Disclosure, Corporate Governence and Foreign Share Ownership on the

To view this page ensure that Adobe Flash Player version 9.0.124 or greater is installed.

Please login or register to make a comment!

The working papers are produced by the Bradford University School of Management and are to be circulated for discussion purposes only. Their contents should be considered to be preliminary. The papers are expected to be published in due course, in a revised form and should not be quoted without the author 9s permission.

Working Paper Series Disclosure, Corporate Governence and Foreign Share Ownership on the Zimbabwe Stock Exchange Musa Mangena Venanico Tauringana Working Paper No 06/43 November 2006 DISCLOSURE, CORPORATE GOVERNANCE AND FOREIGN SHARE OWNERSHIP ON THE ZIMBABWE STOCK EXCHANGE Musa Mangena and Venancio Tauringana* ABSTRACT We investigate the association of foreign share ownership with firm-level disclosure and corporate governance structures in Zimbabwe, a developing country in Southern Africa. Our motivation for the study derives from the literature which suggests that foreign investors: (1) generally have a preference for companies that they are well informed as well as for companies in which their investments are more likely to be protected, and (2) avoid companies in developing countries because of weak corporate governance structures and low disclosure. Using data drawn from companies listed on the Zimbabwe Stock Exchange, we examine the effect of disclosure and corporate governance on foreign share ownership.

We find ... more. less.

that disclosure, proportion of non- executive directors, institutional share ownership, and audit committee independence are all positively and significantly associated with foreign share ownership. Our results also demonstrate that market capitalisation, return on equity and liquidity ratios are also significantly associated with foreign share ownership. These results are consistent with the notion that foreign investors have a preference for companies with effective corporate governance structures, companies with less information asymmetry, as well as companies with healthy cash positions on their balance sheet. The results have implications for policy- makers in developing countries in their endeavour to improve liquidity on stock markets through the participation of foreign investors. The results are also useful to managers in developing countries who are keen to increase the market value of their company thereby reducing their cost of capital. Key words: Disclosure; corporate governance; foreign share ownership; information asymmetry; Developing countries; Zimbabwe Stock Exchange. 3 WORKINGPAPERSERIES *The authors are, respectively, Lecturer in Accounting at the School of Management, University of Bradford and Senior Lecturer in Accounting and Finance, Centre for Corporate Governance, Bournemouth University. We would like to express our sincere gratitude to Mr. Richard Mashava of the Zimbabwe Stock Exchange for providing data on foreign share ownership in Zimbabwean listed companies. We are also grateful to Dr. Mohammad Hudaib for his helpful comments and suggestions. Correspondence should be addressed to Dr. Musa Mangena, Lecturer in Accounting, Accounting and Finance Group, Bradford University School of Management, Emm Lane, Bradford, BD9 4JL, UK. Tel: +44 (0) 1274 234340; Fax: +44 (0) 1274 235680; E- mail: m.mangena@bradford.ac.uk. 1. INTRODUCTION The objective of this study is to investigate foreign share ownership in developing countries by examining whether differences in foreign share ownership across listed companies are related to company-specific differences in disclosure and corporate governance mechanisms. Specifically, we examine foreign share ownership in companies listed on the Zimbabwe Stock Exchange (ZSE), the second largest and most active stock exchange (after the JSE Securities Exchange) in Southern Africa. In the aftermath of the Asian financial crisis of 1997-98, which was blamed largely on poor corporate governance (see for example, Johnson et al., 2000), it has been suggested that some of the key reasons why investors avoid investing in emerging markets are poor corporate governance and transparency (McKinsey and Company, 2001; Gibson, 2003). Examples of investors noting corporate governance and transparency as key reasons are (as quoted in Gibson, 2003), California Public Employees 9 Retirement System (CalPERS) (Engardio, 2002) and the Franklin-Templeton mutual fund group (Mobius, 2001). It is in this context that we investigate the association of foreign share ownership with disclosure and corporate governance in Zimbabwe. Consistent with previous studies (e.g., Kang and Stulz, 1997; Dahlquist and Robertsson, 2001; Lin and Shiu, 2003; Jiang and Kim, 2004), we define foreign share ownership as the percentage shareholding owned by foreign investors. Over the past 25 years, an important regulatory change in both the developed and developing countries has been the liberalisation of stock markets, allowing foreign investors to participate in domestic stock markets (Bekaert and Harvey, 2000; Bekaert et al., 2003). As a result of this liberalisation, participation by foreign investors of domestic stock exchanges has increased over the years (Ramaswamy and Li, 2001). Prior literature suggests that foreign investors play a particularly important role in emerging markets. First, foreign investor participation promotes development of emerging markets through its enlargement of the supply of capital (Bekaert and Harvey, 2000; Bekaert et al., 2001; Ramaswamy and Li, 2001) and thus ensuring liquidity and efficiency of these markets (Bekaert and Harvey, 2000). Second, opening the market to foreign investors may increase the value of local companies, thereby reducing their cost of equity capital (Bekaert and Harvey, 2000; Bekaert et al., 2002). Third, foreign investors improve the inflow of foreign exchange, which is often greatly needed in developing countries to finance imports and other foreign payments (Jefferis, 1995; Doidge et al., 2004). Indeed in the case of Zimbabwe, foreign investors are allowed to participate on the stock exchange if they finance the purchase of shares by inward transfer of foreign currency through normal banking channels. However, whilst foreign investors are important for emerging markets, the currency crises in East Asia in the 1990s added a strong doubt that such portfolio investment flows generate benefits for developing countries (Bhagwati, 2001; Gabriele et al., 2001). Despite liberalisation of stock markets around the world, prior literature documents that share ownership in companies is still mostly held by domestic investors (French and Poterba, 1991; Cooper and Kaplanis, 1994; Tesar and Wesner, 1995; Ahearne et al., 2004). This is particularly more prevalent in developing countries than in the developed countries (Salter, 1998; Ball et al., 2000; World Bank, 2004). The literature provides a number of explanations for this phenomenon, better known as the 8home bias 9 (French and Poterba, 1991; Shukla and van Inwegen, 1995; Dahlquist and Robertsson, 2001). Some of the commonly provided explanations for home bias include transaction costs, information asymmetry (Young and Guenther, 2003), differences in corporate governance across countries (Dahlquist et al., 2003; Klapper and Love, 2004), legal and institutional constraints (Klapper and Love, 2004) and foreign exchange risk (Shukla and van Inwegen, 1995). The majority of the home bias literature has tended to explain foreign share ownership at the cross-country level. Thus, whilst foreign share ownership has been widely explored at the cross- country level, relatively little of this research has examined the issue at the firm-level, particularly in developing countries. However, more recently, there have been a growing number of empirical studies examining foreign share ownership at the firm-level. Kang and Stulz (1997) carried out the first study of foreign share ownership at the firm- level in Japan, a developed country. They document that foreign investors tend to