of environment, we expect that the agency conflicts between controlling and minority shareholders would be more severe than the conflicts observed in more developed markets. More importantly, regulators have been attempting to push forward corporate governance reforms with the hope that good corporate governance can benefit all shareholders. Consequently, many Thai companies have progressively adopted internationally accepted corporate governance practices. In this paper, we study the relation between the quality of corporate governance practices and firm value for Thai family firms and the potential mediating effect of complex ownership structures on the relation between corporate governance and firm value. To accomplish our objectives, we develop a comprehensive measure of corporate governance (the corporate governance index or CGI) appropriate to the Thai context. We establish that, unlike other conventional governance measures, our measure of effective corporate governance is significantly positively associated with

value for the full sample of family firms. The findings from our study highlight the importance of corporate governance in emerging markets. The findings also highlight the difficulty in obtaining relevant measures of good corporate governance in these unique economies which have very different institutional settings. In particular, conventional measures of corporate governance quality, such as board independence, are not particularly effective. Though, on the surface, good governance practices appear to be in place, the effectiveness of these practices can be blunted, for example by extra-contractual arrangements that often exist between families and board members. As a result, conventional governance variables capture only the form, not the substance, of the effectiveness of corporate governance practices among Thai family firms. The situation after the Asian financial crisis in 1997 also provides a unique setting as family firms have become more heterogeneous. After the crisis, a number of families sold their ownership stakes to outside investors or strategic partners. We present empirical evidence showing that Thai family firms employed more pyramidal structures after 1997, potentially to retain control of their firms despite the reductions in ownership levels. Pyramidal ownership structures in environments with weaker protection for minority shareholder rights can have a negative effect on firm value. Our empirical results confirm that the use of pyramidal ownership structures among Thai family firms negatively affects firm value. Furthermore, we find a moderating effect of the pyramidal structure

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

on the relation between the quality of corporate governance practices (CGI) and firm value. Specifically, the CGI shows a positive association with q only for firms that do not employ pyramidal control structures. We interpret this result as providing evidence that public companies in Thailand can improve corporate governance practices in an effort to enhance firm value. At the same time, our results also show that the presence of control-enhancing pyramidal ownership structures subverts the positive effect of good corporate governance practices. In summary, our results show that the link between governance practices and value appears visible only when looking at a comprehensive range of practices, not selective, narrower measures. In

1735

addition, the link is visible only in the absence of pyramidal ownership structures. Appearances can be deceiving, when owners adopt internationally accepted governance practices while also employing unscrupulous tactics, such as pyramidal ownership structures, to reassert their control. These tactics are detrimental to minority shareholders and ultimately to firm value. Finally, future studies linking corporate governance and firm value or performance should now look beyond the overall or prima facie relation between corporate governance practices and firm value. The reason is that there may be institutional or cultural factors in an economy which negate or nullify the effectiveness of the quality of corporate governance practices that are being implemented.

Appendix A Survey question

Scoring

I. Rights of shareholders A. Shareholder rights defined

Total of four items

References Total of 22 items; maximum score = 42 (25% of CGI) Bushman et al. (2004), La Porta et al. (1998), Mallin (2001), and Murphy (1999)

1. Offer other ownership rights beyond Score 2 if equitable share of profits and voting dividends and equitable treatment for share repurchases, 1 if only one right is offered, 0 if neither 2. Shareholders approve the Score 2 if approved, 0 if not remuneration annually 3. Presentation of board remuneration Score 2 if compensation details are to the shareholders provided for every director; 0 if only total/ summation provided 4. Shareholders can elect board Score 2 if yes, 0 if not members individually B. Shareholder rights disclosed

Total of eight items

Bhagat and Brickley (1984), Carcello and Neal (2000), Easterbrook (1984), Fama and Jensen (1983), Gillian and Starks (2000), Gordon and Pound (1993), Jensen (1986), Jensen and Meckling (1976), Karpoff et al. (1996), Klein (2002), Krishnan (2005), Raghunandan and Rama (2003), and Rozeff (1982)

1. Quality of notice to call shareholders’ meeting(s) (a) Appointment of directors Score 2 if names and backgrounds are provided, 1 if only one item is provided, 0 if both items are missing (b) Appointment of auditors Score 2 if name(s), profile, and fees are provided, 1 if 2 items are provided, 0 if one item or none provided (c) Dividend policy amount and Score 2 if both items are provided, 1 if only explanation for payment one item is provided, 0 if both items missing (d) Objective and reason for each Score 2 if included, 0 if omitted item on the shareholders’ meeting agenda (e) Director’s comments and opinion Score 2 if included, 0 if omitted for each agenda item 2. Quality of minutes of shareholders’ meeting(s) (a) Voting method and vote counting Score 2 if declared, 0 if not system declared before the AGM begins Score 2 if both items are included, 1 if time (b) AGM minutes show an for questions is allotted but answers/issues opportunity for shareholders to ask not recorded, 0 if both items are missing questions/ raise issues during the past year, along with a record of questions (continued on next page)

