Managerial Ownership and Informativeness of Earnings: Evidence from Thailand

This paper examines the relation between managerial ownership and the quality of accounting information in an emerging market. It is hypothesized that the relation between managerial ownership and earnings’ explanatory power of returns is negative and different from that documented in developed markets due to a unique institutional setting, leading to relatively high agency conflicts and information asymmetry. However, the relation between managerial ownership and earning informativeness becomes positive under financial distress because of the hypothesized effects of creditor monitoring, minority interest, and management’s desire to reduce further intervention. In addition, managerial ownership is positively related to the use of discretionary accruals. The finding is consistent with the notion that ownership structure affects the quality of accounting information. Corresponding Author: Piman Limpaphayom, Sasin Graduate Institute of Business Administration, Chulalongkorn University, Chula 12, Phyathai Road, Bangkok 10330, Thailand. Telephone: 66(2) 2184028. E-mail: [email protected]. ____________________________________________________________________________ The authors thank Anant Chiarawongse, J. Thomas Connelly, Kenneth A. Kim, Arnat Leemakdej, Nandu Nagarajan, and Patarapong Vongsvivut for valuable comments on earlier drafts of this paper. Financial support from Sasin Graduate Institute of Business Administration of Chulalongkorn University and the Master in Finance Program at Thammasat University is gratefully acknowledged. Managerial Ownership and Informativeness of Earnings: Evidence from Thailand Abstract This paper examines the relation between managerial ownership and the quality of accounting information in an emerging market. It is hypothesized that the relation between managerial ownership and earnings’ explanatory power of returns is negative and different from that documented in developed markets due to a unique institutional setting, leading to relatively high agency conflicts and information asymmetry. However, the relation between managerial ownership and earning informativeness becomes positive under financial distress because of the hypothesized effects of creditor monitoring, minority interest, and management’s desire to reduce further intervention. In addition, managerial ownership is positively related to the use of discretionary accruals. The finding is consistent with the notion that ownership structure affects the qua lity of accounting information.This paper examines the relation between managerial ownership and the quality of accounting information in an emerging market. It is hypothesized that the relation between managerial ownership and earnings’ explanatory power of returns is negative and different from that documented in developed markets due to a unique institutional setting, leading to relatively high agency conflicts and information asymmetry. However, the relation between managerial ownership and earning informativeness becomes positive under financial distress because of the hypothesized effects of creditor monitoring, minority interest, and management’s desire to reduce further intervention. In addition, managerial ownership is positively related to the use of discretionary accruals. The finding is consistent with the notion that ownership structure affects the qua lity of accounting information. Managerial Ownership and Informativeness of Earnings: Evidence from Thailand The Bank of Thailand’s release of the Baht peg and the subsequent devaluation of the Thai Baht on July 2, 1997, was viewed as the onset of the Asian financial crisis, which spreaded through the region. In Thailand, the crisis heralded the closing of financial institutions and resulted in an increase in the number of high non-performing loans and debt restructurings. Under this financial crisis, creditors and outside investors became increasingly concern about default risk and the credibility of information reported in a firm’s valuation. This resulted in several questions concerning the quality of information that a firm provides to outsiders, as updates to the projection of a firm’s performance use a wide variety of information sources. La Porta, Lopez-de-Silanes and Shileifer (1999) posit that, in most East Asian countries, corporate control is enhanced through pyramidal structures and cross-holdings among mainly family-controlled firms. Many researchers have attempted to explain the relationship between corporate ownership and earnings report quality, as corporate ownership structure in East Asian countries is unique in comparison to that in the United States. Fan and Wong (2002) examines the relations between earnings informativeness, measured by the earnings-return relation, and the ownership structure of 977 companies in seven East Asian economies. The results are consistent with two complementary explanations in which concentrated ownership and the associated pyramidal and cross-holding structures create agency conflicts between