Earnings Management and Dividend Policy An Empirical comparison between Pakistani Listed Companies and Chinese listed Companies

Transcription

1 International Research Journal of Finance and Economics ISSN Issue 35 (2010) EuroJournals Publishing, Inc Earnings Management and Dividend Policy An Empirical comparison between Pakistani Listed Companies and Chinese listed Companies Syed Zulfiqar Ali shah ACMA, MBA, MS(Finance), Ph.D (Contd.), Mohammad Ali Jinnah University Assistant Professor International Islamic University Hui Yuan Associate Professor, Xinhua School of Banking and Insurance Zhongnan University of Economics and Law, Wuhan, Hubei China Nousheen Zafar PhD scholar, Mohammad Ali Jinnah University Islamabad Abstract The study is an effort to fiind out the impact of earnings management on dividend payout policy for two coutries that is Pakistan and China. A set of listed Companies in both of the countries have been investigated to analyze the relationship for the year 2003 to 2007 in Pakistan and 2001 to 2007 in China. Dividend policy has been measured by dividend payout. Whereas earnings management has been measured by calculating discretionary accruals. Modified Cross Sectional Jones Model (1995) has been applied to data to determine the Earnings management. Results of Common effect model shows that for both the countries, earnings management has no impact on dividend pay out policy. This enables us to accept two of our hypothesis. However on the basis of these results our third hypothesis has been rejected stating that relationship between earnings management and dividend payout is not same in both countries. Key words: Dividend policy, Earnings Management Jel Classification Codes: G35, M41 Introduction Corporate managers in their daily routine of life are exposed to a number of crucial decisions regarding finances of a company. Among all such decisions, Dividend payout policy is one of the most important financial decisions that come across (Baker and Powell, 1999) as it is perceived to be a symbol of good financial health of accompany. Miller and Modigliani (1961), by assuming perfectly efficient market, proved that firm value can not be increased by changing dividend policy of a firm. However as perfect market do not exist in reality, literature provides number of theories e.g. Clientele theory, bird-in-hand theory, agency theory, and signaling theory that shows dividends increases the value of equity and thus investors are more attracted towards dividend paying firms.

2 International Research Journal of Finance and Economics - Issue 35 (2010) 52 As dividends are paid out of the net earnings of a firm, there could be ways to look at this matter. One view is that dividends can be used as a predictor of earnings whereas another view is that earnings can also be used as a predictor of dividends. Thus both of these concepts are interrelated as both determine each others value. This helps to understand why managers of a firm are more interested in maximizing the firm s earnings. Earnings are the most important item to signal how much firm is involved in value adding services. With this much increased emphasis over importance of earnings for a firm, it is no wonder if company management takes vital interest in the manner their earnings are reported. An increase in earnings depicts the increase in overall value of a company and vice versa (Lev, 1989). Particularly to conceal the losses of a company, earnings are managed to show beneficial situation (Hayn, 1995). This presents the idea of earnings management that is use of accounting choices to amend reported earnings for the sake of managers benefits. Or earnings management can be defined as reasonable as well as legal decision making and reporting of financial results, by managers, with the intention to achieve stability in earnings. According to an estimate 8-12 percent firms with small pre-managed earning decreases manipulate their earnings to give impression of earning increases and percent firms with small pre-managed losses manipulate earnings to show positive earnings (Burgstahler, 1997). Evidences are found that dividend payout decision is not only the outcome of operating performance or the growth opportunities of the company. Lie and Li (2005) stated that the decision to increase or decrease dividend and its magnitude is determined by the premium placed by capital markets on dividends. However Savov and Weber (2006) rejected this hypothesis and claimed dividend increase as a compensation for the poor returns of the shareholders. In light of these results we may conclude that it is not necessary for the managers to manage their reported earnings for signaling high value of their firm as dividend payout can be effected by various factors as well. Present study is a comparative study of earnings management practiced by Pakistani and Chinese listed companies and its impact on their dividend policies. Both are neighboring developing economies however growth rate of China economy is far better than Pakistan. Both of these countries have recently started emphasizing on corporate governance practices, despite of very different business environment of two countries. Earnings management is supposed to be reduced by increased truly implemented corporate governance practices. In Pakistan, as corporate governance can not play its role in true sense (mainly due to family owned business and shareholding concentration), practice of earnings management is supposed to be high. In case of China, literature shows that agency problem prevails in Chinese listed companies which is again a result of poor corporate governance. This also serves as a cause of earnings management. Dividends are considered as a way to mitigate the agency problem by giving shareholders their share. However in Pakistan dividends are considered more like a signal of firm future growth. This study is mainly an effort to find out the trends of business in two neighboring courtiers as they operate in two different environments. China s pace of development is far ahead as compared to us. The results found might help the policy makers of both the countries to find the loop holes and to craft a better policy for future. Also if earnings management proves to have no impact on dividends, it will increase the confidence of investors in both the countries as they would consider their benefits safe. Next section provides us with the evidence about determinates of dividend policies and Earnings management in China and Pakistan. Also relationship of two is discussed in general. Literature review Strong corporate governance practices reduces the chances of earnings management in a country. However some times earnings management is needed to maintain the dividend payout ratio of the firm as it is considered a signal of future growth perspectives of a firm in market (Miller and Rock, 1985; Ambarish et al, 1987). Also dividends are given importance because they are believed to tackle the agency problems rising between corporate insiders and outsiders (Gomes, 1998; Zwiebel, 1996).