1736

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Appendix A (continued) Survey question

Scoring

and answers (c) Minutes show voting results for each agenda item, including both ‘‘for’’ and ‘‘against’’ vote tallies

Score 2 if both items included, 1 if only one item is shown, 0 if missing

C. Shareholder participation in AGM

Total of seven items

References

Ferris et al. (2003), Fich and Shivdasani (2005), Gillian and Starks (2000), and Karpoff et al. (1996)

1. Names of attending board members Score 2 if recorded, 0 if not recorded in the AGM minutes 2. Attendance by chairman of the board Score 2 if chairman attended the last two AGMs; 1 if attended only one meeting; 0 if not attending either 3. Attendance by CEO/Managing Score 2 if chairman attended the last two Director/President (top executive officer) AGMs; 1 if attended only one meeting; 0 if attended the last two AGMs not attending either 4. Attendance by chairman of the audit Score 2 if chairman attended the last two committee AGMs; 1 if attended only one meeting; 0 if not attending either 5. Attendance by chairman of the Score 2 if chairman attended the last two Compensation/Remuneration Committee AGMs; 1 if attended only one meeting; 0 if not attending either 6. Attendance by chairman of the Score 2 if chairman attended the last two Nomination Committee AGMs; 1 if attended only one meeting; 0 if not attending either 7. Additional AGM/EGM agenda item(s) Score penalty of 0 if no items included; 1 included in the meeting but omitted (penalty) if included from the meeting notice D. Takeover rules and anti-takeover defenses

1. Cross shareholding apparent

2. Pyramid holding apparent

3. Board members holdings

Total of three items

Score 2 if no apparent cross-holding, 1 if cross-holdings are likely; 0 if obvious evidence of cross-holding Score 2 if no evidence of pyramidal structure; 1 if pyramid shareholding is likely; 0 if obvious evidence of pyramiding Score 2 if directors in total hold more than 25% of the outstanding shares; 0 if not

II. Treatment of shareholders A. Voting rights for shares

Bhagat and Brickley (1984), Claessens et al. (2002, 2000), Jensen and Meckling (1976), La Porta et al. (1999), McConnell and Servaes (1990), Morck et al. (1988), and Shleifer and Vishny (1986)

Total of three items

Total of 13 items; maximum score = 24 (15% of CGI) Bhagat and Brickley (1984), Givoly and Palmon (1985), Grossman and Hart (1988), La Porta et al. (1997, 1998)

1. Voting rights for shares

Score 2 if only one class of share with oneshare, one-vote; 1 if more than one class of shares has higher, but not excessive, voting rights; 0 if voting rights are excessive, e.g., 50% or more voting rights per 10% of capital 2. Minority shareholders can influence Score 2 if mechanism is offered; 0 if none board composition 3. Cumulative voting used to elect Score 2 if offered (bonus); 0 if not board members B. Shareholder conflict

Total of six items

Cheung et al. (2006), Friedman et al. (2003), Johnson et al. (2000), La Porta et al. (1997, 1998)

1737

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Appendix A (continued) Survey question 1. System established to prevent the use of material inside information and inform all employees, managers, and board members 2. Insider trading cases involving company directors and/or managers in the past two years 3. Raionale/explanation offered for related-party transactions affecting the corporation before conducting relatedparty transactions that require shareholders’ approval 4. Non-compliance case regarding related-party transactions in the past two years 5. Level of business interconnections

6. Related-party transactions to nonsubsidiary companies

C. Proxy voting

1. Proxy voting facilitated

2. Shareholders know the documents required to give proxy 3. Notarization requirement for proxy appointment D. AGM procedures 1. Advance notice of the AGM

Scoring

References

Score 2 if system is established; 0 if not

Score 2 if no instance; 0 if one or more instances Score 2 if no related-party transactions were observed or if company provides full disclosure (name, relationship, policy, value of transaction, and board opinion); 1 if some but not all information is provided; 0 if no rationale provided for transaction(s) Score 2 if no non-compliance cases; 1 if company received a disclosure waiver from the exchange and/or regulator; 0 if noncompliance cases exist Score 2 for lowest level of interconnections;1 for moderate level; 0 for highest level of interconnections Score 0 if no transactions that could be considered as financial assistance to nonsubsidiary companies; 1 (penalty) if transaction(s) exists Total of three items

Brickley (1986), La Porta et al. (1997, 1998), Maug and Rydqvist (2001), and Pound (1991)