controlling owners and outside investors. Consequently, it is perceived that controlling owners report accounting information out of selfinterest, resulting in reported earnings with little credibility with outside investors. Furthermore, concentrated 1 “Quality” is defined as the usefulness of the financial statements in contracting, monitoring, and valuing and other decision-making activities by investors, creditors, managers and other parties contracting with the firm. (Ball & Shivakumar 2001) ownership structure is associated with low earnings report quality as ownership concentration prevents leakage of proprietary information about the firms’ rent-seeking activities. Warfield, Wild and Wild (1995) finds managerial ownership to be directly related with earnings report informativeness for returns and inversely related with the magnitude of accounting accrual adjustment. A study by Yeo, Tan, Ho and Chen (2002) on companies in Singapore consisting of 490 firm-year observations shows that earnings’ quality does not always increase with managerial ownership. At low levels of management ownership, earnings informativeness (the level of discretionary accruals) has positive (negative) relationship with management ownership. However, at higher levels of managerial ownership, this relationship reverses. This suggests the entrenchment effect to have possibly set in. Thailand provides a unique setting to test the relation between managerial ownership and earning informativeness. First of all, the corporate governance structure in Thailand is relatively weak compared with other developed economies. Low transparency and the lack of disclosure are two major problems in the Thai market (Zhuang, Edwards, Webb, and Capulong, 2000). Previous studies in East Asia have found that corporate governance affects firm valuation (Mitton, 2002; Lins, 2003). Consequent ly, the relation between managerial ownership and earnings informativeness in Thailand could be different from that observed in other developed markets. With the economic crisis in Thailand, further, many listed companies in the Stock Exchange of Thailand (SET) were either in the debt restructuring process or bankruptcy court due to an inability to pay back the principle and interest on loans outstanding. Thus creditors monitored closely the companies that were candidates for debt restructuring, leaving open the possibility for further intervention into the affairs of a financially distressed company that had reached agreement upon its debt restructuring process. Financial distress and the intervention of creditors raise questions concerning the relations among the levels of managerial ownership, the use of earnings management, and the informativeness of financial information after the crisis period. Lemmons and Lins (2003) find that the Asian crisis increases the incentives of controlling shareholders to expropriate wealth from minority shareholders. Consequently, the results from Thailand should provide insights on the relation among variables. This paper investigates how the separation of ownership and control affects the informativeness of accounting earnings in an emerging market with a unique institutional setting. Separating equity ownership from the control of corporate decisions is fundamental to contemporary firm theory (Jensen and Meckling, 1976; Fama, 1980; Fama and Jensen, 1983) in explaining principal-agent conflict of interest. It is predicted that, when managers hold less equity in the corporation, incentives arise for managers to pursue non-value-maximizing behaviors such as shirking and perquisite taking. Accordingly, contracts are written to restrict managers’ value-reducing behavior when ownership and control are distinct. This theory on the separation of ownership and control is utilized to formulate the hypothesis predicting how the informativeness of accounting earnings in explaining stock returns varies systematically with the level of managerial ownership in the corporation. This paper makes several contributions. First, several recent accounting studies (Ball, Kothari, and Robin, 2000) have provided evidence that, in addition to accounting standards, features of the institutional environment such as corporate governance as well as legal and financial systems can also explain the differences in the properties of accounting information across countries. Further, the paper extends the work by examining managerial ownership of financially distressed and non-financially distressed firms as a channel that influences reporting quality. This research should have implications for economic reformers and regulators who are striving to improve corporate governance and transparency in emerging economies. In general, the results suggest that managerial ownership affects the properties of accounting information. This paper is organized as follows: Section I presents the literature review of managerial ownership, earnings management and financial distress. Section II describes hypothesis development with respect to the relation between managerial ownership and