participate in companies that are large, low geared and with high export sales ratio. These results are confirmed in a recent study conducted by Jiang and Kim (2004), who in addition, also find that foreign investors tend to avoid companies in which financial institutions have high share ownership. In a Swedish study, Dahlquist and Robertsson (2001) find that foreign investors companies show a preference for 4 WORKINGPAPERSERIES larger firms, firms with low dividend payout, and firms with large cash position on their balance sheet. They also show that foreign investors participate in firms whose shares are liquid and who are also listed on other foreign stock exchanges. In the context of developing countries, Lin and Shiu (2003) provide the first evidence from Taiwan. They reveal that foreign investors hold shares in large companies and in companies with high export sales ratios, and also in companies with a high beta. Consistent with Merton 9s (1987) model, these findings suggest that foreign investors prefer to participate in companies for which they are well informed. We extend this literature on foreign share ownership, as well as literature on the consequences of disclosure (e.g., Leuz and Verrecchia, 2000; Gietzmann and Ireland, 2005) and corporate governance (e.g., Karamanou and Vafeas, 2005; Ajinkya et al., 2005) in three ways. First, whilst most prior research examines developed countries, we undertake our analysis of foreign share ownership using data drawn from Zimbabwe, a developing country in Southern Africa. Southern Africa is of interest to foreign investors because of its vast natural resources and the potential for high returns (Siddiqi, 1998; Bekaert et al., 2003), yet no study has been undertaken in countries of this region. We focus on Zimbabwe because it is one of the most important countries in the region with quite a diversified economy compared to other Southern African countries (excluding South Africa) (OECD, 2003). Additionally, the country is host to a well- developed, mature and active stock exchange, in which participation is open to foreign investors although with some restrictions. 1 Furthermore, firm- level data on foreign share ownership is available in Zimbabwe. The Zimbabwe Stock Exchange (ZSE) maintains a weekly record of foreign share ownership of all listed companies, including the shares traded by foreign investors in each company. 2 Second, we investigate whether differences in foreign share ownership in listed companies are related to firm-specific differences in corporate disclosure. Whilst the prior studies discussed above (e.g., Kang and Stulz, 1997; Lin and Shiu, 2003; Jiang and Kim, 2004) focus on examining firm characteristics that proxy for greater information availability (e.g., size, export sales ratio, multiple listing), we quantify the extent of corporate disclosure using a disclosure index (e.g., Wallace and Naser, 1995; Haniffa and Cooke, 2002; Young and Guenther, 2003; Mangena and Pike, 2005), in order to determine whether disclosure matters for foreign investors 9 participation in companies listed on developing countries 9 stock exchanges. As Bushman and Smith (2001) argue, a firm 9s pre-commitment to high quality disclosure reduces information asymmetry, thereby reducing investors 9 risk of loss from trading with more informed investors. Consistent with this, Young and Guenther (2003) document at the cross country-level that international capital mobility is associated with countries in which there is greater disclosure of value relevant accounting information. Salter (1998) also shows that patchy information about companies listed on stock exchanges of developing countries is the main problem for low foreign investor participation. This problem of transparency in developing countries makes it more difficult for foreign investors to assess the amount, timing and uncertainty of expected future cash flows (Ball et al., 2000; Jiang and Kim, 2004). To the extent that greater disclosure reduces information asymmetry for the foreign investor, we expect to observe a positive relationship between foreign share ownership and firm-level disclosure. Finally, we examine whether firm-level differences in corporate governance mechanisms are associated with foreign share ownership. As Dahlquist et al. (2003) and Klapper and Love (2004) suggest, differences in corporate governance across countries may be used to explain home bias through their impact on share ownership in different countries. Additionally, Shleifer and Vishny (1997) and La Porta et al. (1999) argue that greater investor protection increases investors 9 willingness to provide financing, and indeed to invest in shares of a publicly listed foreign company. Our belief that corporate governance mechanisms can be important to the foreign investors is based on the notion that foreign investors are usually minority shareholders 3 and therefore more susceptible to expropriation by local corporate managers or 5 WORKINGPAPERSERIES 1 The Zimbabwe Stock Exchange (ZSE) opened share trading to foreign investors in 1993 (Bekaert et al., 2003). Although currently the country is undergoing political problems that have had a negative impact on the economy and perception by foreign investors, there is still some trading by foreign investors on the ZSE. 2 We also find in our analysis that most companies provide information on the number of foreign shareholders, the number of shares they hold and also the percentage shareholdings in their annual reports. 3 This is particularly so in some developing countries such as Zimbabwe, where, because of policies on indigenisation of the loca l economy, there is a cap on percentage of foreign shareholdings in any listed company. In Zimbabwe, foreign investors cannot own more than 40% of any company as a group, and not more than 10% of the company as individuals. controlling shareholders (Jensen and Meckling, 1976; Klapper and Love, 2004). Indeed, the World Bank (2004) suggests that low foreign share ownership in developing countries is a result of weak corporate governance structures in these countries. The relationship between firm-level corporate governance and foreign share ownership has not previously been empirically examined in the literature. We expect results at the country-level to be different from those at the firm-level because foreign investors may engage domestic analysts to advise them on which companies to invest (Brennan and Cao, 1997). Domestic analysts may recommend certain companies, not on the basis of corporate governance issues, but on their familiarity with the companies as well as their close relationship with company management. 4 We believe that investigating the determinants of foreign share ownership, particularly the effect of disclosure and corporate governance, is a fundamental research problem, of which the findings may inform public policy decisions, as well as provide guidance for corporate managers on the needs of foreign investors. From the perspective of policy-makers, the importance of investigating foreign share ownership in emerging markets is to suggest areas where efforts to improve the flow of foreign capital into these markets should be concentrated. If disclosure and effective corporate governance are important determinants of foreign share ownership, regulatory changes may be considered to improve disclosure and corporate governance structures. 