Score 2 if proxy voting forms are sent to shareholders along with the AGM notice; 0 if not Score 2 if the AGM notice specifies the documents required; 0 if not Score 2 if appointments are not required to be notarized; 0 if notarization is needed Total of one item Score 2 if shareholders receive notice 30 days or more before the meeting; 1 if 21–30 days’ notice is given; 0 if less than 21 days

III. Role of stakeholders A. Safety and welfare policy/benefits of employees

Score 0.67 if explicitly mentioned with comprehensive coverage; 0.33 if only superficial coverage given, 0 if not mentioned

B. Provident fund/retirement fund provided for its employees

Score 0.67 if provided; 0.33 if not

C. Professional development training programs for employees

Score 0.67 if explicitly mentioned with comprehensive coverage; 0.33 if only superficial coverage given; 0 if not mentioned

D. Role of customers

Score 2 if explicitly mentioned with comprehensive coverage; 1 if only superficial coverage given; 0 if not mentioned

E. Environmental issues

Score 2 if explicitly mentioned, with standards and explanation (e.g., ISO

Total of nine items; maximum score = 14 (10% of CGI) Berman et al., 1999; Connelly and Limpaphayom 2004 and La Porta et al. (1997, 1998)

(continued on next page)

1738

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Appendix A (continued) Survey question

Scoring

References

14000); 1 if disclosure only to the extent required by law; 0 if not mentioned F. Role of suppliers/business partners

Score 2 if explicitly mentioned with comprehensive coverage; 1 if only superficial coverage given; 0 if not mentioned

G. Obligations to shareholders

Score 2 if explicitly mentioned with comprehensive coverage; 1 if only superficial coverage given; 0 if not mentioned

H. Broader obligations to society and/or the community

Score 2 if explicitly mentioned with comprehensive coverage; 1 if only superficial coverage given; 0 if not mentioned

I. Obligations to creditors

Score 2 if explicitly mentioned with comprehensive coverage, 1 if only superficial coverage given, 0 if not mentioned

IV. Disclosure and transparency A. Disclosure of material information

Total of four items

Total of 32 items; maximum score = 40 (25% of CGI) Bushman et al. (2004), Claessens et al. (2002), Himmelberg et al. (1999), La Porta et al. (1998, 1999), and Mallette and Fowler (1992)

Transparency of the ownership structure 1. Breakdown of shareholding structure Score 2 if provided; 0 if not 2. Beneficial ownership Score 2 if easily identified; 1 if shares held by nominees or holding companies total less than 15percent; 0 if shares held by nominees or holding companies total more than 15 percent 3. Directors’ shareholdings Score 2 if disclosed; 0 if not 4. Management shareholdings Score 2 if disclosed; 0 if not B. Quality of the annual report. Does the Total of eight items report include:

1. Financial performance

Score 2 if clear, comprehensive, and informative; 1 if superficial; 0 if not available 2. Business operations and competitive Score 2 if clear, comprehensive, and position informative; 1 if superficial; 0 if not available 3. Operating risks Score 2 if clear, comprehensive, and informative; 1 if superficial; 0 if not available 4. Board member background Score 2 if full coverage with detailed background; 1 if limited to a few items; 0 if not available 5. Identification of independent Score 2 if identified; 0 if not available directors 6. Basis of board remuneration Score 2 if detailed compensation provided for each director; 1 if superficial or compensation shown in aggregate; 0 if not available 7. Disclosure of individual directors’ Score 2 if detailed compensation provided

Boyd (1994), Bushman et al. (2004), Ferris et al. (2003), Fich and Shivdasani (2005), Meek et al. (1995), Ryan and Wiggins (2004), and Singhvi and Desai (1971)

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

1739

Appendix A (continued) Survey question remuneration

8. Board meeting attendance of individual directors

C. External disclosure

Scoring

References

for each director; 1 if superficial or compensation shown in aggregate; 0 if not available Score 2 if detailed attendance record provided for each director; 1 if meeting attendance is listed without breakdown by director; 0 if not available Total of 20 items

Ashbaugh et al. (1999), Bushman et al. (2004), Cheung et al. (2006), Fan and Wong (2005), Farragher et al. (1994), Gregory et al. (1994), Hillier and Marshall (2002), Johnson et al. (2000), Lang and Lundholm (1993, 1996), and La Porta et al. (1997, 1998)