accounting measures. In section III, the data and sample used in this study are discussed. Section IV discusses the methodology and empirical results. Section V concludes the study. I. Literature Review 1.1 Managerial ownership and informativeness of earnings Given that managers are not equity owners, the informativeness of earnings can be explained by examining the conflict of interest within two distinct groups: (1) a firm’s insiders and its outsiders and (2) a firm’s management and its shareholders. Warfield, Wild and Wild (1995) shows that managerial ownership is positively correlated with earnings explanatory power for returns and inversely related to the magnitude of discretionary accounting accruals. When managerial ownership is low, increased demand for accounting-based constraints motivates managers strategically to choose accounting policies and accounting accruals that mitigate accounting-based contractual restrictions. However, Morck, Shleifer and Vishny (1988) shows that high managerial ownership implies sufficient voting power to guarantee future employment but insufficient incentive to motivate managers in making valuemaximizing decisions. The finding that informativeness of earnings improves as managerial ownership increases is based on Jensen and Meckling’s (1976) agency theory, which is explained by the separation of corporate ownership and control. Jensen and Meckling (1976) predict that when managers hold less equity in the corporation, incentives arise for managers to pursue non-value-maximizing behaviour. Accordingly, contracts are written to restrict managers’ value-reducing behaviour when ownership and control are distinct. In contrast to Warfield, Wild and Wild (1995), Yeo, Tan, Ho and Chen (2002) shows that the informativeness of earnings does not always increase with managerial ownership. In a study of Singaporean firms, Yeo, Tan, Ho and Chen (2002) document a nonlinear relation between managerial ownership and earnings informativeness. Specifically, they find that managerial ownership is positively related to earnings informativeness at low levels of ownership. At high levels of ownership, however, the relation becomes negative. The difference can be explained by the entrenchment effect at high levels of ownership concentration (Morck, Shleifer and Vishny, 1998). As controlling owners are entrenched by their effective control of the firm, decisions they make that deprive the rights of minority shareholders are often incontestable, especially within a weak legal system and an ineffective corporate governance mechanism. Moreover, due to complicated pyramidal and cross-holding ownership structures typical in East Asian companies, a significant number of controlling owners in the region actually possess more control than is indicated by their equity ownership stakes, further exacerbating the entrenchment effect. This entrenchment effect in the ownership structure potentially affects the quality of a firms’ financial reporting. Because the controlling owner oversees the accounting reporting policies and is perceived to have strong opportunistic incentives to hold up minority shareholders, the market expects that the owner will not report accounting information accurately. Clasessen et al. (2002) reports the concentrated control and the divergence between ownership and control in public corporations in eight East Asian economies to diminish firm value, indicating the economic significance of the agency problem associated with ownership structures. Consistent evidence is also found in several other studies. La Porta et al. (2002) examined over 300 firms from 27 wealthy economies and found firms having a higher ownership by controlling owners to have higher valuation. Fan and Wong (2002) results also are consistent with the explanations of concentrated ownership and the associated pyramidal and cross-holding structures in creating agency conflicts between controlling owners and outside investors. Consequently, controlling owners are perceived to report accounting information with results serving their self–interest. The ensuing credibility deficit is reflected in investor skepticism regarding the firm’s earnings reports. Moreover, concentrated ownership is associated with low earnings informativeness as ownership concentration prevents leakage of proprietary information about a firm’s rent-seeking activities. 