3 53 International Research Journal of Finance and Economics - Issue 35 (2010) Outside investors always prefer dividends over retained earnings because they fear that retained earnings might be used by insiders for their own benefits against the interest of outsiders. As much the reported earnings of a firm are, as much the dividend expected by the shareholders and thus an overall increase in share value. However if dividend payment becomes a constraint for managers, they might manage to show reduced earnings. U.S. Real Estate Investment Trusts (REIT) are required by federal law to distribute 90% of their taxable income as dividends to common shareholders. It is feasible for them to manage their real earnings to meet regulatory dividend restriction by reducing revenue and increasing expenses. Both of these activities are a source of reduction in REIT s taxable income (Edelstein et al, 2008). Kato et al (2002) in his study provided the evidence of dividend motivated earnings management. They used pure accruals by defining them as accruals and deferrals (A&D) by Japanese banks under the governmental guideline. As a result of this guideline, 40% or less of net incomes was paid by banks as dividend until August The results revealed direct relationship between earnings management and dividend payment policies of Japanese banks. According to them when dividend driven motivation behind earnings management disappears, the extent of income increasing A&D management becomes significantly low. Kasanen et al (1996) were the first one to promote the idea of dividend driven earnings management. According to their implicit contracting view for non-financial firms of Finland, major institutional shareholders demand high yields on their stock holdings and thus expect a smooth dividend stream. This motivates the companies for earnings management to show high enough income for dividend payout. Thus in Finland during , reported earnings were dependent on the dividend-based target earnings. Savov (2006) analyzed earnings management to dig out the relationship between the reporting behavior of corporations and their investment and dividend policies. For a sample of German firms they held that firms having high investments tend to report more discretionary accrual in their earnings and vice versa. However dividend payout was found to be negatively related to both of these two activities. Another finding was that in case of overvalued firms, there is increased sensitivity of earnings management to permanent earnings. Case of Pakistan Pakistan is a developing country and the concept of corporate governance is not very old for it. With most of the large corporations owned and controlled by families and with family members holding key managerial positions, however, the major agency problem exists not between the management and owners in general, but between the management (the controlling family) and minority shareholders. Other corporate governance practices are also not complied with their true essence in Pakistan. It provides an ideal environment to managers for practicing earnings management to benefit themselves on the cost of shareholders. A very recent study conducted by Shah et al (2009) examined the relationship of earnings management with CG practices in context of Pakistan. Results revealed that increase in quality of corporate Governance practices leads to an increase in discretionary accruals. Means there exist a positive relationship among two. The reason for such contradictory results might be the use of very subjective CG score card for research purpose. Leuz et al (2002) in their study showed that presence of outside investor protection limits the chances of earnings management as it confines the abilities and incentives of insider managers to obtain private control benefits. This study was held across 31 countries between 1990 and Pakistan is one of these countries. Calculations of Leuz et al (2002) shows that aggregate earnings management score, (average of earnings smoothing measures and earnings discretion measures) for Pakistan is The highest score is for Austria i-e 28.3 and lowest score is of USA i-e 2.0. All these scores shows that earnings management in Pakistan is neither very strong nor very weak rather it is on moderate level.