5 For companies, foreign investor participation should improve the market value of their shares and thus a lower cost of capital. Understanding the impact of disclosure and effective corporate governance (and indeed other factors) on foreign share ownership will enable companies to effectively address these issues. On the academic front, the findings could improve our understanding of foreign share ownership in developing countries as well as the consequences of greater disclosure and effective corporate governance for listed companies in these countries. Our results suggest that effective disclosure policy and corporate governance structures are important for foreign investors 9 participation on stock exchanges in developing countries. We find that foreign share ownership is more likely to be greater in firms that engage in greater disclosure and those with effective corporate governance structures. Foreign share ownership is positively associated with our measure of disclosure. With respect to corporate governance, our results indicate that foreign share ownership is positively associated with the proportion of non-executive directors, institutional share ownership and audit committee independence. Consistent with previous studies, we find that foreign share ownership is positively related to market capitalisation (a measure of company size), return on equity and liquidity. This suggests that foreign investors also prefer companies that are large, profitable companies and those with a good cash position on their balance sheets (Lin and Shiu, 2003). The rest of this paper is organised in the following manner. The next section discusses the institutional environment for foreign share ownership in Zimbabwe, with particular emphasis on disclosure and corporate governance. In Section 3, we review the literature and develop hypotheses on the relationship of foreign share ownership with disclosure and corporate governance structures. Section 4 reports on the methodology we adopt in the study. In section 5, we present the results of the study. Finally, in section 6 we provide summary and concluding remarks. 2. INSTITUTIONAL ENVIRONMENT FOR FOREIGN SHARE OWNERSHIP IN ZIMBABWE Since opening up their markets to foreign investors in the 1990s, developing countries have experienced increased net foreign investment flows from US$42.6 billion in the 1990s to US$192.3 billion in 2003 (World Bank, 2005). Part of this investment, although small in proportion, has gone into the Southern African region, particularly into South Africa and Zimbabwe (Siddiqi, 1998; World Bank, 2005). 6 These countries are considered the most important in the region and also host two of the oldest and most active stock exchanges in the region, and indeed in Africa: the JSE Securities Exchange (JSE) and Zimbabwe Stock Exchange 6 WORKINGPAPERSERIES 4 Coval and Moskowitz (1999) argue that domestic analysts are privy to inside information because they can talk to managers and may also have close personal ties with managers. 5 This is particularly the case given heightened drive since the 1990s, to attract foreign investors to participate in the stock markets, with many developing countries reducing or eliminating foreign share ownership restrictions (Bekaert et al., 2003). 6 However, Zimbabwe has been embroiled in a severe economic and political crisis since 2000, which is mainly attributed to the compulsory acquisition of farms owned by white people by the government. This has severely affected the flow of foreign equity (ZSE), which are open to foreign equity investment (Bekaert et al., 2003; World Bank, 2005). Given that our study is based on Zimbabwe (due to data availability), we discuss the institutional environment of foreign share ownership in Zimbabwe in the following sections. We address in this respect issues relating to ownership structure, corporate disclosure and transparency, as well as the state of corporate governance. 2.1 Zimbabwe Stock Exchange and listed companies ownership structure The Zimbabwe Stock Exchange (ZSE) was established in 1896, initially to provide a forum through which mining companies could raise equity financing to fund operations. Although the ZSE was originally established to cater for the mining industry, today, the majority of listed companies are non-mining. The exchange is regulated by the Zimbabwe Stock Exchange Act 1996 (Chapter 24:18) and is organised as a body corporate under the supervision of a committee of the exchange which falls under the Ministry of Finance. The exchange is small by international standards, but is the second largest and most active in the Southern African region behind the JSE Securities Exchange. As at December 2001, there were 76 listed companies with a market capitalisation of Z$485.6 billion (US$9.1 billion) (World Bank, 2003). According to the World Bank (2003), as of December 2001, local institutional investors (including government institutions) own 70% of listed shares. Foreign investors, who were first allowed to participate on the exchange in June, 1993 own 15%, with the balance being held by local companies and retail shareholders. Although the ZSE is open to foreign investors, share ownership is restricted to 40% of the shares, and no one individual foreign investor is allowed to hold more than 10% of the shares. However, this restriction only applies to companies that were listed on the ZSE after June 1993, thus companies prior to this date allowed to have more than 40% of their shares owned by foreign investors. Dividend income is taxed at 15%, whilst capital gains are taxed at 10% (Zimbabwe Investment Centre (ZIC), 2005]. 2.2.1 Disclosure and transparency Zimbabwe follows a disclosure-based regime, with corporate disclosure and reporting being influenced by the Zimbabwe Companies Act 1956 (as amended in 1996) (Chapter 24:03) and the accounting profession [i.e., Institute of Chartered Accountants of Zimbabwe (ICAZ)] through its adoption of International Financial Reporting Standards (IFRS) (Chamisa, 2000). 7 The Companies Act requires all companies to provide shareholders with audited annual reports prepared in accordance with generally accepted accounting standards and stipulates the minimum basic requirements of financial reports. To supplement the minimum requirements of the Companies Act, the ICAZ recommends that companies should comply with the requirements of the IFRS. In addition to the Companies Act and ICAZ, the ZSE places a continuing periodic reporting obligation on companies listed on the exchange. Listed companies are required to report any relevant and material information necessary to enable investors to evaluate the performance of companies (ZSE Listing Rules, 2002). Additionally, the listing rules require companies to send their audited annual reports to shareholders as well as to the exchange before the annual general meetings. Furthermore, companies are required to publish interim reports, which are usually not audited or reviewed. However, where an audit or review has been undertaken, the report must be published with the interim reports (ZSE Listing Rules, 2002). The monitoring of financial reporting and compliance with standards is the responsibility of the ICAZ, which employs a review method in monitoring and enforcing compliance with statutory and regulatory disclosure requirements (Owusu-Ansah, 1998; Chamisa, 2000). The main purpose of monitoring process is to ascertain and encourage compliance with IFRS. The problem is that the ICAZ does not have the power to enforce compliance with the IFRS that it adopts, and only works through its membership to persuade them to comply (World Bank, 2003). However, for listed companies, the ICAZ works closely with the ZSE which has the power to suspend or de-list non- complying companies from the exchange. In this regard, the ZSE is one of the strictest stock exchange exchanges in Africa (World Bank, 2003). It is perhaps for this reason that compliance with mandatory disclosure and reporting requirements by ZSE listed companies is high (see Owusu-Ansah, 1998 and Chamisa, 2000). 7 WORKINGPAPERSERIES 7 Whilst, the authority to develop or adopt accounting standards in Zimbabwe rests with ICAZ, all standards are first considered by the Zimbabwe Accounting Practices Board (ZAPB) whose membership is drawn from the accounting profession, the Zimbabwe Stock Exchange and industry (Chamisa, 2000). Once adopted, the measurement and disclosure requirements of the standards are incorporated into the Zimbabwe Companies Act and the standards become legally enforceable (Chamisa, 2000). 2.2.2 Corporate governance In the light of the increased emphasis on corporate governance around the world, the Institute of Directors of Zimbabwe (IODZ) has been promoting the principles of good corporate governance by recommending compliance with the recommendations set out in the UK 9s Cadbury Report (1992) or, the South African 9s King Report (1994). 