1. Public communications of relatedparty transactions

Score 2 if no related-party transactions were observed or if company provides full disclosure (name, relationship, policy, value of transaction, and board opinion); 1 if some but not all information is provided; 0 if no information provided 2. Specific policy requiring directors to Score 2 if a specific policy exists; 0 if policy report their transactions of company does not exist or disclosure is only required shares of managers 3. Annual audit performed using Score 2 if reputable, recognized auditors are independent and reputable auditors used; 1 if auditor is not approved by the exchange; 0 if auditor is not disclosed or affiliated with the company Score 2 if an unqualified opinion; 1 if an 4. Accounting qualifications in the audited financial statements (other than unqualified opinion with special mention items; 0 if a qualified opinion the qualification on Uncertainty of Situation) D. Are multiple channels used to provide access to information? 1. Annual report 2. Company website 3. Analyst briefing(s) 4. Press conference(s)/press briefing(s) 5. Timely disclosure of financial reports during the past 3 years 6. Contents of the company website with up-to-date information: (a) Business operations (b) Financial statements (c) Press releases (d) Shareholding structure (e) Organization structure (f) Corporate group structure, if applicable (g) Downloadable annual report (h) Notice to call shareholders’ meeting (i) Dual-language website E. Contact details provided for a specific Investor Relations person or unit

Score 0.5 if used; 0 if not Score 0.5 if used; 0 if not Score 0.5 if used; 0 if not Score 0.5 if used; 0 if not Score 2 if meeting deadlines every time, 1 if two or fewer delays, 0 if more than two delays

Score Score Score Score Score Score

0.22 0.22 0.22 0.22 0.22 0.22

if if if if if if

used; used; used; used; used; used;

0 0 0 0 0 0

if if if if if if

not not not not not not

Score 0.22 if used; 0 if not Score 0.22 if used; 0 if not Score 0.22 if used; 0 if not Score 2 if provided; 0 if not

F. Regulatory sanctions required revision of Score 0 if no sanctions made or revisions financial statements required during the past year; 1 (penalty) if company was sanctioned (continued on next page)

1740

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

Appendix A (continued) Survey question

Scoring

V. Board responsibilities A. Index of board monitoring/control efforts

Total items = 21

1. Written corporate governance rules describing value system and board responsibilities 2. Board of directors provides a code of ethics or statement of business conduct for all directors and employees; Board ensures all are aware of and understand the code 3. Corporate vision/mission 4. Incidences of regulatory of noncompliance during the past year

Score 2 if rules are board approved and disclosed; 1 if rules exist but have not been approved; 0 if no rules Score 2 if code exists and is effectively communicated; 1 if code exists; 0 if no code exists

5. Internal audit function

6. Line of reporting for internal audit function 7. Quality of the audit committee report in the annual report, containing the following key items: (a) Attendance (b) Internal control (c) Management control (d) Proposed auditors (e) Financial report review (f) Legal compliance (g) Overall concluding opinion 8. Orientation for new directors

Score 2 if present; 0 if not Score 2 of no cases of non-compliance with exchange or regulatory rules; 1 if one case; 0 if two or more cases or one serious offense case Score 2 if a separate unit in the company, 1 if internal audit function was outsourced; 0 if no internal audit function exists Score 2 if reporting to the board audit committee; 0 if reporting to operating management only

Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 0.286 if available; 0 if not Score 2 if provided, with evidence of implementation; 0 if not or no evidence provided 9. Board member training Score 2 if directors have participated in professional/accredited directors’ training; 0 if not 10. Board meeting frequency Score 2 if the board met more than four times in 2005 and more than two times in 2004; 1 if the board met four times in 2005 and two times in 2004; 0 if the board met less than four times in 2005 and once in 2004 11. Attendance of board members Score 2 if greater than 80% average attendance during the past 12 months; 1 if 70–80% average attendance; 0 if below 70 percent. 12. Risk management policy Score 2 if provided; 0 if not 13. Clear distinction between the roles, Score 2 if both board and management duties, and responsibilities of the board roles are delineated; 0 if not and management

References

Total items = 41; maximum score = 50 (30% of CGI) Adams (1994), Boyd (1994), Carcello et al. (2002), Daily et al. (1998), Ferris et al. (2003), Fich and Shivdasani (2005), Ingley and van der Walt (2002), La Porta et al. (1997, 1998), Raghunandan and Rama (2003), Scarbrough et al. (1998), Turpin and DeZoort (1998), Vafeas (1999), and Weller (1988)

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

1741

Appendix A (continued) Survey question

Scoring

References

14. Annual board self-assessment

Score 2 if conducted and documented; 0 if not or undocumented 15. Annual performance assessment of Score 2 if conducted and documented; 1 if CEO/MD/President not or undocumented B. Assessment of conflicts of interest 1. Chairman independence

Total items = 1 Score 2 if the chairman is an independent director; 0 if not

Coles and Hesterly (2000)

C. Assessment of use of independent board committees with independent members

Total items = 15

Bostock (1995), Brick et al. (2006), Carcello et al. (2002), Carcello and Neal (2000), Daily et al. (1998), Klein (1998, 2002), and Krishnan (2005)

1. Presence of an audit committee, including the following items: (a) Charter/Role and responsibilities (b) Profile/Qualifications (c) Independence (d) Performance/Meeting Attendance record 2. Presence of a Compensation/ Remuneration Committee, including the following items: (a) Charter/Role and responsibilities (b) Committee composition