4 International Research Journal of Finance and Economics - Issue 35 (2010) 54 Ahmed and Javid (2009) in their study on dynamics and determinants of dividend policy in Pakistan held that dividend payout for Pakistani listed non-financial firms depends upon both the change in dividends and change in net earnings. However dividend sensitivity towards current earnings is higher than the past dividends. Number of other determinants of dividend was also found by the study. Due to the family owned businesses in Pakistan, nepotism is largely followed at the time of appointment of directors and thus they are paid heavy pays and fringe benefits. This extra cost reduces the net profit of the firm and so firms find it difficult to declare dividends. In all this game who suffers are the minority share holders (Mehar, 2002). In Pakistan earnings management has a clear incentive for the managers because current earnings are considered as the more appropriate and sensitive base for the dividend payout determination. This implies that higher the reported earnings, higher the dividend payout. Case of China China is a rapidly emerging economy in world. Keen interest towards improving the performance of Chinese listed firms and protection of minority shareholders has given birth to the corporate governance practices in china. Most commonly used corporate governance practices in China are concentrated ownership structure, management- friendly boards, inadequate financial disclosure, and inactive take-over markets (Liu, 2005). Most of companies listed in China are those which were previously owned by state. Such firms lack the supervision mechanisms and real owners of the firm. In this environment managers get the chance to exercise their control in their own favor on the cost of shareholder s benefit. This gave birth to agency problem between share holders and managers. Dividend is a way to please shareholders by giving them their share from the profit. Thus dividend payout can mitigate the degree and possibility of this agency problem. For governance mechanism, Control based governance model is used in China in which major share holders, most likely the state, controls the listed firms by implying tight CG practices. This model gives all the powers in hands of controlling shareholders to expropriate the interests of minority shareholders. This shakes the public confidence on the stock market despite its fast growing pace. Thus any firm deviating from this model can perform better in terms of shareholder interest (Liu, 2005). However chances are high for a group controlled firm in China to be engaged in earnings management and to tunnel firm value (Jian and Wong, 2003). The term tunneling was used by Shleifer (2000) referring to the taking away of firm s resources for personal benefits by controlling shareholders. This includes the activities like absolute theft from funds, loan guarantees, selling of assets or products at lower than market prices etc. Controlling shareholders, willing to tunnel the firm value, have a very strong reason for earnings management as their prime objective is to hide their private control benefits from outside investors. thus it is concluded that earnings management is intrinsically related to tunneling in an atmosphere where chances of getting control benefits is high and chances to detect them is low (Liu, 2003). Institutional stock holding is believed to be a channel that compels the managers to pay out dividends. This stops the managers to hurt the interest of stockholders and thus agency problem is reduced. Chinese manufacturing listed companies, for the period 2001 to 2003, showed positive relationship between institutional ownership and payout of cash dividends (Li, 2007). Chances of earnings management arise whenever there exist weak corporate governance mechanism. An environment, where controlling stock holders (normally insiders) are in a position to use the firm resources in their benefit, gives birth to the idea that actual value of a firm must be concealed by the outside shareholders. Liu (2003) suggested that main reason behind earnings management in Chinese listed companies is the presence of agency conflicts between controlling shareholders and outside investors.

