CORPORATE GOVERNANCE, LEGAL ENVIRONMENT, FIRM VALUATION AND FINANCIAL PERFORMANCE IN SOUTH-EAST ASIA M

CORPORATE GOVERNANCE, LEGAL ENVIRONMENT, FIRM

VALUATION AND FINANCIAL PERFORMANCE IN SOUTH-EAST

ASIA

M

M

S

I

E

&

B

F

E

B

U

G

A

Boudewijn J. Quick

T

Prof. Dr. H. van Ees

1

A

2

T

1. INTRODUCTION 4

2. THEORY AND HYPOTHESES 8

2.1 Research in the U.S. 8

2.1.1 Internal corporate governance mechanisms and financial 8 performance and valuation

2.1.2External corporate governance mechanisms and financial

performance and valuation 9

2.2 Research in South-East Asia and emerging markets worldwide 11 2.2.1 The legal system, investor protection and its influence on

corporate governance 11

2.2.2 Corporate governance and firm valuation and financial

valuation 13

2.2.3 Country-level legal environment and firm valuation and

financial performance 15

2.2.4 Corporate governance and legal environment:

complementary or substitutionary? 17

2.2.5 State predation, repression and firm valuation and

firm valuation and financial performance 19 2.2.6 Financial and economic development and firm

valuation and financial performance 20

2.2.7 Firm-level control variables 21

2.3 Determinants of governance 23

2.3.1 Country-level variable hypotheses 23

2.3.2 Firm-level variable hypotheses 24

3. DATA AND METHOD 26

3.1 Data 26

3

3.1.2 Dependent variables 26

3.1.3 Independent country-level variables 27

3.1.4 Independent firm-level variables 28

3.2 Method 30

4. METHODOLOGY IN RELATED LITERATURE 33

5. RESULTS 36

5.1Determinants of corporate governance 36

5.1.1 Descriptive statistics determinants of corporate governance 36 5.1.2 Analytical results determinants of corporate governance 38

5.1.3 Diagnostic checks 44

5.2 Corporate governance, legal environment, firm valuation and financial

performance 46

5.2.1 Descriptive statistics corporate governance and firm valuation

and financial performance 46

5.2.2 Analytical results corporate governance and firm valuation 48 5.2.3 Analytical results corporate governance and firm financial

performance 54

5.2.4 Two-stage least squares and diagnostic checks 58

6. DISCUSSION ANDCONCLUSION 63

7. REFERENCES 69

4

1.

I

Broadly speaking, corporate governance is the system by which companies are directed and controlled (Cadbury Committee, 1992). More specifically, it deals with the separation of ownership and control of a firm. This separation is at the core of the well-known agency problem, also known as the principal-agent problem. Managers (agents) have incentives to expropriate a firm’s assets by undertaking projects that benefit themselves personally but impact shareholder (principal) wealth adversely (Brown & Caylor, 2004). Corporate governance encompasses a broad spectrum of mechanisms intended to mitigate agency problems by increasing the monitoring of managements’ actions, limiting managers’ opportunistic behavior, and reducing the information risk borne by shareholders (Ashbaugh et al., 2004). The term corporate governance has been catapulted from relative obscurity into the public consciousness due to a succession of crises, from the Asian financial crisis of 1997-1998, to corporate accounting scandals in the US and Europe at the beginning of the century, to the current financial crisis. It is widely acknowledged that these crises have at least partly been caused by deficiencies in corporate governance frameworks, both on the level of individual firms and on a country level.

A way of classifying corporate governance mechanisms is by characterizing them as being either internal or external to the firm. The first principal internal governance mechanism is the ownership structure of the firm, i.e. the identity of the shareholders and the size of their stake in the firm. In the US and the UK the typical firm has a widely dispersed ownership base with professional managers, unaffiliated with any shareholders, running the firm. In many other parts of the world, both in developed and developing countries, firms have private or public blockholders such as families, banks, institutional investors and government. These blockholders can influence the actions and decisions taken by management. The second principal internal governance mechanism is the board of directors. The board of directors has the authority to hire, fire, compensate and monitor management. In an Anglo-Saxon setting the goal of the board of directors is generally considered to be the maximization of shareholder value through the minimization of the loss of value associated with a separation between ownership and control. In other settings, such as continental Europe and Japan, the board tends to take into account a more holistic vision based on a more inclusive stakeholder approach.

5 well protected, such as in the US and the UK, equity capital markets are deeper and ownership is more widely dispersed. The market for corporate control can also serve as a powerful mechanism to reduce costs associated with the principal-agent problem. When internal control mechanisms fail to a large enough degree – i.e. when the gap between the actual value of a firm and its potential value is sufficiently large – there is an incentive for outside parties to seek control of the firm (Denis and McConnell, 2003). This disciplining mechanism is much more important in countries with a dispersed ownership base than in countries with firms which are controlled by blockholders.

Corporate governance mechanisms are essential in assuring that the suppliers of finance to corporations get a reasonable return on their investment (Shleifer & Vishny, 1997). One might therefore expect that there exists a clear and relatively easy to establish positive relationship between the quality and effectiveness of corporate governance on the one hand, and firm performance and valuation on the other. This is not the case, at least not in studies which have focused on firms operating in western, developed nations. Although prior work has provided some insight into the role of corporate governance, the results of similar studies are frequently contradictory and a consistent set of results is yet to emerge regarding the importance of corporate governance for understanding managerial behavior and organizational performance (Larcker et al., 2005).

Studies focused on firms operating in developing countries, and some developed countries, have exposed a different agency problem. In many emerging markets the central corporate governance problem is not between management and shareholders (principal-agent), but between majority and minority shareholders; the principal-principal problem. Poor investor protection afforded to small shareholders by the country legal system has led to blockholders who control the majority of firms in these countries. This makes sense; if the law does not protect the owners from the controllers, the owners will seek to be controllers (Denis & McConnell, 2003). In this case the owners of the company privately enforce property rights instead of the state publicly enforcing those rights. The problem in this situation is that often minority shareholders are expropriated by majority shareholders in a variety of ways.

6 shareholders is most likely and most egregious: during times of crisis. These results are consistent with a 2002 McKinsey Global Investor Opinion survey that measured the premium investors were willing to pay for a well governed company. It reports that this premium ranged from 11% for a well governed Canadian company, to 18% for a well governed UK or US company, to 41% for a well governed Moroccan company. It further stated that investors put corporate governance on a par with financial indicators when evaluating investment decisions.