8 The efforts of the IODZ received a major boost with the publication in 2001 of a code entitled 8Principles for Corporate Governance in Zimbabwe: Manual of Best Practice 9 9 and, also with the amendment of the ZSE Listing Rules (2002). Both the code and the ZSE Listing Rules (2002) incorporate the Cadbury Report (1992) and King Report (1994) and companies are encouraged to adopt either of the two. Consistent with the UK, the code is not mandatory. However, like in the UK, the ZSE Listing Rules (2002) (section 7.F.5) compels companies to include a statement in their annual reports indicating the extent to which they comply to enable shareholders and potential investors to evaluate how the principles have been applied. Where the recommended governance structures are not in place, a clear explanation has to be provided in the annual reports to shareholders. The use of corporate governance structures developed in the UK reflects, not only the desire by the ZSE to improve corporate governance, but also the influence of UK business culture on former colonies (Owusu-Ansah, 1998; Chamisa, 2000). 3. LITERATURE REVIEW AND HYPOTHESES In the sections that follow, we develop our hypotheses on the relationship between foreign share ownership and firm-specific disclosure practices as well as corporate governance mechanisms. We first discuss how differences in firm-disclosure of information can lead to differences in foreign share ownership by drawing from the literature on information asymmetries. Second, we draw from the corporate governance literature and discuss how differences in firm- specific corporate governance structures can be linked to differences in foreign share ownership in listed companies. 3.1 Foreign share ownership and disclosure Theoretical research shows that disclosure of information may influence the information costs of investors. For example, Diamond (1985) and Lundholm (1991) conclude that greater disclosure reduces incentives for investors to pay for costly private information. Lev (1988) argue that greater disclosure of accounting information is more likely to favour the interests of the less informed investors than the more informed investors. Foreign investors are more likely to be less informed than domestic investors and the cost of gathering information about companies is also likely to be higher. As Brennan and Cao (1997) argue, whilst information about the domestic economy (and so domestic companies) can be easily acquired, information about foreign economies (and so foreign companies) requires considerably more effort and resources to acquire. Harris and Ravenscraft (1991) provide evidence indicating that because of the additional information costs, foreign investors are at a disadvantage relative to domestic investors. They show that foreign investors purchasing shares in foreign stock exchanges pay a much higher premium than domestic investors. Shukla and van Inwegen (1995) also show that foreign managed funds perform worse than their US counterparts due in part to the information disadvantage of foreign investors. Carlos and Lewis (1995) demonstrate that information considerations were of first order importance in explaining British investment in Canadian railroads in the 19th Century. Furthermore, Chuhan (1992) reports that market participants list limited information on emerging capital markets as one of the major impediments to investing in those markets. Consistent with this, Ahearne et al. (2004) document empirical evidence showing that US investors consider the cost of information gathering as an important determinant of their bias against investing in foreign shares. Finally, other studies document that foreign investment in equities is concentrated in large companies (Kang and Stulz, 1997; Dahlquist and Robertsson, 2001) and in companies with higher export sales ratios (Lin and Shiu, 2003). These results are consistent with foreign investors having relatively less information about smaller and non- exporting companies than domestic investors. If foreigners invest on the basis of the trade-off between expected gains and costs, high 8 WORKINGPAPERSERIES 8 The King Report (1994) of South Africa which was the first code to be issued in Africa, was replaced by the King Report (2002) in an attempt to keep standards of corporate governance in South Africa in step with those in the rest of the world and makes recommendations regarding issues such as non-executive directors, audit committee, and remuneration committee. 9 Although the code is a private sector initiative supported by the International Finance Corporation (IFC), an arm of the World Bank, and other organisations, the code was launched by the Minister of Industry and International Trade indicating government support for the initiative. information costs are more likely to discourage foreign share ownership. The reason for this is that when information costs are high, foreign investors will be less informed and hence face higher risks of misjudgement (Shukla and Inwegen, 1995; Young and Guenther, 2003) or they will pay higher prices to be as informed as domestic investors (Ahearne et al., 2004). In both these cases, foreign investors 9 expected costs are higher. Evidence exists to demonstrate that when foreign investors are less informed than domestic investors, less foreign share ownership occurs in the stock market (see Gordon and Bovenberg, 1996). The above discussion supports the literature on disclosure which suggests that disclosure helps to reduce agency conflicts by bridging the information gap that exists between managers and shareholders (see Healy and Palepu, 2001) and between the informed and uninformed investors (Leuz and Verrecchia, 2000). We therefore argue, drawing from Young and Guenther (2003) that greater firm disclosure should reduce the information costs more for foreign investors than for domestic investors, and thus reduce their information disadvantage. 10 In companies exhibiting greater disclosure levels, the ability for investors to use the information to value the firm is likely to be high. Since, both domestic investors and foreign investors have access to similar information about the company, differences in costs of being informed between domestic and foreign investors will be relatively low (Young and Guenther, 2003). In contrast, for companies with low disclosures, the information asymmetry between corporate managers and shareholders and also between the informed and uniformed investors is high (Coller and Yohn, 1997; Leuz and Verrecchia, 2000). Therefore, given that it is more costly for foreign investors to collect additional information from other sources, the difference in the information costs between domestic and foreign investors will be relatively high (Brennan and Cao, 1997). The following hypothesis summarises our expectations: H1:There is a positive significant association between corporate disclosure in annual reports and foreign share ownership in Zimbabwean listed companies. 