5 55 International Research Journal of Finance and Economics - Issue 35 (2010) Xiaopeng and Xuejie (2002) in their survey of listed companies of china gave some main reasons that serve as a motivation for the earnings management. These reasons as stated by him are: the exaggerating financial packing for initial public offering (IPO), for the purpose of rationed share acquisition, escape from being delisted, special treatment (ST), particular transfer (PT), inside dealing and the manipulation of stock prices. Generally dividend policy of a firm is determined by number of factors like earning level, share prices, financial ability of a firm etc. but in case of China s listed companies, dividend policies are mainly effected by strong information asymmetry and agency problem between managers and shareholders, along with the other financial factors. It is yet to be observed that whether earnings management impact dividend policy or not and if yes, to how much extent. Present study covers this aspect of relationship between two. Methodology The sample consists of 55 Chinese listed companies and 120 Pakistani listed companies. All the Pakistani companies are listed at Karachi stock Exchange however in case of China some companies are listed in shanghai stock exchange and some in Shenzhen stock exchange. We have collected the data of Pakistani listed companies from SBP (State Bank of Pakistan) annual balance sheet analysis, Companies annual reports and web sites and Data for Chinese companies has been collected from companies annual reports on their web sites. The entire sample consists of non financial companies. We have excluded financial companies as they have very different capital structure and profits. Also the companies for whom data was not available are excluded. For the purpose of study we have hypothesized that: H 1 = Earnings management has no impact on dividend policy in Pakistan H 2 = Earnings management has no impact on dividend policy in China H 3 = Relationship between earnings management and dividend policy is not same in Pakistan and China. Following variables have been selected to test the hypothesis on the basis of intensive literature review: - Dependant Variable Variable Abbreviation Description Dividend Payout DPO Dividend Paid/ Net Profit after Tax Independent Variables Earnings Management EM Calculated by using modified Jones Model (1995). Details are at the end of this table. Control Variables Return on Equity ROE Net Profit/ Shares Holders equity. Size of the firm Size Ln of total Assets following Scott and Martin (1975) Self Finance Ratio SFR Retained Earnings/ change in capital employed following John and Willaims (1985) and Ahmed and attiya (2009) We have calculated the earnings management using modified Jones model (1995) as follows: - Earnings Management (Dependant variable) Discretionary accruals have been taken as a proxy of earnings management by a number of researchers. And Discretionary accruals are calculates as the difference between total accruals and non discretionary accruals. Total accruals can be calculated by following two methods: - o Balance sheet approach (Healey (1985) & Jones (1991)) TAt = ΔCAt ΔCasht ΔCLt + ΔDCLt DEPt Where: ΔCA t is change in current assets in year t

6 International Research Journal of Finance and Economics - Issue 35 (2010) 56 ΔCash t is the change in cash and cash equivalents in year t ΔCL t is the change in current liabilities in year t ΔDCL t is the change in debt included in the current liabilities in year t DEP t is depreciation and amortization expense in year t o Cash flow statement approach TA t = N.I t - CFO t Where: TA t is total accruals in year t N.I t is Net Income in year t CFO t is cash flows from operating activities in year t We have used cash flow statement approach to calculate total accruals in this study. After calculating total accruals we calculated nondiscretionary accruals by using the following equation. 1 ΔREVt ΔRECt PPEt NDAt = α ( III) A + α + t 1 A α t 1 A t 1 Where: TA t is total accruals in year t scaled by lagged total assets Δ REV t is revenues in year t less revenue in year t-1 PPE t is gross property plant and equipment at the end of year t ΔREC t is net receivables in year t less net receivable in year t-1. All of the variables have been scaled by lagged total assets. A t-1 is total assets at the end of year t-1 α1,α2,α3 are firm specific parameters ε is the residual, which represents the firm specific discretionary portion of total accruals Discretionary accruals have been calculated by taking the difference between Total accruals and Non discretionary accruals. DAt = TAt NDAt Where:- NDA t is non discretionary accruals, DA t is discretionary component of accruals Regressing Discretionary accruals with dividend policy To test the hypothesis common effect model in panel data analysis has been used. DPO it = α + β1 (DA it ) + β2 (SFR it ) + β3 (ROE it ) + β4 (size of the F it ) + µ it Where: DPO is dividend payout DA is discretionary accruals ROE is Return on equity Size of the f is size of the firm. Results Following table shows the results of descriptive statistics: - Descriptive Statistics Descriptive statistics for Pakistan and China are as follows:

7 57 International Research Journal of Finance and Economics - Issue 35 (2010) Pakistan DPO DA Size SFR ROE Mean Standard Deviation Minimum E Maximum China DPO Size DA SFR ROE Mean Standard Deviation Minimum Maximum Descriptive statistics shows us that mean value for discretionary accruals is far higher in case of Pakistan i-e as compared to that in case of China i-e However standard deviation for discretionary accruals is also higher in case of Pakistan as compared to China. On the contrary, average dividend payout in China i-e is higher than that of Pakistan i-e and standard deviation for Pakistan is higher for dividend payout as compared to China. In order to check multi collinearity among variables, correlation test has been performed. Results indicate that in case of both the countries, no multicolinearity amongst independent variables exist. Correlation Matrix Pakistan China DPO DA size SFR ROE DPO Size DA SFR ROE DPO 1 1 DA size SFR ROE As mentioned in the above table, SFR (self finance ratio) is statistically significant with dividend payout but its direction is in the opposite direction in both of the countries. That is in Pakistan it is negatively related with dividend payout but in china it is positively correlated. Size is also slightly significantly impacting on dividend payout. No any other independent variable is significantly correlated with dependant variable. Presence of non correlation between independent variables proves the emancipation of model from multicolinearity. After correlation for in depth analysis we applied the panel data analysis technique. Results of the common effect model are presented as follows: -

8 International Research Journal of Finance and Economics - Issue 35 (2010) 58 Common Effect Model Pakistan China Variable Description Coefficient T statistics Coefficient T statistics Intercept ** * DA size * * SFR ** *** ROE F statistics R Square Adjusted R Square *significant at 95 % level of confidence **significant at 99% level of confidence ***significant at 90% level of confidence Above table shows that discretionary accruals does not have any impact on dividend payout in both of the countries. However coefficients for DA in both the countries move in positive direction. This result is contrary to the results of Savov (2006) who showed negative relationship between earnings management, investment and dividend payout. Also Kato et al (2002) showed direct relationship between two for Japanese banks. Based on our results we can say that earnings in both of the countries are not managed to avoid or announce the dividends. There could be a number of other intentions behind earnings management. On basis of these results we accept our first and second hypothesis. However third hypothesis is not accepted as our results for both countries goes in same direction. Size of the firm is positively related with dividend payout which is statistically significant relationship in Pakistan which shows that big companies announce more dividends. In china this relationship has been found negatively related with dividend payout and shows that small companies announce more dividend and big companies announce less dividend. Self finance ratio is showing the same trend as size in a way that it has different relationship in both of the countries. In Pakistan it has been found having negative relationship with dividend payout which seems logical as well because if a company would pay dividend it would not finance it self from its internal resources. But in China we have found positive relationship between these two variables which shows the companies which pay more dividends have more self financing. Here we can say that the companies keep balance between dividend payment and self retention. All of the other variables have been found insignificant relationship with dividend payout in both of the countries. R square for our study is very low i-e around 3% in case of China but in case of Pakistan it is around 93 % which is good. But as the data consists of time series and cross section data so, it may be an acceptable range in case of China as well. Significant F statistics in both of economies confirms the validity of the model. Conclusion After having the detailed econometric analysis we have come to the conclusion that: Discretionary accruals do not have any impact on dividend policy in both of the countries. Big companies announce more dividends in Pakistan while small companies announce more dividends in China. Control variable self finance ratio has also shown the significant impact in both of countries but in contrary direction. A very important behavior of firms have come in front of us after this study that earnings management although are present in both of the countries but they are not for the purpose of dividend announcement or dividend avoidance. It may be due to the reason of laws or presence of family controlled firms in both of countries which doesn t allow the manipulation of data to restrict them from