Why is this the case? The political economy of a country, proxied by the country-level legal environment (which determines investor protection), level of state predation (shareholder expropriation) and political repression (level of autocracy) appears to be the key. The most apparent explanation for the premium placed on corporate governance in markets with weaker legal protection of investors is that good corporate governance is valued higher where it is scarce; in weaker legal regimes (Durnev & Kim, 2005). Since developing countries tend to have weaker legal environments and investor protection than developed nations (especially the US), this would indicate that good corporate governance matters more in weaker legal environments; firm-level governance and country-level investor protection are substitutionary. It is not immediately apparent however that this is indeed the case. Shareholders may prefer to invest in companies whose country of incorporation guarantees better protection in the eventuality of legal disputes, irrespective of the company corporate governance practices (Bruno & Claessens, 2007). There are a number of other papers that also dispute this notion of substitutionarity and assert that the relation between country-level legal environment and firm-country-level governance is complementary (Mitton (2004), Doidge et al. (2006)).

7 to financial- and valuation information on these firms. This information will be combined with data on the country-level legal environment, state expropriation and political repression, and financial and economic development of the countries in which these firms operate.

This thesis makes a number of contributions to the literature. The first is the use of the anti-selfdealing index as a proxy for the legal environment, developed by Djankov et al. (2008). This alternative to the often used anti-director index (La Porta et al., 1998) has only been used in one other paper (Durnev & Fauver, 2007) as far as I have been able to ascertain. Second, this thesis uses measures of state predation (government expropriation) and state repression of political and civil liberties to assess their effect on the relationship between corporate governance and firm performance. The concept of state predation was introduced by Stulz (2005), and used by Durnev & Fauver (2007), but the issue of how government expropriation activity influences the behavior of management and the relationship between governance and performance has otherwise been largely ignored. Third, on the debate whether corporate governance and legal environment are complementary or substitutionary, this thesis confirms that firm-level governance can substitute for deficiencies in investor protection provided by the legal system.

8

2.

This section discusses some of the literature with regard to corporate governance, legal environment and firm performance. Hypotheses are then developed from this literature. First, research focused on the U.S. is highlighted. Second, Section 2.2.1 briefly discusses the features of corporate governance in emerging markets, with particular attention for South-East Asia. Third, sections 2.2.2-2.2.7 develop hypotheses to test the relationships between corporate governance, country-level environment and firm financial performance and valuation; the main objective of this thesis. Fourth, section 2.3 provides hypotheses to investigate what the determinants of corporate governance are; the second aim of this thesis.

2.1 RESEARCH IN THE U.S.

2.1.1 INTERNAL CORPORATE GOVERNANCE MECHANISMS AND FIRM VALUATION AND FINANCIAL PERFORMANCE

Initial research into the impact of corporate governance on financial performance and valuation took place mostly in the US. A well functioning board consisting of a majority of independent directors who bring their judgment to bear on issues of strategy, performance and resources is thought to be one of the most important internal governance mechanisms (Cadbury Committee, 1992). Empirical evidence however is mixed, which is to say inconclusive.

Bhagat & Black (2002), in a large scale study of large American firms, find that firms with a more independent board do not perform better (financially of valuation wise) than other firms. Brown & Caylor (2004) find mixed evidence; board independence is actually negatively correlated with Tobin’s Q, but positively with other firm performance measures such as profit margins and return on equity. Hermalin & Weisbach (2003), in a review of the literature on US boards, also report mixed evidence. Higher proportions of outside directors are not associated with superior firm performance, but are associated with better decisions concerning such issues as acquisitions, executive compensation and CEO turnover. Ashbaugh et al. (2004), in a paper that studies the impact of corporate governance on the cost of equity, also document a positive impact of an independent board; they find a negative relation between the cost of equity and the independence of the board.

9 firms to establish that better governed firms are relatively more profitable, more valuable and pay out more cash to their shareholders. They show that executive and director compensation are most highly associated with superior firm performance. Additionally, Larcker et al. (2005) also find that compensation based on accounting measures of performance is positively associated with return on assets. In contrast Core et al. (2003), in a survey of the literature, state that despite considerable prior research the performance consequences of executive equity ownership remain open to question. Another important internal governance mechanism is the ownership structure of a firm. Traditionally, the typical public US firm is thought of as having widely dispersed ownership with professional managers (with small equity stakes, if any) at the helm. However, research has shown that approximately 20% of exchange-listed corporations in the US are owned by insiders (officers and directors) or blockholders (Holderness, 2003). These insiders or blockholders could have an incentive to increase their private benefits of control, to the detriment of the minority shareholders. Minority shareholders would price this expropriation into the value of the share, since the future cash flows to which they are entitled decrease. This then leads to a drop in firm value. Holderness (2003) concludes however that the impact of the presence of blockholders on firm value in sometimes positive, sometimes negative, and that there is little evidence that this impact is very pronounced. Ashbaugh et al. (2004) do find that consistent with potential rent extraction, concentrated ownership in the form of the percentage of shares held by institutions and the number of five-percent blockholders are positively related to the cost of equity.

2.1.2 EXTERNAL CORPORATE GOVERNANCE MECHANISMS AND FIRM VALUATION AND FINANCIAL PERFORMANCE

An essential external corporate governance mechanism in the US is the market for corporate control, although this mechanism is considered a ‘court of last resort’ for assets that are not being utilized to their full potential (Denis & McConnell, 2003). Thus, this instrument is often only used when the above mentioned governance channels fail to improve company performance, although the threat alone of a takeover might be enough to discipline and invigorate management.

10 provisions they believe to be the most important, and confirm that for the period 1990-2003 firms with weaker shareholder rights were associated with large negative abnormal returns. They also provide suggestive evidence that these firms are characterized by economically significant reductions in firm value. Nevertheless, when Core et al. (2006) repeat the Gompers et al. study for the years 2000-2003 they find that in this period the higher returns for firms with stronger shareholder rights do not materialize, and they conclude that the abnormal returns in the period 1990-1999 might in fact be explained by an omitted variable problem. They do however confirm that for 2000-2003 weak governance firms continue to have lower operating performance.

The second crucial external governance mechanism is the country-level legal system, specifically the laws protecting the property rights of minority shareholders and the enforcement of these laws. The protection of shareholders (and creditors) by the legal system is central to understanding the patterns of corporate finance in different countries (La Porta et al., 2000). When suppliers of finance feel like they are well protected by law they are willing to pay more for assets such as debt and equity because they realize that with better legal protection expropriation is limited and more of the firm’s profits would come back to them in the form of dividends and interest (La Porta et al., 2002). Of course, contrary to the governance mechanisms mentioned above this is a country-level governance issue rather than a firm-level one and thus impacts all firms incorporated in one country equally.