3.2 Foreign share ownership and corporate governance Agency theory (e.g., Jensen and Meckling, 1976) suggests that corporate managers or controlling shareholders act in their interests at the expense of outside investors or minority shareholders. Thus when outside investors purchase shares in a company, they face the risk that the returns on their investments may fail to be realised because of expropriation by corporate managers or controlling shareholders (Klapper and Love, 2004). Although the company 9s past performance and other factors may induce outside investors to provide finance to the company, to a large extent, investors would provide finance to companies in which their investments are more likely to be protected against expropriation. Corporate governance, to a large extent, is a set of mechanisms that are designed to protect outside investors against this expropriation by corporate managers (Daily et al., 2003; Ajinkya et al., 2005; Karamanou and Vafeas, 2005). Given this agency problem, and particularly following high profile financial scandals and corporate failures such as Maxwell in the UK, Enron and WorldCom in US, and Parmalat in Italy, the importance of effective corporate governance structures has received increased attention. In the context of developing countries, the Asian financial crisis of 1997-98 highlighted the need for effective corporate governance structures in these countries. In both the developed and developing countries, corporate governance codes (e.g., Cadbury Committee, 1992; Blue Ribbon Committee, 1999; Kings Report, 2002; Higgs Report, 2003) suggest that effective corporate governance structures are important in monitoring the activities of corporate managers, thus protecting shareholders against expropriation. Indeed a large body of prior empirical research documents evidence to suggest that effective corporate governance structures help to align managers 9 interests with those of shareholders. For example, studies document evidence that the presence of non-executive directors reduces the likelihood of fraud (Dechow et al., 1996; Beasley et al., 2000) and earnings management (Klein, 2002; Ajinkya et al., 2005). Prior work also find a negative association between earnings management and institutional share ownership (Ajinkya et al., 2005) and board size and audit committee size (Karamanou and Vafeas, 2005), 9 WORKINGPAPERSERIES 10 The information gap between the domestic and foreign investors cannot be eliminated because domestic investors still have the advantage in collecting additional information given that they can easily talk to employees, managers and suppliers of the firm , and also can easily obtain important information from the local media, and perhaps have personal ties with companies 9 executives (Coval and Moskowitz, 1999). and audit committee independence (Karamanou and Vafeas, 2005). Mangena and Pike (2005) also show a positive relationship of voluntary disclosure with audit committee independence and financial expertise. To the extent that corporate governance will constrain corporate managers and/or controlling shareholders from expropriating other investors by monitoring and fostering an environment of greater transparency (e.g., Cadbury Committee, 1992; Ajinkya et al., 2005), foreign investors are more likely to be more dependent on effective corporate governance structures. The rationale for this argument is that foreign investors are usually minority shareholders (La Porta et al., 1999; Klapper and Love, 2004) and face higher risks of being expropriated by corporate managers and/or controlling shareholders. This problem is also compounded by distance (Brennan and Cao, 1997), so that whilst local minority shareholders are able (to a limited extent) to monitor corporate managers easily, monitoring costs for foreign investors could be considerably high. Effective corporate governance structures should reduce foreign investors 9 risks and increase their confidence and willingness to invest in particular listed companies. In addition, they should help to ensure that minority shareholders receive reliable information about the performance of the company and that the value of their investment is not expropriated by managers and controlling shareholders (Bushman and Smith, 2003). Therefore, on balance, there is reason to expect a positive relationship between foreign investor share ownership and corporate governance structures. Hence: H2:There is a positive significant association between effective corporate governance mechanisms and foreign share ownership in Zimbabwean listed companies. 4. DATA AND METHODOLOGY 4.1 Sample selection The analysis we perform in this study is based on data for two years drawn from non-financial companies listed on the Zimbabwe Stock Exchange (ZSE). We select non-financial companies only for three reasons. First financial companies operate under the Banking Act of 1999 (Chapter 24:20), which compels them to have certain corporate governance structures which non-financial companies are not obliged (although recommended) to have. Second, financial companies are also required to disclose certain additional information by regulators (e.g., the Reserve Bank of Zimbabwe) which non- financial companies are not required to disclose. Finally, previous studies (e.g., Dahlquist and Robertsson, 2001; Jiang and Kim, 2004) have also employed non-financial companies only and thus to enable comparison with prior studies, we also exclude non-financial companies. As of 31 December, 2004 the total number of companies listed on the Zimbabwe Stock Exchange, including both financial and non- financial companies was 87. Given this small population of listed companies, we decided to employ all the non-financial companies in this study. We requested for the annual reports and received 51 reports for 2002 and 67 for 2003 (Table 1). 4.2 Measure of foreign share ownership Data for foreign share ownership (FORESHARE) was obtained from the Zimbabwe Stock Exchange. The ZSE maintains a record of foreign share ownership for all listed companies, and these were provided to the researcher for 2001, 2002, 2003 and 2004. However, because we only have annual reports for 2002 and 2003, we run our regression analysis using 2003 and 2004 foreign share ownership data. The reason for this is that if 10 WORKINGPAPERSERIES TABLE 1 3 SAMPLE OF COMPANIES foreign investors are influenced by disclosure and corporate governance structures, it is highly likely that they use the previous year 9s data to assess the firm. We are unable to run the analysis for 2001 and 2002 because we failed to get the applicable annual reports for 2000 and 2001, respectively. The foreign share ownership data for 2002 and 2003 as provided by the ZSE was compared with data in annual reports of the companies, where reported. Under section 8.52 (e) of the ZSE Listing Rules (2002), all listed companies are required to provide an analysis of their shareholding by category. Whilst, the ZSE Listing Rules (2002) do not specifically require companies to explicitly provide shareholding by foreign investors, some companies do provide this information in their annual reports. In our sample of companies, we find that the ZSE data is the same with that reported in the annual reports, suggesting that our measure of foreign share ownership is reliable. 11 4.3 Measure of disclosure We measure disclosure using a disclosure checklist developed on the basis of the one compiled by Gray et al. (1995). Gray et al. (1995) developed the most comprehensive disclosure checklist, taking into consideration international trends, observation of standards of reporting practice and relevant research studies. We adjusted the checklist to take account of the reporting environment in Zimbabwe, and also taking account of the information items identified in prior studies as important for foreign investors (see Choi and Levich, 1990). From the resultant list we eliminated all mandatory items given that mandatory disclosure by Zimbabwean listed companies is high (Owusu-Ansah, 1998; Chamisa, 2000). Our final checklist comprised 86 voluntary disclosure items. Consistent with previous studies (e.g., Wallace and Naser, 1995; Haniffa and Cooke, 2002), disclosure for each company was measured using a dichotomous procedure, where an item is scored 1 if it is disclosed or 0 if it is not disclosed. To minimise the possibility of penalising companies for disclosures that were not applicable or relevant, a review of the entire annual report was undertaken (e.g., Wallace and Naser, 1995; Haniffa and Cooke, 2002). After scoring the entire annual report, the total disclosure score was divided by the total possible score for each company to produce a disclosure index (DSCORE). 