1.2 Earnings Management Schipper (1989) defines earnings management as a “...purposeful intervention in the external financial reporting process, with the intent of obtaining some private gain”. Management can use discretional accrual either to signa l private information or to manipulate earnings. To the extent that management uses its discretional accruals to manipulate earnings, data will become less informative. Rangan (1998) and Teoh et al. (1998) show that managers appear to overstate earnings prior to the issuance of seasoned equity offerings. Sweeney (1994) finds that managers make such choices when they approach potential violations of their debt covenants with outside creditors. Further, there are incentives for insiders to maintain information asymmetries to prevent outsider interference in their rent-seeking activities. Managers and controlling owners attempt to gain privately by misusing financial reporting discretion to conceal the firm’s true performance from outsiders. Masking firm performance creates in effect, information asymmetries between insiders and outsiders, weakening outsiders’ ability to monitor and discipline insiders. The information asymmetry between insiders and outsiders reduces both the likelihood of insider replacement and the threat of firm takeovers (Jensen and Ruback, 1983; Shleifer and Vishny, 1989). Moreover, insiders become entrenched in diverting profits by consuming perquisites and shirking and building empires at the expense of shareholders (e.g., Jensen and Meckling, 1976). Therefore, insiders engaged in asset diversions have the incentive to manage both the level and variability of reported earnings to reduce the likelihood of outside intervention given that outsiders rely on earnings in financial reports to monitor a firm’s performance, and excessive variability in earnings alarms is due cause for outsider intervention. There exist many studies on earnings management. The literature attempts to understand why managers manipulate earnings, how they do so, and the consequences of their behavior. There is broad interest in these findings. The starting point for measuring discretionary accrual is total accrual, assuming the generation of the non-discretionary component. According to Dechow (1995), there are several models for the process that generates non-discretionary accruals (Healy, 1985; DeAngelo, 1986; Jones, 1991; Dechow and Sloan, 1991). The most commonly used model in the literature is the model by Jones (1991). This paper focuses on discretionary accruals as a measure of manager earnings manipulations during import relief investigations under the notion that total accruals capture a large portion of manager manipulation. Insiders’ financial reporting choices can smooth earnings as these bookkeeping entries are at the insiders’ discretion and can hide real changes in the firm’s economic performance. Insiders that have incentive to avoid fluctuations in earnings can use both real operating decisions and financial reporting discretion to smooth variations in economic income. This smoothing measure captures the degree of using discretion to alter the “accounting” component of reported earnings, namely accruals, in order to reduce the variability of operating earnings. 1.3 Financial Distress Traditionally, the literature in financial economics has portrayed financial distress as a costly event important in determining firms’ optimal capital structures. Financial distress is seen as costly because it creates a tendency for firms to do things that are harmful to debt holders and non-financial stakeholders, impairing access to credit and raising the costs of stakeholder relationships. However, more recent studies have argued that financial distress can improve corporate performance and advocate changes in corporate form that are financed primarily with debt. Therefore, financial distress can cause significant losses in some cases and motivate value-maximizing choices in others. It is difficult to quantify the overall costs and benefits of financial distress. Altman (1984) attempts to measure the indirect costs of financial distress by examining a sample of firms that later went bankrupt. Opler and Titman (1994) examines the indirect costs of financial distress in a way that minimizes the problem of reverse causality. Their finding indicates that highly leveraged firms lose market share to their less leveraged competitors in industry downturns in a way that is consistent with a number of different interpretations. One possibility suggests that this phenomenon reflects reluctance by customers to do business with financially distressed firms. Wruck (1990) defines financial distress as a situation where cash flow is insufficient to cover current obligation. In pursuing their own interests, claimants have incentives to present biased and inaccurate data as though the data were unbiased and accurate. The result is a conflict of interest between a firm’s insiders and its outsiders. The agency problem is an essential element of the contractual view of the firm by Jensen and Meckling (1976), and Fama and Jensen (1983). Financial distress is resolved in an environment of imperfect information and conflicts of interest by either private workouts or legal reorganization. In order to reduce the likelihood of creditor interference when a firm is under financial distress, firm insiders have incentive to manage the information to hide the firm’s true economic performance. Most research on financial distress is focused on distressed costs and financial restructuring, but it is possible that financial distress could result in beneficial outcomes. The legal rules of bankruptcy create conflicts of interest among claimholders. These conflicts lead to complex information and inference