11 by now and has a significant impact on a financial development, ownership structure and some performance and valuation variables.

2.2RESEARCH IN SOUTH-EAST ASIA AND EMERGING MARKETS WORLDWIDE

2.2.1 THE LEGAL SYSTEM, INVESTOR PROTECTION AND ITS INFLUENCE ON CORPORATE GOVERNANCE

There exist two fundamental differences between governance systems in the US and governance systems in emerging markets. The first is the protection offered to shareholders by the legal system. The second is the ownership structure, which is a corollary of the strength of the legal system. Emerging markets generally suffer from a lack of shareholder and creditor protection and have poorly developed legal systems (Lins, 2003). As discussed above, this results in higher ownership concentration. Of course, initially there are majority owners in many companies. With the growth of the firm the need to go public arises. However, because of weak investor protection provided by the law, fewer smaller investors participate and capital markets remain underdeveloped. Blockholders then remain the norm in the ownership structure of many companies. Another way to see blockholding as a natural result of lack of shareholder protection in emerging market settings is as follows. If the law does not protect the owners from the controllers, the owners will seek to be controllers (Denis & McConnell, 2003). In this case the owners of the company privately enforce property rights instead of the state publicly enforcing those rights. This ownership structure also leads to a change in the parties that are engaged in an agency conflict. In the United States corporate governance mechanisms are generally designed to deal with agency problems between management and dispersed shareholders; the principal-agent problem. In emerging markets the major corporate governance issue is the agency problem between majority shareholders or blockholders (who often are, or control the management of the firm) and minority shareholders. This is known as the principal-principal problem.

12 Problems arise when there is a divergence between cash flow rights (ownership) and control rights. South-East Asian firms show a sharp divergence between these two rights; the controlling shareholder is often able to control a firm’s operations with a relatively small direct stake in its cash-flow rights through the use of pyramid structures, cross-holdings among firms and sometimes through dual-class shares (Claessens et al., 2002). This is known as the entrenchment effect. This divergence creates an incentive for the controlling shareholder to expropriate firm assets because he would receive the entire benefit but only bear a fraction of the cost through a lower valuation of his cash-flow ownership (Claessens & Fan, 2002). Some examples of expropriation are overcompensation, excessive consumption of perquisites, installation of family members in management positions, empire building, ‘tunneling’ through non-arms length transactions with affiliated companies and outright theft of company assets. When corporate governance issues such as expropriation are recognized by shareholders they discount the price of the share since the cash flow to which they are entitled decreases. This then leads to lower firm value, which means that controlling shareholders or managers also bear some of the costs associated with these governance problems. Financial performance can also be affected, for example because of direct and indirect costs associated with tunneling or self-dealing, or because the management installed by the controlling shareholder is not the most qualified. Since blockholding in the form of family ownership is common across Asian economies (Rajan & Zingales (1998), Bertrand et al. (2002), Claessens et al. (2002)), in Asia corporate governance is, to a large extent, a set of mechanisms through which outside investors protect themselves against expropriation by the insiders (La Porta et al., 2000).

13 2.2.2 CORPORATE GOVERNANCE AND FIRM VALUATION AND FINANCIAL PERFORMANCE

Claessens et al. (2002) evaluate the effect of ownership structure on firm valuation in eight East Asian economies1, and find that families are often the dominant shareholders in firms in emerging markets. Nearly 70% of the firms in their sample are family owned, and they find that divergence between cash flow rights and control rights is especially pronounced in firms owned by families. This ownership structure in combination with high control rights can have a direct impact on the effectiveness of traditional internal and external governance mechanisms. Majority shareholders often control the management of a company by virtue of their large equity stake alone or by installing themselves or family members on the board of directors. This leads to a board that lacks independence and that is beholden to the interests of the controlling shareholder at the expense of other shareholders. In such instances it is a weak governance mechanism unable to look after the interests of minority shareholders. Traditional board duties such as determining remuneration and succession and monitoring the auditing process are also compromised and rarely serve to align the interests of majority and minority shareholders.

Consistent with this Yeh et al. (2001) find that in Taiwan there is a positive valuation effect when controlling families hold less than 50% of a firm’s board seats. Lins (2003) finds that when a management group’s control rights exceed its cash flow rights firm values are lower in firms operating in emerging markets across the world. Claessens et al. (2002) confirm these results for firms in eight East Asian countries. They provide evidence that for the largest shareholders the difference between control rights and cash flow rights is associated with a value discount and that the discount generally increases with the size of the wedge between control rights and cash flow rights.

The flip side of this coin is that controlling shareholders are able to increase firm value by signaling to the equity market that they plan to limit their level of expropriation. They can do this either by increasing board independence as mentioned, but also by decreasing the wedge between cash flow (ownership) rights and control rights, either by increasing their ownership stake or by decreasing their control rights. Lins (2003) and Claessens et al. (2002) document evidence that this signal is indeed interpreted in this way by the market.

There are more ways for individual firms to use corporate governance mechanisms to distinguish themselves in the field of corporate governance when operating in an environment with weakly enforced shareholder rights. One way is through dividend policy, which decreases potential ‘free

14 cash flow’ and empire building problems by returning funds to shareholders. Mitton (2004) shows that firms from emerging markets worldwide with higher corporate governance ratings have higher dividend payouts, which is consistent with the agency model of dividends which predicts that when investors are better protected (more empowered) they use this power in the form of dividend extraction. He also finds a stronger negative relation between dividends and growth opportunities among firms with better governance scores. This would also be consistent with theory. If investors feel protected they worry less about rent extraction by insiders and would allow a firm with growth opportunities to use retained earnings to capitalize on these opportunities, instead of claiming the funds in the form of dividends. Furthermore, he documents that firms with stronger governance are more profitable.