4.4 Measures of corporate governance mechanisms For corporate governance mechanisms, we employ the widely used measures of effective corporate governance mechanisms of board size, proportion of non-executive directors, institutional share ownership, audit committee, and top-ten shareholders (a measure of ownership concentration). These measures have been used in other previous studies (e.g., Haniffa and Cooke, 2002; Karamanou and Vafeas, 2005; Ajinkya e al., 2005; Mangena and Pike, 2005) and indeed are recommended by corporate governance codes in different countries (e.g., Cadbury Committee, 1992; Blue Ribbon Committee, 1999; King Report, 2002; Higgs Report, 2003). We obtain these measures from the annual reports of the companies in our sample. In Zimbabwe, all companies are required by the Companies Act 1996 and the ZSE Listing Rules (2002) to disclose this information in the annual reports to shareholders. 4.5 Control Variables According to Bartov et al. (2000) failure to control for confounding variables could lead to falsely rejecting the hypothesis when in fact it should be accepted. We draw from prior research on foreign share ownership and control for a number of other variables that are likely to be determinants of foreign share ownership. Dahlquist and Robertsson (2001) and Jiang and Kim (2004) provide evidence supporting a positive relationship between foreign share ownership and company size (measured in terms of market capitalisation), profitability and gearing ratio. Lin and Shiu (2003) also show that liquidity ratio is positively related to foreign share ownership. We therefore control for company size (market capitalisation), return on equity, liquidity ratio and gearing ratio in our study. 4.6 Model specification To examine whether differences in share ownership are influenced by differences in disclosure and corporate governance, we use the following ordinary least squares (OLS) regression model. FORESHARE = ² 0 + ² 1 DSCORE + ² 2 BSIZE + ² 3 PROPNED+ ² 4 INSHARE+ ² 6 TOP10SHARE + ² 6 ACSIZE + ² 7 ACIND + ² 8 MCAP + ² 9 ROE + ² 10 LIQUID + ² 11 GEAR + µ j (1) 11 WORKINGPAPERSERIES 11 We find that in our sample of companies, at least 92% provided this information in their annual reports. Since we do not have the annual reports for 2001 and 2004, we are unable to do the comparison for these years. All the variables used in the regression model are defined in table 2. 5. EMPIRICAL FINDINGS OF THE STUDY 5.1 Descriptive Statistics Summary descriptive statistics of foreign share ownership on the Zimbabwe Stock Exchange are presented in table 3. The mean for foreign share ownership ranges from 15.1% in 2001 to 7.9% in 2004. These results indicate that there has been a significant decrease in foreign share ownership on the ZSE since 2001. This decline is expected given the political and economic problems prevailing in the country since 2000, which have made Zimbabwe a risky country to invest in. Table 4 presents summary statistics for the independent variables. With regard to the main independent variables of interest, the summary statistics show that voluntary disclosure is very low, although it slightly increased in 2003. For the corporate governance variables, we find that on average the size of boards in ZSE listed companies is seven, whilst non-executive directors constitute about 49%. The non-executive director percentage is consistent with those reported in the UK (see for example, Young, 2000; Mangena and Pike, 2005). This is not surprising particularly given that the ZSE recommends that listed companies comply with the UK Cadbury Report 9s (1992) recommendations. We observe that all companies in our sample had an audit committee 12 WORKINGPAPERSERIES TABLE 2 3 DEFINITION OF VARIABLES INCLUDED IN THE MULTIPLE REGRESSION MODEL 1 in place, and the audit committee size is about three directors, of which 49.5% are non-executive directors. Another noticeable finding of this study is that the top-ten shareholders (our proxy for ownership concentration) in each company dominate share ownership in ZSE listed companies, with a mean share ownership of about 84%. These shareholders, who include companies and in certain cases government bodies, may own the shares for long-term benefit (Jiang and Kim, 2004) and perhaps strategic reasons and are more unlikely to trade actively on the exchange. This may provide an explanation for the liquidity problems encountered by the ZSE and indeed other stock exchanges in developing countries (see World Bank, 2003). The summary statistics in tables 3 and 4 suggest that the variables in our model are not normally distributed. In addition, an analysis of the statistics on skewness and kurtosis, as well the Kolmogorov-Smirnov (K-S Lilliefors) normality test statistics (not reported) also suggests that the assumptions of normality were not met. 12 Furthermore, an analysis of residuals, plots of the studentised residuals against predicted values, as well as the Q-Q plots indicated the assumption of homoscedasticity, linearity and normality were violated (see Field, 2000). We, therefore, transformed both the dependent and independent variables using normal scores (Cooke, 1998; Haniffa and Cooke, 2002) 13 and performed our regression analysis with the transformed variables. 5.2 MULTIPLE REGRESSION RESULTS 5.2.1 Correlation matrix and multicollinearity In Table 5, we provide the Pearson correlations among the variables. The results reveal several significant relationships between the dependent variable and independent variables, as well as among independent variables. With regard to the independent variables, although the correlation coefficients are significant, they are all below 0.50 for both the two years, suggesting that data may have minimal unusual influences on the estimated regressions. 14 6.2.2 Results and Discussion Table 6 presents the multiple regression analysis results of the hypotheses of the relationship of foreign share ownership with voluntary disclosure and corporate governance mechanisms. The estimations are carried out on each of the two years 2003 and 2004, as well as in a pooled regression. All the multiple regression models explain a significant explanatory amount of the variations in foreign share ownership. The Adjusted R 2 of the models ranges from 32.8% to 36.7%. These explanatory powers are better than those reported in Dahlquist and Robertsson (2001) on Swedish data and Lin and Shiu (2003) on Taiwanese data, but slightly lower than Jiang and Kim (2004) who report an Adjusted R 2 ranging from 39.2% to 45.7%. The results suggest that foreign share ownership is more likely to be greater in companies with greater voluntary disclosure and effective corporate governance structures, thus supporting the findings of previous studies at the country level (e.g., Young and Guenther, 2003; Klapper and Love, 2004) indicating that foreign investors invest in countries with greater disclosure as well as good corporate governance. The results are similar in the regressions for the two years 2003 and 2004, as well as in the pooled regression. We find that foreign ownership is positively and significantly associated with disclosure, and thus supporting Hypothesis 1. This finding is consistent with the argument that foreign investors 13 WORKINGPAPERSERIES TABLE 3 3 SUMMARY STATISTICS FOR FOREIGN SHARE OWNERSHIP ON THEZIMBABWE STOCK EXCHANGE 12 K-S Lilliefors with significance of >0.05 indicates that the distribution is approximately normal (Field, 2000). 13 Cooke (1998) proposes the normal scores method as the most appropriate in transforming datasets that reveal non-linear monotonic relationships between the independent and dependent variables. 14 As a rule-of-thumb, multicollinearity in regression analysis is considered harmful only when they exceed 0.8 (Field, 2000). 