Score 2 if present; 0 if missing Score Score Score Score

0.5 0.5 0.5 0.5

if if if if

present; present; present; present;

0 0 0 0

if if if if

missing missing missing missing

Score 2 if present; 0 if missing

Score 0.5 if present; 0 if missing Score 0.5 if composed of a majority of independent directors; 0 if not. (c) Committee chairman Score 0.5 if an independent director; 0 if independence not. (d) Performance/Meeting attendance Score 0.5 if present; 0 if missing record 3. Presence of a Nomination Score 2 if present; 0 if missing Committee, including the following items: (a) Charter/Role and responsibilities Score 0.5 if present; 0 if missing (b) Committee composition Score 0.5 if composed of a majority of independent directors; 0 if not. (c) Committee chairman Score 0.5 if an independent director; 0 if independence not. (d) Performance/Meeting attendance Score 0.5 if present; 0 if missing

D. Definition of board independence

Total items = 1

Beasley (1996), and Mallette and Fowler (1992)

1. ‘Director independence’ defined in public communications

Score 2 if defined; 0 if not

E. Assessment of communication 1. Separate Board of Director’s report issued, describing the board’s responsibilities in reviewing the firm’s financial statements

Total items = 1 Score 2 if the report is issued; 0 if not

Beasley (1996)

F. Management incentive scheme

Total items = 1

Core and Guay (2001), DeFusco et al. (1990), and Yermack (1995)

1. Incentive for top management through option scheme

Score 2 if exercise period over is over three years and exercise price(s) are above the market value at the time of the award; 0 if not or no option scheme

G. Regulatory compliance 1. Non-compliance cases

Total items = 1 0 if no cases were serious offense during the past year; 1 (penalty) if otherwise

La Porta et al. (1997, 1998)

1742

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743

References Adams, M.B., 1994. Agency theory and the internal audit. Managerial Auditing Journal 9, 8–12. Agrawal, A., Nagarajan, N.J., 1990. Corporate capital structure, agency costs and ownership control: the case of all equity firms. Journal of Finance 45, 1325– 1331. Allen, F., Carletti, E., Marquez, R.S., 2009. Stakeholder capitalism, corporate governance and firm value. Working Paper, University of Pennsylvania. Anderson, R.C., Reeb, D.M., 2003. Founding family and firm performance. Evidence from the S&P 500. Journal of Finance 58, 1301–1328. Anderson, R.C., Duru, A., Reeb, D.M., 2009. Founders, heirs, and corporate opacity in the United States. Journal of Financial Economics 92, 205–222. Ashbaugh, H., Johnstone, K.M., Warfield, T.D., 1999. Corporate reporting on the internet. Accounting Horizons 13, 241–257. Ball, R., 2001. Infrastructure requirements for an economically efficient system of public financial reporting an disclosure. In: Litan, R., Herring, R. (Eds.), Brookings-Wharton Papers on Financial Services. Brookings Institution Press, Washington, DC. Beasley, M.S., 1996. An empirical analysis of the relation between the board of director composition and financial statement fraud. The Accounting Review 71, 443–465. Bebchuk, L., Cohen, A., Ferrell, A., 2009. What matters in corporate governance? Review of Financial Studies 22, 783–827. Berman, S.L., Wicks, A.C., Kotha, S., Jones, T.M., 1999. Does stakeholder orientation matter? The relationship between stakeholder management models and firm financial performance. Academy of Management Journal 42, 488–506. Bertrand, M., Johnson, S., Samphantharak, K., Shoar, A., 2008. Mixing family with business: a study of Thai business groups and the families behind them. Journal of Financial Economics 88, 466–498. Bhagat, S., Brickley, J.A., 1984. Cumulative voting: the value of minority shareholder voting rights. Journal of Law and Economics 27, 339–365. Black, B., Jang, H., Kim, W., 2006a. Does corporate governance affect firm value? Evidence from Korea. Journal of Law, Economics, and Organization 22, 366–413. Black, B., Love, I., Rachinsky, A., 2006b. Corporate governance indices and firms’ market values: time-series evidence from Russia. Emerging Markets Review 7, 361–379. Black, B., Kim, W., Jang, H., Park, K.S., 2009. How corporate governance affects firm value: evidence on channels from Korea. Working Paper, University of Texas. Bostock, R., 1995. Company responses to Cadbury. Corporate Governance: An International Review 3, 72–77. Boyd, B.K., 1994. Board control and CEO compensation. Strategic Management Journal 15, 335–344. Brick, I.E., Palmon, O., Wald, J.K., 2006. CEO compensation, director compensation, and firm performance. Evidence of cronyism? Journal of Corporate Finance 12, 403–423. Brickley, J.A., 1986. Interpreting common stock returns around proxy statement disclosures and annual shareholder meetings. Journal of Financial and Quantitative Analysis 21, 343–349. Brown, L., Caylor, M., 2006. Corporate governance and firm operating performance. Journal of Accounting and Public Policy 25, 409–434. Bushman, R.M., Piotroski, J.D., Smith, A.J., 2004. What determines corporate transparency? Journal of Accounting Research 42, 207–252. Carcello, J.V., Neal, T.L., 2000. Audit committee composition and auditor reporting. The Accounting Review 75, 453–467. Carcello, J.V., Hermanson, D.R., Neal, T.L., 2002. Disclosures in audit committee charters and reports. Accounting Horizons 16, 291–304. Cheung, Y.L., Rau, R., Stouraitis, A., 2006. Tunneling, propping, and expropriation: evidence from connected party transactions in Hong Kong. Journal of Financial Economics 82, 343–386. Cheung, Y.L., Connelly, J.T., Limpaphayom, P., Zhou, L., 2007. Do investors really value corporate governance? Evidence from the Hong Kong market. Journal of International Financial Management and Accounting 18, 86–122. Cheung, Y.L., Jiang, P., Limpaphayom, P., Lu, T., 2010. Corporate governance in China: a step forward. European Financial Management 16, 94–123. Cheung, Y.L., Connelly, J.T., Jiang, P., Limpaphayom, P., 2011. Does corporate governance predict future performance? Evidence from Hong Kong. Financial Management 40, 159–197. Claessens, S., Djankov, S., Lang, L.H.P., 2000. The separation of ownership and control in East Asian corporations. Journal of Financial Economics 58, 81– 112. Claessens, S., Djankov, S., Fan, J., Lang, L.H.P., 2002. Disentangling the incentive and entrenchment effects of large shareholdings. Journal of Finance 57, 2741–2771. Credit Lyonnais Securities Asia (CLSA), 2002. Corporate governance watch: corporate governance in emerging markets (February). Research & Sales Offices, CLSA Ltd. Coles, J.W., Hesterly, W.S., 2000. Independence of the chairman and board composition: firm choices and shareholder value. Journal of Management 26, 195–214. Connelly, J.T., Limpaphayom, P., 2004. Environmental disclosure and firm performance. Evidence from Thailand. Journal of Corporate Citizenship 13, 137–149. Core, J.E., Guay, W.R., 2001. Stock option plans for non-executive employees. Journal of Financial Economics 61, 253–287.