9 59 International Research Journal of Finance and Economics - Issue 35 (2010) dividend. These results can also help investors of both the countries to have faith on management especially in China. It is interesting to know that nature of relationship between earnings management and dividend payout is same for both the countries means Pakistan is also following same track as China has and thus can progress this way. Also these results are in accordance with the literature that showed various factors as determinants o f dividend payout in both the countries and earnings management was not among them. More over this study offers another way to look at the things that is to check out the impact of dividend payout on earnings management. It would be interesting to know whether dividend payout induces the firms to manage earnings. References [1] Baker, H. and G. Powell (1999), How Corporate Managers View Dividend Policy, Quarterly Journal of Business and Economics, Vol. 38, pp [2] Miller, M. and F. Modigliani (1961), Dividend Policy, Growth and the Valuation of Shares, Journal of Business, Vol. 34, pp [3] Burgstahler, D., I. Dichev, (1997), Earnings Management to Avoid Earnings Decreases and Losses, Journal of Accounting and Economics, 24:101 [4] Hayn, C. (1995): The information content of losses. Journal of Accounting and Economics. vol. 20, pp [5] Lev, B., (1989), On the Usefulness of Earnings and Earnings Research:Lessons and Directions from Two Decades of Empirical Research, Journal of Accounting Research, 27, supplement 1989, [6] Edelstein, Robert H., Liu; Peng, Desmond Tsang, (2008), Real Earnings Management and Dividend Payout Signals: A Study for U.S. Real Estate Investment Trusts, January 1, 2008, (CAAA) 2008 Annual Conference Paper. [7] Kato;Kazuo, Michio Kunimura,Yasushi yoshida, (2002), Dividend Behavior and Pure Accrual Management of Japanese Banks, Working paper no: [8] Kasanen, E., J. Kinnunen, and J. Niskanen, (1996), Dividend-based earnings management: Empirical evidence from Finland, Journal of Accounting and Economics, 22, pp [9] Lie, E., and W. Li, 2005, Dividend Changes and Catering Incentives, Working paper, [10] Henry B. Tippie College of Business, University of Iowa. [11] Savov, S., and M. Weber, 2006, Dividend Increases and Dividend Initiations: What Role [12] for Fundamentals and Market Movements, Working paper, University of Mannheim. [13] Savov; sava, (2006), Earnings Management, Investment, and Dividend Payments, Working paper, July 16, [14] Shah; Syed Zulfiqar Ali, Butt; Safdar Ali, Hasan ;Arshad, (2009), Corporate Governance and Earnings Management an Empirical Evidence Form Pakistani Listed Companies, European Journal of Scientific Research, Vol.26, No.4, [15] Leuz;Christian, Nanda;Dhananjay, Wysocki ;Peter D., (2002), Earnings Management and Investor Protection: An International Comparison, Working Paper, October 1999, Forthcoming Journal of Financial Economics. [16] Ahmed ;Hafeez, Javid ;Attiya, (2009), Dynamics and Determinants of Dividend Policy in Pakistan (Evidence from Karachi Stock Exchange Non-Financial Listed Firms), International Research Journal of Finance and Economics, Issue 25, [17] Liu; Qiao, (2005), Corporate Governance in China: Current Practices, Economic Effects, and Institutional Determinants, Working paper, Draft: May 9, 2005 [18] Li;Qiang, Huang;Guoliang, (2007), Influence of Institutional Ownership on Cash Dividend Policy of China Listed Companies, The Sixth Wuhan International Conference on E-Business - International Finance Track, 2007, [19] Liu; Qiao, Lu ;Zhou (Joe), (2003), Earnings Management to Tunnel: Evidence from China s Listed Companies, Working Paper, draft: August 2003.

10 International Research Journal of Finance and Economics - Issue 35 (2010) 60 [20] Shleifer, Andrei and D. Wolfenzon, (2002), Investor protection and equity market, Journal of Financial Economics, 66(1), [21] Jian, Ming, and T. J. Wong, (2003), Earnings management and Tunneling through related party transactions: evidence from Chinese corporate groups, working paper. [22] Mehar, Ayub, (2002), Corporate governance and dividend policy, MPRA Paper No. 619, available at: [23] Miller, Merton, and Kevin Rock, (1985), Dividend policy under asymmetric information, Journal of Finance, 40, [24] Ambarish, Ramasastry, Kose John, and Joseph Williams, (1987), Efficient signaling with dividends and investments, Journal of Finance, 42, [25] Gomes, Armando, (1998), Going public with asymmetric information, agency costs, and dynamic trading, Mimeo, Wharton School. [26] Zwiebel, Jeffrey, (1996), Dynamic capital structure under managerial entrenchment, American Economic Review, 86, [27] Xiaopeng; Zou and Xuejie; Chen, (2002), Earnings Management of Chinese Listed Companies: A Survey of Empirical Studies, China & World Economy, No:4,