problems for claimholders that try to value and intervene in a distressed firm. According to Kose John (1993), a firm is in financial distress at a given point in time when its liquid assets are insufficient in meeting the current requirements of its hard contracts. If these payments are not made on time, the firm is considered in violation of the contract, allowing its claimholders to act on specified and unspecified legal recourse to enforce the contract. Since financial distress results from a mismatch between currently available liquid assets and the current obligations of its “hard” financial contracts, mechanisms for managing financial distress rectify the mismatch either by restructuring the assets or restructuring the financial contracts (or both). II. Managerial Ownership, Earnings Informativeness and Discretionary Accruals This section discusses how ownership structure and financial distress affects accounting earnings and shapes the firms’ agency problems, leading to a hypothesis pertaining to the relationship among ownership structure, earnings informativeness and earnings management under a situation of financial distress. 2.1 Managerial ownership and discretionary accruals The ownership of companies in Thailand is typically concentrated in the hands of large shareholders. This concentrated control is achieved through complicated ownership arrangements, i.e., stock pyramids and cross-shareholdings like those used in other countries in East Asia. The separation of ownership from control has given managers incentive to take 2 An example of a hard contract is a coupon debt contract that specifies periodic payments by the firm to the bondholders. actions in their own best interest rather than those of the firm’s owners. This issue has resulted in the use of contracts that normally involve accounting-based constraints in determining management rewards and monitoring and controlling the actions of management. Even though contracts can be used to monitor the actions of management, they do not eliminate all possibilities of management’s non-value-maximizing behaviour, as implementing such a “perfect” system would be costly in terms of contract and monitoring control. Just as ownership structure describes a firm’s agency problems, so too does it impact firm reporting. Managerial compensation builds market expectations that managers will capitalize upon the latitude permitted by contracts and accepted accounting procedures in reporting accounting numbers. As the separation of ownership and control leads to agency cost, the theory predicts lower ownership to be associated with both increased contractual constraints that are often denominated in accounting numbers and management motivation to either relax restrictions or capitalize upon incentives. This yields the hypothesis that the informativeness of accounting earnings as an explanatory variable for returns is systematically related to the level of managerial ownership. After the economic crisis, many companies found themselves in financial distress to the extent that debt restructuring or bankruptcy court was required. Although, creditor intervention serves as another monitoring control on manager actions that should lower the non-valuemaximizing behavior, in reality it is never completely eliminated. Consequently, I expect the positive relationship between earnings’ informativeness and ownership for financially distressed firms to be smaller than that fo r non-financially distressed firms. For example, the coefficient of ownership-financial distress-earnings interaction is predicted to be negative, but the summation of coefficients for this term and ownership-earnings interaction term is still positive. Moreover, the relationship between managerial ownership and informativeness of earnings should remain unchanged even after adding other variables. The sign for each coefficient preceding each additional variable should be consistent with the sign predicted in previous studies in which a firm’s returns carry a negative relationship with its size, its debt level, and its earnings’ variations. According to size effect theory, risk is associated with the level of debt and the level of earnings’ variation. Return also should have a positive relationship with risk, growth and earnings persistence—the higher the risk, the higher the return. 