14 WORKINGPAPERSERIES TABLE 4 3 DESCRIPTIVE STATISTICS FOR THE INDEPENDENT VARIABLES TABLE 5: CORRELATION MATRIX participate in companies for which more information is available (Dahlquist and Robertsson, 2001; Young and Guenther, 2003). As for corporate governance variables, we find that foreign share ownership is positively associated with the proportion of non-executive directors, institutional share ownership and audit committee independence. We interpret these results as an indication that foreign investors invest in companies in which their interests are more likely to be protected. The positive relationship of foreign share ownership with the proportion of non-executive directors suggests that foreign investors perceive non-executive directors as effective in monitoring and controlling the opportunistic behaviour of managers, and thus protecting the interests of shareholders. In the case of institutional share ownership, we interpret the results in two ways. First, the results suggest that foreign investors perceive institutional investors in Zimbabwe as well suited to monitor managers, perhaps due to the usually information advantage they have (Karamanou and Vafeas, 2005) and also because they usually have higher investment which lead to their incentives to monitor managers. Second, foreign investors may have an interest in companies whose shares are tradable or liquid. In the context of Zimbabwe, institutional investors are the most active market makers, and the shares that are more likely to be actively traded on the ZSE are those held by institutional investors (World Bank, 2003). The positive association of foreign share ownership with audit committee independence suggests that foreign investors perceive independent audit committees as free from the influences of management and therefore more likely to ensure that shareholders are provided with quality and credible financial 15 WORKINGPAPERSERIES TABLE 6 3 REGRESSION RESULTS OF THE RELATIONSHIP OF FOREIGN SHARE OWNERSHIP WITH DISCLOSURE AND CORPORATE GOVERNANCE information. This is consistent with the literature (e.g., Karamanou and Vafeas, 2005; Mangena and Pike, 2005) and corporate governance codes (e.g., Cadbury Committee, 1992; King Report, 2002) suggesting that independent audit committees are more effective monitors of the financial reporting process. As for board size and audit committee size, we find no significant relationship with foreign share ownership, although the coefficients are positive as predicted. These results suggest that the size of the board or the audit committee is not an important consideration in the decision process of the foreign investor. This is inconsistent with the literature which suggests that the size of the board or audit committee results in better monitoring of managers (e.g., Karamanou and Vafeas, 2005; Mangena and Pike, 2005), and thus reducing expropriation of shareholders. Our results also indicate that foreign share ownership is not significantly related to share ownership by the top-ten shareholders, thus contradicting Dahlquist and Robertsson (2001) who find a negative significant relationship. However, the coefficient of top-ten shareholders is negative as predicted, providing modest support indicating that foreign investors under-invest in companies with concentrated shareholding. With respect to the control variables we observe that all, except gearing ratio, are significant and of the expected direction. Consistent with previous studies (e.g., Kang and Stulz, 1997; Dahlquist and Robertsson, 2001; Lin and Chiu, 2003; Jiang and Kim, 2004), we find that company size, measured in terms of market capitalisation, has a positive significant effect on foreign share ownership. The coefficients of return on equity and liquidity are all positive and significant, which is also consistent with previous studies (e.g., Kang and Stulz, 1997; Dahlquist and Robertsson, 2001; Lin and Chiu, 2003) that show that greater foreign share ownership in companies with higher returns and also in companies with large cash balances on their balance sheet. Although gearing ratio is not significantly associated with foreign share ownership, and thus inconsistent with Dahlquist and Robertsson (2001) and Jiang and Kim (2004), the direction is negative suggesting that foreign investors may avoid companies with high debt. 6.2.3 Robustness of Results We perform some additional analyses to determine if our results are robust. First, we re- estimate our regressions with industry dummy variables since no control variable has been included in the regression models for industry. We classify our sample companies into four industries based on the ZSE classification and used 0; 1 dummies to indicate each company 9s industry group: (1) mining and construction, (2) manufacturing, (3) consumer goods and services, and (4) agriculture. The results of these re- estimations (not reported) support our main findings that foreign share ownership is positively and significantly associated with disclosure and corporate governance variables, suggesting that our results are not affected by industry effects. We also re-run our regression analyses with companies in which foreign share ownership is more than the 40% threshold excluded from our sample. The reason for this is that companies that were listed prior to June 1993 are allowed to exceed the 40% threshold (ZIC, 2005). Therefore, it is possible that our results are influenced by the data drawn from these companies. In this analysis, we find that our results are largely unaffected as indicated in table 7. Except for gearing ratio, which becomes significant at the 10% level, our foreign share ownership measure continues to be significantly associated with voluntary disclosure, proportion of non-executive directors, institutional share ownership and audit committee independence. This suggests that our results are robust in supporting our hypotheses. Finally, Myers (1990) and Field (2000) suggest that even when the correlations between the independent variables are not very high, some degree of multicollinearity can still exist. We therefore further examine the extent to which multicollinearity presents a problem in the estimation of the relationship between our dependent and independent variables by computing the variance inflation factors (VIFs) for each independent variable (Myers, 1990; Field, 2000). The VIFs should not exceed the critical value of 10 (Myers, 1990; Field, 2000). Since all the VIFs in our model do not exceed 2.5 (not reported here), we conclude that multicollinearity is not a major problem in our analysis. 7. SUMMARY AND CONCLUDING REMARKS Our focus in this study is to examine the relationship of foreign share ownership and disclosure and corporate governance mechanisms in Zimbabwe, a developing country in Southern Africa. We are motivated by the literature which suggests that foreign investors have a preference for companies that they are well informed as well as for companies in which their investments are more likely to be protected. Given this, it is 16 WORKINGPAPERSERIES suggested in the literature that foreign investors avoid investing in companies in developing countries due to weak corporate governance structures as well as low disclosure. Using a dataset relating to listed Zimbabwean companies for the period 2001 to 2004, we find that foreign share ownership is positively related with voluntary disclosure, proportion of non-executive directors, institutional share ownership and audit committee independence. Thus our results are consistent with the notion that foreign investors have a preference for companies with greater disclosure and effective corporate governance structures. We also find, consistent with previous studies that foreign share ownership is associated size, profitability and liquidity. The findings of this study must be interpreted in the light of the following potential limitations. First, our measure of foreign share ownership could be problematic. We do not distinguish between portfolio equity investment and foreign direct investment (FDI) because of data limitations. In contrast to FDI, portfolio equity investment is not motivated by a long-term interest in controlling the destination company, but by financial returns (World Bank, 2004). It is possible that there may well be differences in the determinants of these two components of foreign share ownership. A second limitation regards the possibility of omitted variables in the regression models. Variables such as export sales ratios, beta, dividend yield, volatility and volume of shares traded could also influence foreign share ownership. These have not been examined owing to data limitations. Third, we have not investigated the possible endogenous relationship of foreign share ownership and firm-level 17 WORKINGPAPERSERIES TABLE 7: REGRESSION