Daily, C.M., Johnson, J.L., Ellstrand, A.E., Dalton, D.R., 1998. Compensation committee composition as a determinant of CEO compensation. Academy of Management Journal 41, 209–220. DeFusco, R.A., Johnson, R.R., Zorn, T.S., 1990. The effect of executive stock option plans on stockholders and bondholders. Journal of Finance 45, 617–627. Demsetz, H., Lehn, K., 1985. The structure of corporate ownership: causes and consequences. Journal of Political Economy 93, 1155–1177. Doidge, C., Karolyi, G.A., Stulz, R., 2007. Why do countries matter so much for corporate governance? Journal of Financial Economics 86, 1–39. Drobetz, W., Schillhofer, A., Zimmerman, H., 2003. Corporate governance and expected stock returns: evidence from Germany. European Financial Management 10, 267–293. Durnev, A., Kim, E.H., 2005. To steal or not to steal: firm attributes, legal environment, and valuation. Journal of Finance 60, 1461–1493. Easterbrook, F.H., 1984. Two agency-cost explanations of dividends. The American Economic Review 74, 650–659. Fama, E., Jensen, M.E., 1983. Separation of ownership and control. The Journal of Law and Economics 26, 301–325. Fan, J.P.H., Wong, T.J., 2005. Do external auditors perform a corporate governance role in emerging markets? Evidence from East Asia. Journal of Accounting Research 43, 35–72. Farragher, E.J., Kleiman, R., Bazaz, M.S., 1994. Do investor relations make a difference? The Quarterly Review of Economics and Finance 34, 403–412. Ferris, S.P., Jagannathan, M., Pritchard, A.C., 2003. Too busy to mind the business? Monitoring by directors with multiple board appointments. Journal of Finance 58, 1087–1111. Fich, E., Shivdasani, A., 2005. Are busy boards effective monitors? Journal of Finance 61, 689–724. Friedman, E., Johnson, S., Mitton, T., 2003. Propping and tunneling. Journal of Comparative Economics 31, 732–750. Gillian, S.L., Starks, L.T., 2000. Corporate governance proposals and shareholder activism: the role of institutional investors. Journal of Financial Economics 57, 275–305. Givoly, D., Palmon, D., 1985. Insider trading and the exploitation of inside information: some empirical evidence. Journal of Business 58, 69–87. Gompers, P., Ishii, J., Metrick, A., 2003. Corporate governance and equity prices. Quarterly Journal of Economics 118, 107–155. Gordon, L.A., Pound, J., 1993. Information, ownership structure, and shareholder voting: evidence from shareholder-sponsored corporate governance proposals. Journal of Finance 48, 697–718. Gregory, A., Matatko, J., Tonks, I., Purkis, R., 1994. UK directors’ trading: the impact of dealings in smaller firms. The Economic Journal 104, 37–53. Grossman, S., Hart, O., 1988. One share-one vote and the market for corporate control. Journal of Financial Economics 20, 175–202. Gugler, K., 2003. Corporate governance, dividend payout policy, and the interrelation between dividends, R&D, and capital investment. Journal of Banking & Finance 27, 1297–1321. Harris, M., Raviv, A., 1988. Corporate governance. Voting rights and majority rules. Journal of Financial Economics 20, 203–235. Healy, P.M., Palepu, K.G., 2001. Information asymmetry, corporate disclosure, and the capital markets: a review of the empirical disclosure literature. Journal of Accounting and Economics 31, 405–440. Hillier, D., Marshall, A.P., 2002. The market evaluation of information in directors’ trades. Journal of Business Finance & Accounting 29, 77–110. Himmelberg, C.P., Hubbard, A.G., Palia, D., 1999. Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics 53, 353–384. Hwang, B., Kim, S., 2009. It pays to have friends. Journal of Financial Economics 93, 138–158. Ingley, C., van der Walt, N., 2002. Board dynamics and the politics of appraisal. Corporate Governance: An International Review 10, 163–174. James, H., 1999. Owner as manager, extended horizons and the family firm. International Journal of the Economics of Business 6, 41–56. Jensen, M.C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76, 323–329. Jensen, M.C., 2002. Value maximization, stakeholder theory, and the corporate objective function. Business Ethics Quarterly 12, 235. Jensen, M.C., Meckling, W., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 13, 305–360. Johnson, B.W., Magee, R.P., Nagarajan, N.J., Newman, H.A., 1985. An analysis of the stock price reaction to sudden executive deaths: implications for the managerial labor market. Journal of Accounting and Economics 7, 151–174. Johnson, S., La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2000. Tunneling. American Economic Review 90, 22–27. Karpoff, J.M., Malatesta, P.H., Walkling, R.A., 1996. Corporate governance and shareholder initiatives: empirical evidence. Journal of Financial Economics 42, 365–395. Khanna, T., Palepu, K.G., Srinivasan, S., 2004. Disclosure practices of foreign companies interacting with US markets. Journal of Accounting Research 42, 475–508. Khantavit, A., Polsiri, P., Wiwattanakantang, Y., 2004. Did families lose or gain control after the East Asian financial crisis? Evidence from Thailand. In: Fan, J., Hanazaki, M., Teranishi, J. (Eds.), Designing Financial Systems in East Asia and Japan: Toward a Twenty-First Century Paradigm. Routledge, London.