2.2 Discretionary accounting accruals under financial distress Jensen and Meckling’s (1976) agency theory predicts the separation of ownership and control, and the consequent manager-owner incentive problems, to be related to the extent of contractual constraints denominated in the accounting numbers. Therefore, the second hypothesis is that the magnitude of adjustments in managers’ accounting choices is related to managerial ownership. When managers hold less equity in the corporation, incentives arise for managers to pursue non-value-maximizing behaviour. Accordingly, contracts are written to restrict managers’ value-reducing behaviour when ownership and control are distinct. The increase in demand for accounting-based constraints motivates managers strategically to choose accounting policies and determine accounting accruals in an attempt to mitigate accounting-based contractual restrictions. Agency conflict theory states that information asymmetry between insiders and outsiders reduces the likelihood that insiders are replaced and outsiders intervene. Furthermore, insiders are responsible for the quality and credibility of financial reports, and are therefore the ones who may engage in earnings management to affect these reports to reduce such interventions. This motive indicates that an incentive exists for insiders to manage earnings reports. Therefore, when an owner effectively controls a firm, the owner also controls the production of the firm’s accounting information and reporting policies. I hypothesize that managerial ownership is positively correlated with earnings explanatory power for returns and inversely related to the magnitude of discretionary accounting accruals. Moreover, the relationship between managerial ownership and the magnitude of discretionary accounting accruals should remain unchanged even after adding other variables. There should be a positive relationship between risk and discretionary accruals. The higher risk, the higher discretionary accruals that management is likely to use in order to reduce outsiders’ intervention. Additionally, discretionary accruals possess a negative relationship with each of these variables: a firm’s size, its debt level, its growth, its earnings’ variation, and earnings’ persistence. The negative relationship between size and discretionary accruals should be consistent with the Chaney and Jeter (1992) interpretation that the market perceives larger firms’ earnings as more relevant due to increased scrutiny of these firms’ accounting choices. Therefore, large firms should have lower discretionary accruals; a higher debt level implies a higher monitoring role by creditors, which should lower discretionary accruals. The greater the company’s growth, the lower the risk of intervention by outsiders and the level of discretionary accruals. These relationships also explain earnings’ variation. In conclusion, it is predicted that Thai firms with a high managerial ownership structure are more likely to manage earnings than firms with low managerial ownership. Therefore, there should be a positive relationship between earnings’ informativeness and managerial ownership after taking into consideration of principal-agency conflict theory and information asymmetry. Moreover, it is hypothesized that the effect of ownership on earnings explanations is lower in financially distressed firms due to its creditor’s use of earnings in monitoring. Another hypothesis relates to how financial distress affects the relation between managerial ownership and the incentives to manage earnings disclosed in financial reports by using discretionary accounting accruals. III. Data The sample consists of companies listed on the Stock Exchange of Thailand during 1998-2000. The ownership data are manually collected from the companies’ annual report and the ISIM database, compiled by the Stock Exchange of Thailand while earnings and stock returns data are drawn from ISIM database. To be included in the sample, companies must have met the following selection criteria: (1) available ownership data fo r the period 19982000; (2) available data necessary to compute stock returns and other variables; (3) not a classified as banks or financial institutions; and (4) not have negative earnings in any of the sample years. These criteria yield 69 companies with a total number of 207 firm-year observations. To be classified as a financially distressed company, the firm’s annual report must contain the auditor’s opinion report that has an emphasis paragraph showing that the company has a going concern problem by entering either into the debt restructuring process or bankruptcy. This criterion yields 9 financially distressed companies from 69 sample size. IV. Methodology and Results The average proportion of managerial ownership in the sample is 36.7 percent with a median of 42.6 percent. From the data, 25 of the companies (36.23 percent of the sample) reported zero managerial ownership while 14 companies (20.29 percent of the sample) reported 5 to 20 percent managerial ownership. Managers owned 20 percent or more of 28 companies in the sample (40.58 percent of sample). Table1 provides descriptive statistics for all variables used in the analyses. -----------------------------------Insert Table 1 Here -----------------------------------4.1 Explanatory power of earnings conditional on managerial ownership The hypothesis predicts that the informativeness of earnings is systematically related to the level of managerial ownership. One measure of the informativeness of earnings is its explanatory power for returns (see Table 2). Consistent with results from previous research, the correlation between earnings and returns for the entire sample irrespective of ownership is positive (0.129) and statistically significant. However, the magnitude of the