Voluntary transparency and disclosure of company information can serve to reduce information asymmetry between majority shareholders or management and minority shareholders. Firms can improve the quantity and quality of information provided to investors on their own, but they can also employ external reputational agents (e.g. accountants and auditors) to do so. Lower transparency of firms results in a higher level of asymmetric information which provides greater opportunities for expropriation of minority shareholders by managers or controlling shareholders (Mitton, 2002). Unsurprisingly, public corporations in Asia typically have low levels of transparency and disclosure quality (Claessens & Fan, 2002). This often leads to reported earnings that are less informative, especially when controlling owners possess high voting rights and when voting rights substantially exceed cash flow rights (Fan & Wong, 2002). Low earnings informativeness works in conjunction with high voting rights to protect rent seeking activities by controlling shareholders. There is evidence however that individual firms are able to overcome information asymmetry and estimation risk issues by voluntarily adopting more stringent governance provisions. Chen et al. (2003) investigate the effect of voluntary adoption of disclosure and other corporate governance mechanisms on a firm’s cost of equity capital for firms operating in nine Asian emerging markets2_. Lower cost of capital and associated higher firm valuation is an important advantage for firms when adopting better governance provisions. The authors find that a high score on disclosure and a high score on non-disclosure corporate governance mechanisms are associated with a lower cost of equity capital. Investors pay a premium for firms with good corporate governance by discounting higher expected cash flows with a lower rate.

As mentioned, firms can also employ reputational agents such as accountants and auditors to improve the quality and quantity of information provided to investors. Firms can do this on a

15 voluntary basis, or be forced to do so via regulatory means. In Korea, government assigned external auditors were found to constrain the ability for income-increasing earnings management (Kim et al., 2002). With regard to voluntary retention of high quality external auditors, defined as the Big Five, Fan & Wong (2002) conclude that for eight Asian economies3 _{auditors play a crucial monitoring role} to mitigate agency problems in companies where these problems are most likely to occur.

Almost all emerging market studies surveyed so far focus on one or two specific corporate governance mechanisms. The more recent availability of composite corporate governance indices that measure a wide variety of governance provisions at the firm level, published by financial and shareholder institutions, have allowed researchers to explore the connection between overall firm-level corporate governance and firm value and financial performance. In Korea and Russia (Black et al. (2005), Black et al. (2006b), respectively) time-series evidence points to a positive relation between a number of firm-level governance measures and firm valuation. There have been a number of multi-country studies (Klapper & Love (2004), Durnev & Kim (2005), Bruno & Claessens (2007)) that have used other composite governance indices assessing such things as board structure and independence, separation of CEO and Chairman, transparency and accountability. They confirm a positive association between (components of) corporate governance and financial performance and valuation across a large number of countries and companies.

Hypothesis 1: Corporate governance quality is positively related to both firm value and firm financial performance.

2.2.3 COUNTRY-LEVEL LEGAL ENVIRONMENT AND FIRM VALUATION AND FINANCIAL PERFORMANCE

The quality of the legal system as a determinant of investor protection, and its subsequent overarching effects on the characteristics of corporate governance and the development of equity financing in emerging economies has been discussed above. Just as in the US, the kinds of laws protecting shareholders and, crucially, their enforcement, have a direct bearing on firm performance and valuation issues in South-East Asia.

Intuitively, the importance of the legal system in influencing company valuation and financial performance is straightforward. When their rights are better protected by the law, outside investors are willing to pay more for financial assets such as equity and debt. They pay more because they recognize that, with better legal protection, more of the firm’s profits would come back to them as dividends or interest as opposed to being expropriated by the entrepreneur who controls the firm

16 (La Porta et al., 2002). Additionally, similar to better firm-level governance, a better country-level legal system reduces expropriation opportunities for managers and blockholders and can prevent the installation of incompetent crony managers, thereby improving profitability.

La Porta et al. (1998, 2000, 2002) were influential in documenting the importance of country-level investor protection. They show that higher investor protection at the country level is associated with greater access to finance, higher capital markets development, and higher company valuation (Bruno & Claessens, 2007). Subsequent studies have provided mixed results. The focus here is on the micro effects, i.e. company valuation and profitability. Klapper & Love (2004) find that one measure of the legal environment is positively associated with firm valuation in emerging markets around the world, but that other measures are not. Durnev & Kim (2005) use data on firms operating in both emerging markets and developed nations. They also find that the strength of the legal framework and firm values are positively related. However, this positive relation becomes insignificant when firm-level governance variables are added to the regression. Bruno & Claessens (2007), using panel data on firms operating in advanced economies worldwide, provide more robust evidence that the country-level protection of minority shareholders is positively and significantly associated with company valuation.

South-East Asia also provides an interesting research setting for researchers to establish a possible relation between the quality of the legal environment, corporate governance and firm performance during a crisis. Crises generally negatively impact firms’ investment opportunities, raising the incentives of controlling shareholders to expropriate minority shareholders (Lemmon & Lins, 2003). When investor protection is lower, expropriation will be higher. In turn, this leads to capital flight out of countries affected by the crisis which are then more likely to see a collapse in currency and stock prices. Johnson et al. (2000) find that corporate governance in general, and the de facto protection of minority shareholder rights in particular, matters a great deal for the extent of exchange rate depreciation and stock market decline in 1997-1998. In countries with weaker investor protection net capital inflows were more sensitive to negative events that adversely affect investors’ confidence. Interestingly, in cross-country regressions the authors show that corporate governance variables explain more of the variation in exchange rates and stock market performance during the Asian crisis than do macroeconomic variables.

17 Hypothesis 2: The quality of investor protection afforded by the country-level legal environment is positively related to both firm value and firm financial performance

2.2.4 CORPORATE GOVERNANCE AND LEGAL ENVIRONMENT: COMPLEMENTARY OR SUBSTITUTIONARY?

In emerging markets there is wide variation in the quality of corporate governance, but the degree of variation is not systematically related to countries’ legal environment, i.e. there are well-governed firms in countries with weak legal systems and badly governed firms in countries with strong legal systems (Klapper & Love, 2004). However, there is a strong positive correlation between corporate governance and the legal environment; on average countries with better investor protection have firms with better governance. This would seem to indicate a complementary relationship between the two.

On the other hand, there is evidence of a substitutionary relationship. This can be seen as follows. As discussed in the introduction and section 2.1, there is a lack of consistent evidence of a relationship between corporate governance and performance in the US, a country with high investor protection. This positive relationship appears to be much stronger and consistent in studies which looked at firms operating in emerging economies with their commensurate weaker legal regimes. This can be interpreted as a premium that is being placed by investors on firms with superior corporate governance operating in weaker legal environments. In this instance investors see good firm-level governance as a substitute for deficiencies in investor protection afforded by the legal system. This would indicate a substitutionary relationship.

18 firm’s (ADR) cross-listing premium is inversely related to the quality of investor protection in its home country.

These results all point towards a substitutionary relationship between corporate governance and the country legal environment. It is not immediately apparent however that this is the case. Shareholders may prefer to invest in companies whose country of incorporation guarantees better protection in the eventuality of legal disputes, irrespective of the company corporate governance practices (Bruno & Claessens, 2007). Two papers provide contrary evidence to the substitutionary relationship proposed by the papers mentioned above. Mitton (2004), in examining the relationship between governance, dividend payouts and profitability finds that the positive relation between governance and dividend payout is primarily limited to countries with strong investor protection; this would indicate that country-level and firm-level protection are complementary.

Doidge et al. (2006) come to the conclusion that country characteristics are the most important determinant of a firm’s governance. The adoption of good governance at the firm level has costs and benefits, and these costs and benefits are directly influenced by the level of investor protection granted by the state and the level of financial and economic development of a country. Lower cost of capital is probably the most important benefit derived from good governance. The problem is that this benefit is worth less in countries with low levels of financial and economic development since capital markets are less deep and liquid. The costs associated with improving firm-level governance might be prohibitively high in countries with low development and investor protection since mechanisms required by firms to credibly commit themselves to higher quality governance may be either unavailable or too expensive. This implies that there is a threshold level of development below which the benefits for firms to individually improve upon investor protection granted by the state are too small to justify the costs; for low levels of economic and financial development and investor protection country-level and firm-level governance are complementary. On the other end of the spectrum, once country-level investor protection is very high there would be very little gain for firms to improve upon that level of protection. In this case country-level and firm-level protection and governance are substitutes. This last prediction is less relevant for most countries however since very few have such a high level of country-level investor protection that it would preclude individual firms from improving their corporate governance.

19 contention that strong corporate governance is valued higher in countries with lower country-level investor protection.

Hypothesis 3: The relationship between firm-level governance and country-level legal environment is substitutionary.

2.2.5 STATE PREDATION, REPRESSION AND FIRM VALUATION AND FINANCIAL PERFORMANCE In addition to the establishment and enforcement of laws aimed at protecting investors, the government affects the valuation and financial performance of firms operating within its borders in another crucial way. Managers and majority shareholders often engage in expropriation of other stakeholders when given the opportunity. However, they are not the only parties who exhibit this sort of activity; the state can also engage in predatory behavior.

The principal-principal problem endemic in countries with weak shareholder protection is widely recognized. There are fewer papers that take the existence of additional government expropriation into account. This is known as the ‘twin-agency problem’ (Stulz, 2005). In fact the two are complementary, one reinforces the other. When investors recognize that the state will expropriate profits, they will consent to expropriation by managers, who will have greater incentives to do so. This result emerges because it is more difficult for governments to seize profits that managers have already stolen and hidden from the owners (Durnev & Fauver, 2007). Owners could consent to this because they might prefer managerial diversion to government expropriation if it is easier for the governments to expropriate a fixed fraction of profits from larger or more profitable companies, resulting in even greater seizure of profits (Durnev & Fauver, 2007). The risk that managers will steal hidden profits is preferable to the certainty that the government or the mafia will take even more after honest disclosure (Black & Kraakman, 1996). Higher levels of state predation thus give managers or blockholders greater incentives for expropriation, with the consent of minority owners. This does not necessarily mean that increased managerial expropriation can completely preclude or rule out state predation, it will likely only mitigate it.

Hypothesis 4a: State predation is negatively related to firm valuation and financial performance.

20 transparency or disclosure, has the potential to increase firm value and improve financial performance. However, the same governance mechanisms or procedures can also aid expropriation by the state, which has a negative impact on valuation and financial performance. The moderating influence of state expropriation on the relationship between governance and performance is therefore important to take into account, but few papers have done so.

Hypothesis 4b: The positive relationship between corporate governance and firm valuation and financial performance is weaker in countries with more predatory governments.

In addition to state predation, another factor that is of importance here is the level of autocracy or repression in a country. It is a measure of the effectiveness of political and societal institutions to limit government predation. Durnev & Fauver (2007) argue that firms operating in more democratic countries perform better since there is less expropriation by the government. This is because in pursuit of self-enrichment, autocratic rulers are more likely to set up extortion regimes and are less subject to checks and balances from democratic institutions by, for example, inhibiting independent media (Egorov et al., 2007). When countries are more democratic, manager or majority shareholder expropriation is less necessary as a mechanism to protect from state expropriation. Less manager expropriation and less state predation are expected to have a positive effect on valuation and performance.

Hypothesis 5: State repression is negatively related to firm valuation and financial performance.

2.2.6 FINANCIAL AND ECONOMIC DEVELOPMENT AND FIRM VALUATION AND FINANCIAL PERFORMANCE

The final country characteristics that are important to include in an analysis of the relationship between corporate governance, country legal environment and firm performance are measures for financial (stock market size) and economic (GNI per capita) development. Higher financial and economic development means a deeper and more liquid capital market. This, in turn, means a lower cost of capital for firms, which allows them to invest in more projects and obtain a higher correlation between investment opportunities and actual investments (Shleifer & Wolfenzon, 2002). This should have a positive effect on both valuation and financial performance.

21 variables correlated with shareholder protection, such as corruption, financial market development, and GDP, rather than shareholder protection itself.

Hypothesis 6: Financial and economic development is positively related to firm valuation and financial performance.

2.2.7 FIRM-LEVEL CONTROL VARIABLE HYPOTHESES

This section deals with firm characteristics that need to included in any governance-performance regressions in order to avoid establishing a spurious relationship between governance and valuation/performance. These variables conceivably have an effect on both governance and performance, and therefore are also included in the determinants of governance regression discussed in the next section. Hypotheses are briefly presented.

Firm size. The hypothesis for this variable is ambiguous. Larger firms might benefit from economies of scale, increased market power or international investor attention. This would have a positive impact on valuation and financial performance. On the other hand, small companies may have better growth opportunities, and tend to have higher market-to-book ratios (Bruno & Claessens, 2007).

Sales growth. This variable controls for growth opportunities, and is expected to be positively related to valuation and performance. Firms with better growth opportunities are valued higher by the market, and higher sales growth can also indicate higher profitability.

Tangibility. This variable relates to the composition of a firm’s assets; it measures the relative amount of tangible assets to total sales. The hypothesis for this variable is that it has a negative relation to valuation and financial performance. In general the market values intangibles higher than their book value which may result in a higher Tobin’s Q (Klapper & Love, 2004). Firms with a relatively many intangible assets also tend to be high-growth firms, and a lower ratio (less tangible assets) indicates higher productivity, which means higher efficiency, and more profitability.

23

2.3

The secondary objective of this thesis is to assess the influence of various country-level and firm-level variables on the corporate governance of the firms under scrutiny. These are the determinants of corporate governance quality. Corporate governance is determined endogenously, and assessing the antecedents of corporate governance at the level of the firm will provide some insight into which variables have what influence. This is important because these same variables can also have an influence on the valuation and financial performance of a company. Assessing their effect on corporate governance while at the same time using them as independent variables in governance-performance/valuation regressions will shed light on their role in both issues. First, the expected impact of country-level variables on governance quality will be presented, followed by firm-level variables.

2.3.1COUNTRY-LEVEL VARIABLE HYPOTHESES

Legal system. The legal system, and the level of investor protection that it provides, has been established as having a direct influence on a range of issues. These include such diverse elements of countries' financial systems as the breadth and depth of their capital markets, the pace of new security issues, corporate ownership structures, dividend policies, the efficiency of investment allocation and company valuation (La Porta et al. (2000), Bruno & Claessens (2007)). Differences in legal investor protection across countries also shape the ability of insiders to expropriate outsiders (Djankov et al., 2008), and have been found to positively correlate with a number of corporate governance rankings (e.g. Doidge et al., 2006). One way to think about legal protection of outside investors is that it makes the expropriation technology less efficient (La Porta et al., 2000). As the diversion technology becomes less efficient, the insiders expropriate less, their private benefits of control diminish and the quality of corporate governance increases. The investor protection provided by the country-level legal environment should be positively related to the quality of corporate governance.

24 State repression. Durnev & Fauver (2007) argue that firms operating in more democratic countries practice better corporate governance since there is less expropriation by the government. Political and societal institutions designed to provide a set of checks and balances to deter state expropriation are more developed in less autocratic countries. When countries are more democratic and less predatory manager or majority shareholder expropriation is superfluous as a mechanism to protect from state expropriation, which in turn allows minority shareholders to push for higher quality corporate governance practices. State repression is negatively related to the quality of corporate governance

Financial/economic development. Finally, the economic and financial development of a country is important to consider when looking at corporate governance determinants. This is a relatively recent insight. A 2006 paper by Doidge et al. was the first to take into account the effect economic and financial development has on the costs and benefits firms face when choosing the level of quality of their governance mechanisms. They show that when development is weak, improving firm-level governance has high costs because of a lack of supporting institutional infrastructure, while at the same time the benefit to the firm (lower cost of capital) is low because capital markets lack depth. When development is higher, the costs of voluntarily adopting better governance go down, since supporting institutions make it easier for firms to credibly commit to higher quality governance. At the same time, the benefits of access to capital markets on better terms is more relevant because of better developed financial markets. Both economic and financial development should be positively related to the quality of corporate.

2.3.2FIRM-LEVEL VARIABLE HYPOTHESES

In addition to the country-level variables that have been shown to have an influence on corporate governance, there are a number of firm-level variables that affect the choices firms make with regard to the quality of their governance. In countries with weaker legal environments or less developed capital markets not all firms are badly governed, and vice versa. Country-level variables or characteristics are thus unable to explain all variation in corporate governance scores. A number of firm-level variables that act as additional determinants of corporate governance quality have been identified.

25 external financing might induce firms to improve their governance in order to lower their costs of capital (Klapper & Love, 2004). The relationship between firm size and corporate governance quality is ambiguous.

Sales growth. The second firm-level variable is sales growth, which proxies for growth opportunities. Firms with better growth opportunities might have a greater need for external financing to capitalize on these opportunities and in order to lower their cost of capital might implement better governance mechanisms. Sales growth is positively related to the quality of corporate governance.

Tangibility. The third variable is tangibility, the ratio of tangible assets to total sales. Tangible assets are easier to monitor and harder to steal than intangible assets. Companies operating with higher proportions of tangible assets (and lower proportions of intangible assets) may find it less optimal to adopt stricter governance mechanisms to signal to investors that they intend to prevent the future misuse of intangible assets (Doidge et al., 2006). Tangibility is negatively related to quality of corporate governance.

ADR. The fourth variable is an ADR (American Depositary Receipt) dummy variable. Doidge et al. (2004) show that ADR issuance tends to bond foreign firms to US standards of governance, which are often higher than those prevalent in a firm’s home country. ADR issuance is positively related to the quality of corporate governance.

26

3.

D

Section 3.1 provides an overview of the sample, and defines and operationalizes the various variables involved in the analysis. Section 3.2 presents the regression equations.

3.1 DATA

3.1.1 THE SAMPLE

The basis for the sample was the 2002 Credit Lyonnais Securities Asia (CLSA) CG Watch Report. This report assesses the corporate governance quality in 2001 of 475 firms in twenty countries in Asia, Eastern Europe, the Middle East, Africa and Latin America. The focus here is on ten Asian countries4_{, where a total of 373 firms are assessed. For an overview of the number of firms per} country and per sector see table A1 in the appendix. These corporate governance scores were then matched with firm financial data from the Thomson Financial Datastream/Worldscope database. All financial data is based on values as of 31/12/2001. After matching a sample of 285 firms remained. Firms that could not be positively matched, or for which information was insufficient to calculate Tobin’s Q, are excluded from the sample. Firms with some missing information (e.g. missing sales growth information) are included in the sample. The missing information is coded as N/A in Excel and Eviews. For the ten home countries of these firms data on the country-level legal environment, state predation, political and civil liberties, and financial and economic development was then gathered.

3.1.2DEPENDENT VARIABLES

Tobin’s Q. Tobin’s Q is the dependent variable that is the proxy for firm value, and is chosen because it is used often in the literature. It is defined as the sum of total assets plus the market value of equity less book value of equity, divided by total assets. Total assets is Datastream item DWTA. DWTA is defined as the sum of total current assets, long term receivables, investment in unconsolidated subsidiaries, other investments, net property plant and equipment and other assets. Market value of equity is Datastream item Market capitalization (08001); Market Price-Year End * Common Shares Outstanding. Book value is Datastream item Common equity (03501). It is defined as common shareholders’ investment in a company. It is in local currency.

Return on assets. Return on assets, ROA, is the proxy for firm financial performance. It is defined as earnings before interest, taxes, depreciation and amortization (EBITDA) divided by total assets.

27 Total assets is Datastream item DWTA. EBITDA is Datastream item DWED and represents the earnings of a company before interest expense, income taxes and depreciation. It is calculated by taking the pretax income and adding back interest expense on debt and depreciation, depletion and amortization and subtracting interest capitalized. It is in local currency.

Corporate governance. Corporate governance is a dependent variable in the determinants of governance regressions. The CLSA corporate governance score for each firm is the proxy for quality of firm-level corporate governance. It is taken from the 2002 CLSA CG Watch Report. It is based on a questionnaire consisting of 57 questions covering seven weighted sub indices. These are: management discipline (15%), transparency (15%), independence (15%), accountability (15%), responsibility (15%), fairness (15%) and social awareness (10%). The questions are answered/scored in binary form by CLSA analysts using their best judgment. This results in a score, ranging from 0 to 100, for each weighted sub index, which together form the overall governance score. For a complete overview of the questionnaire the reader is referred to the appendix.

With regard to the choice of weights, the report states: ‘The first six criteria were given an equal weight of 15% and the last, social responsibility, was given a lower weight of 10%, owing to the split response from fund managers as feedback on whether they did or did not see this as part of CG.’ With regard to subjectivity owing to the nature of the questions, the report states the following:‘The balance we have arrived at is that 16 of the 57 questions - just under 30% - is an assessment of the commitment of the company to particular aspects of CG where there is some interpretation required of the analyst. (Nevertheless, the analyst has to provide a definite yes/no answer to reduce the degree of subjectivity for assessing these issues.) The other 70% of the questions are based on hard facts, like whether the chairman is independent, whether there are independent directors heading nomination and remuneration committees, whether the board meets at least four times a year, etc.’ 3.1.3INDEPENDENT COUNTRY-LEVEL VARIABLES

28 higher the hurdles, the higher is the anti-selfdealing index.’ It takes a value between 0 and 1, with 1 representing strong private enforcement mechanisms to contain selfdealing. A full overview of the components of the index can be found in Table A2 in the Appendix.

Predation. This variable is the proxy for predatory government behavior. It is constructed by Durnev & Fauver (2007) and is based on the sum of seven indices: 1) corruption; 2) risk of government expropriation; 3) lack of property rights protection; 4) government stance towards business; 5) freedom to compete; 6) quality of bureaucracy; and 7) impact of crime. Data frequency is annual from 1996 through 2004. The first four indices are from the International Country Risk Guide, the remaining indices are from the Economist Intelligence Unit. This variable ranges from 7 through 36. Larger numbers indicate a greater degree of government predation.

Repression. Repression measures the level of democracy in a country in 2001. It is a cumulative score based on political rights and civil liberties prevalent in a country, provided by Freedom House5_{. The score takes a value from 2 to 14, with 2 representing the highest degree of freedom and} 14 the lowest.

Market capitalization. This is a measure of the level of financial development. Based on the World Development Indicators database compiled by the Worldbank6_{, it measures stock market} capitalization to GDP in the year 2000. Stock market capitalization is the share price times the number of shares outstanding of listed domestic companies. Listed companies are the domestically incorporated companies listed on the country’s stock exchanges at the end of the year.

Per capita income. GNI per capita proxies for economic development. It is also based on World Development Indicators database compiled by the Worldbank for the year 2000. GNI per capita is based on purchasing power parity (PPP). PPP GNI is gross national income (GNI) converted to international dollars using PPP rates. The figures are divided by a hundred to aid coefficient interpretation.

3.1.4INDEPENDENT FIRM-LEVEL VARIABLES

Corporate governance. Corporate governance is an independent variable in the governance-valuation/financial performance regressions. See section 3.1.2 for details on this variable.

5 _{http://www.freedomhouse.org/template.cfm?page=15} 6

29 Size. Net sales or revenues is the proxy for firm size. It is based on the Datastream item net sales or revenues (01001). It is defined by Datastream as gross sales and other operating revenue less discounts, returns and allowances. It is recorded in local currency. It is then converted to US dollars based on the exchange rate in December 2001. As is usual in the literature, the natural log of this figure is taken to arrive at the variable SIZE.

Sales growth. Sales growth proxies for growth opportunities. It is the percentage sales growth or decline of company sales between 31/12/1998 to 31/12/2001. It is also based on Datastream item Net sales or revenues (01001). It is recorded in local currency.

Tangibility. The variable tangibility measures the ratio of tangible assets to net sales. Tangible assets are property, plant and equipment. It is based on Datastream item Property, plant and equipment (02501), defined as gross property, plant and equipment less accumulated reserves for depreciation, depletion and amortization. This item is divided by Datastream item net sales or revenues (01001) to create tangibility. Both are recorded in local currency.

Debt-to-equity. The debt-to-equity ratio is based on Datastream items Total liabilities (03351) and Market capitalization (08001). Total liabilities are defined as all short and long term obligations expected to be satisfied by the company. Market capitalization is defined as market price year end * common shares outstanding. Both are recorded in local currency.

ADR. The variable ADR is a dummy variable that takes the value 1 when a firm has issued ADRs as of 31/12/2001. All types of ADRs are included, from private placement, to over-the-counter to listings on a US stock exchange. Information on whether or not a firm had placed an ADR issue is obtained from the JPMorgan Chase and Thomson Reuters website www.adr.com and from the Bank of NY Mellon Depositary Receipts directory7_.

Industry. The final variable is an industry dummy variable. Industry dummies are included to account for differences in asset structure, accounting practices, government regulation and competitiveness, each of which may affect firm valuation and performance (Durnev & Kim, 2005). The dummy variables cover ten sectors: Energy, Materials, Industrials, Consumer Discretionary, Consumer Staples, Healthcare, Financials, Information Technology, Telecommunications, and Utilities. Each sector is assigned a dummy. Banks and insurance companies are removed from the sample. Real estate companies and non-industrial multi-sector holding companies are classified as Financials. Healthcare is left out and serves as a baseline. Industry dummies are based on the Global

30 Industry Classification Standard developed by Morgan Stanley Capital International and Standard & Poor’s8_.

3.2METHOD

A multiple regression model will be used to analyze cross-sectional data using ordinary least squares (OLS) for both the governance-performance regressions and the determinants of governance regressions. Two-stage least squares will also be used as additional analysis to investigate the relationship between corporate governance and valuation and profitability. A discussion of the assumptions on which OLS relies, and whether these are met, is provided in section 5.1.3 and section 5.2.4, for determinants of governance regressions and governance-performance regressions respectively. Methodology in related literature is reviewed in section 4, and problems associated with studies of this kind are highlighted.

In section five, the first issue for which results are presented is the determinants of governance regressions. They will determine the influence of various country-level and firm-level variables on corporate governance quality. The following multiple regression model will be used to establish relationships between variables.

1. Determinants of governance CGC I = α + β1ASD C _{+ β} 2REPRESSION C _{+ β} 3PREDATION C _{+ β} 4GNIP.C. C _{+ β} 5MARKETCAP C + β₆SIZEC I + β7SALESGR C I + β8TANG C I + β9ADR C I + β10D/E C I + β11INDUSTRY J _{+ ε} where c indexes country, i indexes company, and j indexes industry.

Second, the regressions below assess the influence of firm-level variables and country-level variables on company valuation and financial performance. When investigating the relation between firm-level and country-level variables and company valuation, the dependent variable is Tobin’s Q. When investigating the relation between firm-level and country-level variables and financial performance, the dependent variable is ROA (Return on Assets). Firm value and financial performance are each chosen as dependent variables since corporate governance has a clear theoretical relationship with both. Corporate governance can increase firm value by reducing value destroying private benefits of control associated with the entrenchment effect, such as tunneling, overcompensation and empire building. Corporate governance can also increase financial performance, for example by reducing

31 direct and indirect costs associated with extraction of private benefits, and by making sure the most qualified managers are appointed to the board.

2. Governance-firm valuation QC I = α+ β1CG C I + β2ASD C _{+ β} 3REPRESSION C _{+ β} 4PREDATION C _{+ β} 5(PREDATION C_*CGC I) + β6(ASD C_*CGC I) + β7GNIP.C. C _{+ β} 8MARKETCAP C _{+ β} 9SIZE C I + β10SALESGR C I + β11TANG C I + β12ADR C I + β13D/E C I + β14INDUSTRY J + ε

where c indexes country, i indexes company, and j indexes industry. 3. Governance-financial performance ROACI = α+ β1CG C I + β2ASD C + β3REPRESSION C + β4PREDATION C + β5(PREDATION C_*CGC I) + β6(ASD C_*CGC I) + β7GNIP.C. C _{+ β} 8MARKETCAP C _{+ β} 9SIZE C I + β10SALESGR C I + β11TANG C I + β12ADR C I + β13D/E C I + β14INDUSTRY J_{+ ε} where c indexes country, i indexes company, and j indexes industry.

Dependent variables

CG = firm-level corporate governance score Q = Tobin’s Q

ROA = Return on assets Country-level variables

ASD = legal environment: anti-selfdealing index REPRESSION = level of autocracy

PREDATION = level of state predation GNI P.C.= GNI per capita

MARKETCAP = stock market capitalization/GDP Firm-level variables

CG = firm-level corporate governance score SIZE = ln sales

SALESGR = net sales growth

32 ADR = American Depository Receipt dummy

D/E = debt-to-equity ratio INDUSTRY = industry dummies Interaction variables

The valuation and financial performance regressions includes two interaction variables. The interaction variable ASD*CG in this regression will help answer the question whether corporate governance matters more, or less in countries with a weak legal environment. If corporate governance matters more, the relation is substitutionary, and the interaction term will be negative. If corporate governance matters less, the relation is complementary, and the interaction term will be positive. This can be seen as follows. If the interaction term is negative, and the legal environment is of high quality (ASD = 1), an increase in corporate governance quality (CG ↑) has less effect on the dependent variable than when ASD is smaller. When the interaction term is positive, a high ASD score amplifies the effect a higher CG score has on the dependent variable.

33

4.

In comparable studies different regression methods are used to estimate relationships between variables. Many studies, especially ones that take into account countries with a weaker legal environment, have found positive relationships, or correlations, between corporate governance and accounting- and market-based performance measures. Examples of papers that only use OLS regressions are Claessens et al. (2002), Gompers et al. (2003), Brown & Caylor (2004) and Klapper & Love (2004). However, a well known problem with studies of this kind is endogeneity. The term endogeneity is used broadly to describe any situation when an explanatory variable is correlated with the error term. There are three sources of endogeneity: omitted variables, simultaneity (reverse causality), and measurement error (Wooldridge, 2006). When this occurs, the assumption that the error ε has an expected value of 0, or E (ε|x)=0, is violated. In this case the OLS estimators will be biased and inconsistent. The problems that come up when trying to move from establishing correlation towards establishing causation are summed up in Figure 1 on the next page.

The omitted variable problem can at least partly be alleviated by including relevant regressors in the equation. Even so, it is unlikely that all relevant regressors can be put into an equation because data is either unavailable or it is very hard to find a good proxy variable. For example, managerial behavior, ability, or corporate culture can be very hard to summarize in a variable. This problem is acknowledged by many papers. If governance is affected by the same unobservable features ofmanagerial behavior or ability that are linked to ownership and performance; then it is in this sense that governance and performance are endogenous (Bhagat et al., 2008). It should be noted that failure to mitigate endogeneity problem does not necessarily prevent an article from being widely discussed, see for example Gompers et al. (2003).

34 solution is limited to the market they study; South Korea. They do state that their results provide evidence of a causal relation running from the overall corporate governance index they use to higher share prices in emerging markets. Durnev & Kim (2005) also claim that the causation runs from better corporate governance to higher valuation.

Figure 1.

Source: Chi (2005). NB. Shareholder rights (G) refer to the index developed by Gompers et al. (2003).

The use of panel data is a second method for dealing with endogeneity. Using panel data, fixed effects estimation or first differencing can be used to estimate the effect of time-varying independent variables in the presence of time-constant omitted variables (Wooldridge, 2006). Chi (2005) uses time series data to investigate the link between and corporate governance and Tobin’s Q. using fixed effects models he shows that when a firm puts more restrictions on shareholder rights, i.e. lowers its governance standards, its value decreases. Black et al. (2005) report time series-based evidence that an overall corporate governance index is an important and likely causal factor in explaining firms’ market values. This confirms their earlier OLS and IV evidence. Black et al. (2006b) investigate the relationship between corporate governance and firm market value in Russia using panel data and confirm a causal connection between firm-level governance and firm market value in a firm fixed effects framework.

35 and firm performance. The papers cited above provide evidence of a causal link between corporate governance and performance, but causation can also run the other way. For example, it is possible that poorly performing managers try to avoid being disciplined by restricting shareholder rights (Chi, 2005). There is some evidence that this can indeed be the case. Berglof & Pajuste (2005) find that for firms in Central and Eastern Europe disclosure is viewed as a financial cost that can be cut during bad times; financially constrained firms disclose less. Conversely, well performing firms might improve their governance to try to improve their performance even further. An example of a paper that uses a framework with four simultaneous equations that take into account the relations between governance, performance, capital structure and ownership structure is Bhagat & Bolton (2008). They use IVs for the four dependent variables in order to identify the system through two- and three stage least squares. They find that the indices developed by Gompers et al. (2003) and Bebchuk et al. (2004) are significantly positively correlated with better contemporaneous and subsequent operating performance, but not with future stock market performance.