J.T. Connelly et al. / Journal of Banking & Finance 36 (2012) 1722–1743 Klapper, F., Love, I., 2003. Corporate governance, investor protection, and performance in emerging markets. Journal of Corporate Finance 195, 1–26. Klein, A., 1998. Firm performance and board committee structure. Journal of Law and Economics 41, 275–303. Klein, A., 2002. Economic determinants of audit committee independence. The Accounting Review 77, 435–452. Krishnan, J., 2005. Audit committee quality and internal control: an empirical analysis. The Accounting Review 80, 649–675. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 1997. Legal determinants of external finance. Journal of Finance 52, 1131–1150. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 1998. Law and finance. Journal of Political Economy 106, 1113–1155. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 1999. Corporate ownership around the world. Journal of Finance 54, 471–517. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 2000a. Agency problems and dividend policies around the world. Journal of Finance 55, 1–33. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 2000b. Investor protection and corporate governance. Journal of Financial Economics 58, 3–27. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 2002. Investor protection and corporate valuation. Journal of Finance 3, 1147–1170. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2006. What works in securities laws? Journal of Finance 51, 1–32. Lang, M., Lundholm, R., 1993. Cross-sectional determinants of analyst ratings of corporate disclosures. Journal of Accounting Research 31, 246–271. Lang, M.H., Lundholm, R.J., 1996. Corporate disclosure policy and analyst behavior. Accounting Review 71, 467–492. Lehn, K., Patro, S., Zhao, M., 2007. Governance indexes and valuation: which causes which? Journal of Corporate Finance 13, 907–928. Limpaphayom, P., Connelly, J.T., 2004. Thailand country report. In: Nam, S.W., Nam, I.C. (Eds.), Corporate Governance in Asia: Recent Evidence from Indonesia, Republic of Korea, Malaysia, and Thailand. Asian Development Bank Institute, Tokyo. Lins, K.V., 2003. Equity ownership and firm value in emerging markets. Journal of Financial and Quantitative Analysis 38, 159–184. Mallette, P., Fowler, K.L., 1992. Effects of board composition and stock ownership on the adoption of ‘‘poison pills’’. Academy of Management Journal 35, 1010–1035. Mallin, C., 2001. Institutional investors and voting practices: an international comparison. Corporate Governance: An International Review 9, 118–126. Maug, E., Rydqvist, K., 2001. What is the function of the shareholder meeting? Evidence from the US proxy voting process. Working Paper, Humboldt University of Berlin. McConnell, J., Servaes, H., 1990. Additional evidence on equity ownership and corporate value. Journal of Financial Economics 27, 595–612. Meek, G.K., Roberts, C.B., Gray, S.J., 1995. Factors influencing voluntary annual report disclosures by US, UK, and continental European multinational corporations. Journal of International Business Studies 26, 555–572. Mitton, T., 2002. A cross-firm analysis of the impact of corporate governance on the East Asian financial crisis. Journal of Financial Economics 64, 215–241. Mitton, T., 2004. Corporate governance and dividend policy in emerging markets. Emerging Markets Review 5, 409–426. Morck, R., Shleifer, A., Vishny, R., 1988. Management ownership and market valuation. Journal of Financial Economics 20, 293–315. Murphy, K.J., 1999. Executive compensation. In: Ashenfelter, O., Card, D. (Eds.), Handbook of Labor Economics, vol. 3b. Elsevier, Amsterdam.