correlation between earnings and returns declines as managerial ownership increases. Specifically, only the correlation in the 0-5 percent ownership range is positive and statistically significant. The results suggest that earnings are only informative when managerial ownership is low. High levels of managerial ownership can result in sufficient voting power to guarantee future employment and increased earnings management in line with the entrenchment effect as indicated by Morck, Shleifer and Vishny (1988). -----------------------------------Insert Table 2 Here -----------------------------------The second test of earnings informativeness is the examination of the variation in the earnings coefficient (the slope coefficient from regression of returns on earnings) conditional on managerial ownership (see Table 3). To test for the differential in earnings informativeness, as reflected in the earnings coefficient and conditional on ownership, the following pooled least squares regression model with an ownership interaction term is used: R i,t = 0 γ + 1 γ *E i,t /P i,t -1 + 2 γ OWN i,t + 3 γ * OWN i,t *E i,t /P i, t -1 + t i , ε (1) Where Ri,t is the return of firm i for the twelve-month period starting nine months prior to the fiscal year-end t and extending through three months after the fiscal year-end t. Ei,t is earningsper-share of firm i at the end of period t. Pi,t -1 is the price-per-share of firm i at the end of period t-1. OWN i,t is the percentage of managerial ownership of firm i for period t. The 3 γ parameter measures the interaction between ownership and earnings’ informativeness reflecting the extent to which the informativeness of earnings is affected by ownership levels. -----------------------------------Insert Table 3 Here -----------------------------------From the results, 3 γ is negatively significant at 0.05 level. This negative relation between managerial ownership and earnings’ explanatory power of returns can be explained by agency conflicts resulting from controlling owners’ manipulation of earnings for outright expropriation. In addition to managerial ownership, financial distress is added to the model. The test relies on the following pooled least squares regression model where the financial distress factor is also one of interaction to ownership and earnings: R i,t = 0 γ + 1 γ *E i,t /P i,t -1 + 2 γ OWN i,t + 3 γ * OWN i,t *E i,t /P i, t -1 + 4 γ OWN i,t * FINi + 5 γ OWN I,t * FINi * E i,t /P i,t-1 + t i , ε (2) where FINi is equal to one when the company is in financial distress and zero otherwise. Financial distress is defined as the condition in which the cash flow is insufficient to cover current obligations. These obligations can include unpaid debts to suppliers and employees, actual or potential damages from litigations, and missed principal or interest payments under borrowing agreements (Wruck, 1999). As no Chapter 7 and Chapter 11 bankruptcy proceedings exist in Thailand as they do in the United States, the financially distressed firm is defined as a firm that is in the debt restructuring process as disclosed in auditor’s opinion report with emphasis paragraph. Table 4 presents the result of a pooled least squares regression estimation of Equation (2). -----------------------------------Insert Table 4 Here -----------------------------------It is hypothesized that the relation between earnings’ informativeness and managerial ownership differs between firms that are in financial distress and those that are not. The sign of 3 γ remains negative and statistically significant. The coefficient of financial distressmanagerial ownership is negative and statistically significant indicating that, for financiallydistressed firms, high managerial ownership leads to low stock returns. Furthermore, the financial-distress-managerial and ownership-earnings interaction coefficient is positive and statistically significant. A possible explanation is that monitoring by creditors lower the entrenchment effect of ownership in the earnings coefficient for a financially-distressed firm. The results can be explained by the notion that information asymmetry between insiders and outsiders reduces the likelihood that insiders are replaced by outsiders and intervention occurs. Therefore, insiders have incentive to manage earnings presented to outsiders, namely the firm’s minority (or arm’s length) shareholders and creditors. While a negative relation exists between earnings’ informativeness ( 3 γ equals –2.738 is significantly greater than zero at the 0.001 level), the positive 5 γ (3.684) parameter demonstrates that the effect of ownership on the earnings coefficient is significantly higher for financially distressed firms. The increased effect of ownership for financially distressed firms on earnings’ explanatory power of returns can be explained by monitoring role of creditors. The earnings-ownership interaction parameter for financially distressed firm equals 0.946( 3 γ + 5 γ ) which changes the effect of managerial ownership on earnings’ explanatory power from negative to positive. This implies significant