RESULTS OF THE RELATIONSHIP OF FOREIGN SHARE OWNERSHIP WITH DISCLOSURE AND CORPORATE GOVERNANCE (EXCLUDING COMPANIES IN WHICH FOREIGN SHARE OWNERSHIP EXCEEDS 40%) disclosure and firm-level corporate governance structures. Whilst foreign investors may have a preference to invest in companies with greater disclosure and effective corporate governance structures, it is possible that they may also influence disclosure policy (Haniffa and Cooke, 2002) as well as effective corporate governance structures (Ramaswamy and Li, 2001). This endogenous effect could influence the interpretation of our results. Finally, our measure of voluntary disclosure assumes that foreign investors attach equal weights to all items, yet in real life some items may be considered more important than others. Furthermore, our measure of disclosure is based on information in published annual reports only. Companies disclose through a variety of ways (e.g., internet, the press). However, in spite of these limitations, this study makes some important contributions to the literature examining foreign share ownership, consequences of voluntary disclosure and effective corporate governance structures. First, we show that greater voluntary disclosure and effective corporate governance are related to foreign share ownership. Whilst evidence exists at the country-level that international capital mobility is associated with countries in which disclosure is greater (Young and Guenther, 2003) and in which corporate governance is stronger (Klapper and Love, 2004), there is no evidence to suggest the same at the firm-level. Thus, our study provides the first evidence that foreign share ownership is related to firm-level disclosure and corporate governance. Second, we conduct our analysis on a developing country, whereas most studies undertake their analyses in developed countries. Our study is the first such study in Southern Africa, and indeed in Africa, a continent generally considered as risky (World Bank, 2005). We conclude, in the light of our findings, that our results are consistent with the notion that disclosure and corporate governance matter for foreign investors. In view of the efforts to improve corporate governance in developing countries, our results are likely to have policy implications. For developing countries such as Zimbabwe, fostering an environment of greater disclosure and effective corporate governance could have positive effects on foreign investor participation in listed companies. However, given that this is the first research to investigate the relationship of foreign share ownership with voluntary disclosure and corporate governance at the firm-level, more research need to be carried out in the context of other developing countries. Future research could extend this study to other countries in the Southern Africa region (e.g., South Africa, Botswana) and indeed to other countries where data on foreign share ownership is available. 18 WORKINGPAPERSERIES REFERENCES Ahearne, A.G., W.L. Griever and F.E. Warnock, cInformation costs and home bias: an analysis of US holdings of foreign equities, d Journal of International Economics 62 (2004), pp. 313-336. Ajinkya, B., S. Bhojraj and P. Sengupta, cThe association between outside directors, institutional investors and the properties of management forecasts, d Journal of Accounting Research 43 (2005), pp. 343-376. Ball, R., S.P. Kothari and A. Robin, cThe effect of international factors on properties of accounting earnings, d Journal of Accounting and Economics 29 (2000), pp. 1-51. Bartov, E., F.A. Gul and J.S.L. Tsui, cDiscretionary- accruals models and audit qualifications, d Journal of Accounting and Economics 30 (2000), pp. 421-452. Beasley, M.S., J.V. Carcello, D.R. Hermanson and P.D. Lapides, cFraudulent financial reporting: Consideration of industry traits and corporate governance mechanisms, d Accounting Horizons 14 (2000), pp. 441-454. Beakaert, G. and C.R. Harvey, cForeign speculators and emerging equity markets, d Journal of Finance 55 (2000), pp. 565-613. Bekaert, G., C.R. Harvey and C.T. Lundblad, cEmerging equity markets and economic development, d Journal of Development Economics , 66 (2001), pp 465-504. Bekaert, G., C.R. Harvey and R.L. Lumsdaine, cDating the integration of world equity markets, d Journal of Financial Economics 65 (2002), pp. 203-247. Bekaert, G., C.R. Harvey and C.T. Lundblad, cEquity market liberalization in emerging markets, d Review-Federal Reserve Bank of St. Louis 85 (2003), pp. 53-80. Bhagwati, J., The Winds of the Hundred Days: How Washington Mismanaged Globalization , (Cambridge: MIT Press, 2001) Blue Ribbon Committee, Report and Recommendations of the Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees, (New York: New York Stock Exchange, 1999). Brennan, M.J. and H.H. Cao, cInternational portfolio Investment flows, d Journal of Finance 52 (1997), pp. 1851-1880. Bushman, R. and A. Smith, cFinancial accounting information and corporate governance, d Journal of Accounting and Economics 32 (2001), pp. 237- 333. Bushman, R. and A. Smith, cTransparency, financial accounting information, and corporate governance, d FRBNY Economic Policy Review (2003), pp. 65-87. Cadbury Committee, Report of the Committee on the Financial Aspects of Corporate Governance, (London: Gee and Company, 1992). Carlos, A.M. and F.D. Lewis, cForeign financing of Canadian railroads, d in Michael D. Bordo and Richard Sylla, Eds. Anglo-American Financial Systems , (New York: Irwin, 1995) Chamisa, E., cThe relevance and observance of the IASC standards in developing countries and the particular case of Zimbabwe, d International Journal of Accounting 35 (2000), pp. 267-286. Choi, F. and R.M. Levich, The capital market effects of international accounting diversity, (Homewood, IL: Dow Jones-Irwin, 1990). Chuhan, P., cSources of portfolio investment in emerging markets, d (Working Paper, World Bank, Washington DC, 1992) Coller, M. and T.L. Yohn, cManagement forecasts and information asymmetry: an examination of bid-ask spreads, d Journal of Accounting Research 35 (1997), pp. 181-191. Cooper, I. and E. Kaplanis, cHome bias in equity portfolios, inflation hedging, and international capital market equilibrium, d Review of Financial Studies 7 (1994), pp. 45-60. Coval, J.D. and T.J. Moskowitz, cHome bias at home: Local equity preference in domestic portfolios, d Journal of Finance 54 (1999), pp. 2045-2073. Dahlquist, M. and G. Robertsson, cDirect foreign ownership, institutional investors, and firm characteristics, d Journal of Financial Economics 59 (2001), pp. 413-440. 19 WORKINGPAPERSERIES Dahlquist, M., L. Pinkowitz, R.M. Stulz and R. Williamson, cCorporate governance and the home bias, d Journal of Financial and Quantitative Analysis 38 (2003), pp. 87-110. Daily, C.M., D.R. Dalton, and A.A. Cannella,Jr, cCorporate governance: decades of dialogue and data, d Academy Management Review 28 (2003), pp. 371-382. Dechow, P.M., R.G. Sloan and A.P. Sweeney, cCauses and consequences of earnings manipulation: An analysis of firms subject to enforcement actions by the SEC, d Contemporary Accounting Research 13 (1996), pp. 1-36. Diamond, D, cOptimal release of information by firms, d Journal of Finance, 40 (1985), pp. 1071- 1094. Doidge, C., G. Karolyi and R.M. Stulz, cWhy are foreign firms listed in the U.S. worth more? d Journal of Financial Economics 71 (2004), pp. 205-238. Field, A., Discovering Statistic Using SPSS for Windows. (London: Sage, 2000). French, K.R. and J.M. Poterba, cInvestor diversification and international equity markets, d The American Economic Review 81 (1991), pp. 222-226. Gabriele, A., K. Boratav and A. Parikh, cInstability and volatility of capital flows to developing countries, d World Economy 23 (2000), pp. 1031- 1056. Gibson, M.S., cIs corporate governance ineffective in emerging markets? d Journal of Financial and Quantitative Analysis 38 (2003), pp. 231-250. Gietzmann, M. and J. Ireland, cCost of capital, strategic disclosures and accounting choice, d Journal of Business Finance and Accounting 32 (2005), pp. 599-634. Gordon, R.H. and L. Bovenberg, cWhy is capital so im

less