1743

Nam, S.W., Nam, I.C., 2004. Corporate Governance in Asia: Recent Evidence from Indonesia, Republic of Korea, Thailand, and Malaysia. Asian Development Bank Institute, Tokyo. OECD (Organization of Economic Co-operation and Development), 1999. Principles of Corporate Governance. OECD, Paris. OECD (Organization of Economic Co-operation and Development), 2004. Principles of Corporate Governance. OECD, Paris. Perez-Gonzalez, F., 2006. Inherited control and firm performance. American Economic Review 96, 1559–1588. Pound, J., 1991. Proxy voting and the SEC: investor protection versus market efficiency. Journal of Financial Economics 29, 241–285. Raghunandan, K., Rama, D.V., 2003. Audit committee composition and shareholder actions: evidence from voting on auditor ratification. Auditing: A Journal of Practice & Theory 22, 253–263. Romano, R., 2005. The Sarbanes-Oxley Act and the making of quack corporate governance. Yale Law Journal 114, 1521–1611. Rozeff, M., 1982. Growth, beta and agency costs as determinants of dividend payout ratios. Journal of Financial Research 5, 249–259. Ryan, H.E., Wiggins, R.A., 2004. Who is in whose pocket? Director compensation, board independence, and barriers to effective monitoring. Journal of Financial Economics 73, 497–524. Scarbrough, D.P., Rama, D.V., Raghunandan, K., 1998. Audit committee composition and interaction with internal auditing: Canadian evidence. Accounting Horizons 12, 51–62. Sersansia, S., Nimmansomboon, W., 1996. Public Limited Companies Act BE 2535. Nititham Publishing House, Bangkok. Shleifer, A., Vishny, R.W., 1986. Large shareholders and corporate control. The Journal of Political Economy 94, 461–488. Shleifer, A., Vishny, R.W., 1997. A survey of corporate governance. Journal of Finance 52, 737–783. Singhvi, S.S., Desai, H.B., 1971. An empirical analysis of the quality of corporate financial disclosure. The Accounting Review 46, 129–138. Stein, J., 1989. Efficient capital markets, inefficient firms: a model of myopic corporate behavior. Quarterly Journal of Economics 103, 655–669. Studwell, J., 2007. Ties that bind. Newsweek 23, 24–33, July. Turpin, R.A., DeZoort, F.T., 1998. Characteristics of firms that include an audit committee report in their annual report. International Journal of Auditing 2, 35– 48. Vafeas, N., 1999. Board meeting frequency and firm performance. Journal of Financial Economics 53, 113–142. Villalonga, B., Amit, R., 2006. How do family ownership control and management affect firm value? Journal of Financial Economics 80, 385–417. Weller, S., 1988. The effectiveness of corporate codes of ethics. Journal of Business Ethics 7, 389–395. Wiwattanakantang, Y., 2001. Controlling shareholders and corporate value: evidence from Thailand. Pacific-Basin Finance Journal 9, 323–362. Yermack, D., 1995. Do corporations award CEO stock options effectively? Journal of Financial Economics 39, 237–269. Yermack, D., 1996. Higher market valuation of companies with a small board of directors. Journal of Financial Economics 40, 185–211. Zhuang, J., Edwards, D., Webb, D., Capulong, M.V., 2000. Corporate governance and finance in East Asia. Asian Development Bank, Manila.