creditor roles in improving earnings’ explanatory power of returns. In the next model, other variables are included to control for other factors that can affect stock returns. These variables are firm size, systematic risk, leverage, growth opportunities, earnings variability, and earnings persistence. The following pooled least squares regression model is used: Ri,t = 0 γ + 1 γ *E i,t /P i,t -1 + 2 γ OWN i,t + 3 γ * OWN i,t *E i,t /P i, t -1 + 4 γ OWN i,t * FINi + 5 γ OWN i,t * FINi * E i,t /P i,t -1 + 6 γ SIZEi,t + 7 γ RISKi,t + 8 γ DEBTi,t + 9 γ GROWTHi,t + 10 γ VARi,t + 11 γ PERSi,t + t i , ε (3) SIZE is the natural logarithm of a firm’s market value of equity. RISK is a firm’s 24-month market model systematic risk. DEBT is the firm’s ratio of total debt to total assets. GROWTH is the market value of equity scaled by book value. VAR is the variability of earnings for the eight quarters prior to the sample year. PERS is persistence of earnings as measured by the first order autocorrelation in earnings for the eight quarters before the sample years. All other variables are as defined in Equation (1) and Equation (2). -----------------------------------Insert Table 5 Here -----------------------------------Table 5 shows that the inclusion of control variables does not alter the relationship between managerial ownership and earnings’ informativeness. Consistent with Table 3 and 4, the relation between managerial ownership and earnings’ informativeness remains negative and statistically significant. This result implies that earnings’ informativeness is lower for firms with higher managerial ownership, ceteris paribus. Among the control variables, earnings persistence coefficient is positive and statistically significant. Overall, the negative relationship between earnings’ informativeness and managerial ownership is robust to the inclusion of these other factors. Further, the parameter on financial-distress-managerial ownership ( 4 γ ) is negative while the parameter of financial-distress-managerial ownershipearnings ( 5 γ ) is positive and statistically significant. Thus, the effect of managerial ownership on the earnings coefficient is significantly higher for a financially distressed firm. The increased effect of ownership for a financially-distressed firm can be explained by the possibility that the management of a firm in financial distress tries to reduce the possibility of further creditor intervention after the completion of debt restructuring negotiations by engaging in earnings management. A final test is performed to control for the effect of the sample years. A series of year dummy variables is added to the model as shown below: Ri,t = 0 γ + 1 γ *E i,t /P i,t -1 + 2 γ OWN i,t + 3 γ * OWN i,t *E i,t /P i, t -1 + 4 γ OWN i,t * FINi + 5 γ OWN i,t * FINi * E i,t /P i,t -1 + 6 γ SIZEi,t + 7 γ RISKi + 8 γ DEBT i,t + 9 γ GROWTHi,t + 10 γ VARi + 11 γ PERSi + 12 γ YEAR 1999 + 13 γ YEAR 2000 + t i, ε (4) In the end, empirical evidence presented in Table 5 shows that the nature of the relation remain intact after controlling for the year effect. 4.2 Managers’ accounting choices conditional on ownership This section presents the results of tests directed at revealing a manager’s accounting choices through the manager’s reliance on discretionary accruals. The magnitude of discretionary accrual adjustments serves to measure the extent that a manager adjusts reported accounting numbers. Tests are performed, conditional on the level of managerial ownership. As before, the greater likelihood of accounting-based contractua l constraints in firms with greater separation of ownership and control yields the hypothesis that managers’ accounting accrual adjustments are positively related to the level of managerial ownership. There are additional factors in managers’ accounting choices as ownership structure is not the only determinant of their accounting choices. Accordingly, a pooled cross-sectional time series regression model is used to investigate the joint interaction of management ownership, financial distress and discretionary accruals. DAi,t = 0 γ + 1 γ * OWN i,t + 2 γ * OWN i,t * FINi + 3 γ *SIZEi,t + 4 γ * RISKi + 5 γ *DEBTi,t + 6 γ GROWTHi,t + 7 γ * VARi + 8 γ * PERSi + t i , ε (5) Where DAi,t is the magnitude of discretionary accruals of firm i for the period t. To define the discretionary accrual, the total accrual and non discretionary accrual must be defined. Total accrual follows the method used in Dechow et al. (1995) which is consistent with Jones (1991), the total accruals are computed as: TAit = ( ∆CAit − ∆Cashit − ∆CLit + ∆STDit − Depit )/At-1 (5.1) Where: TAit = Total accrual ∆CAit = change in total current assets ∆Cashit = change in cash/cash equivalents ∆CLit = change in total current liabilities ∆STDit = change in short term debt included in current liabilities Depit = depreciation and amortization expense Discretionary accruals are then estimated by subtracting the predicted level of nondiscretionary accruals (NDA) from total accruals obtained from equation (5.1) standardized by lagged total assets: