Existence Subordinate Relation "Leader-Follower" in Investigation of Eearnings Management and Dividend Policies Scopes

Transcription

1 Available online at International Journal of Advanced Studies in Humanities and Social Science Volume 1, Issue 12, 2013: Existence Subordinate Relation "Leader-Follower" in Investigation of Eearnings Management and Dividend Policies Scopes Sarollah Eghdamimoghadam * 1, S. Yusef AhadiSerkani (Ph.D.) 2 *1 Department of Accounting, Science and Research Branch, Islamic Azad University, Qazvin, Iran 2 Departments of Accounting, Firoozkooh Branch, Islamic Azad University, Firoozkooh, Iran Abstract The present study attempted to find a relationship regarding the adherence of the following companies (follower) of the leader firms (Progressive) in earnings management and dividend policies field. The study aimed to investigate the status of earnings management and dividend policies in two groups of the leaders companies and the followers firms and identify and explain the relationship between the behaviors of firms, that the behavioral outcomes reflected in reported performance by of leader firms Behavior on the following firms is analyzed. To do this, data were gathered from seven Industries and 124 firms listed in TSE during 2004 to Research hypotheses were tested by multiple-variable analysis of regression and Validation tests. The results indicated that there is a positively significant correlation between management earnings in leaders and follower firms. Also, there is a positively significant association between the changes in announced earnings, retained earnings changes and changes in the ratio of dividends to earnings per share in leaders and following companies. According to finding such relationships, the results of this research could be the basis for future studies in expansion of relationship between the behavior of two companies and it is in the behavioral studies field. Keywords: Earnings Management, Dividend Policy, Leaders firms (progressive) and followers firms Introduction As the investors consider earnings as one of the important factors in their decision makings, earnings management is one of the challenging and attractive issues in accounting researches. Some of accounting standards including the standards of using accruals in accounting allow the managers judge and determine in financial reporting process of various methods. These judgments by managers and their authority in financial reporting process are called earnings management (Barth et al., 2001). For many years, accounting researches consider earnings management and its outcomes. The present study investigated the status of earnings management and dividend policies in leaders and followers firms. The purpose of the study is that behavioral effects reflected in reported performance by leader fir ms on behavior of follower firms are analyzed. This management Corresponding Author sadegh77_moghadam77@yahoo.com 2597 Page

2 behavioral is compatible with the social learning theory. It means that people decisions are based on previous actions (Jones, 1991). According to social learning theory, follower firms perceive the condition of leader companies in relation to expected performance of the industry as the inclination that follower firms show for discretionary reporting is affected by positive or negative reactions of market (optimistic or pessimistic behaviors)arising from earnings announcement by leader firms. The evaluation of earnings management and dividend policies in leader firms (progressive) and following companies (followers) and the study of subordinate relationship of leader-follower on 7 active industries in TSE were done with the purpose of transparency of the required relations. The main problem of the study focused on the point that whether earnings management of the defined companies as follower firms (with emphasis on accruals) is dependent upon earnings management applied by the firms with leadership capability in industry. On the other hand, is there any subordinate relation in dividend policies and retained earnings over time is observed for two mentioned groups? Theoretical basics of the study The approaches of theoretical frameworks of the present study based on its special nature and required scientific method are based on some of the theoretical views. This study emphasized on the methods of divided and earnings management and inference and generalization based on observations. Thus, it is based on deductive approach. On the other hand, as in this study the behavior of some of the companies to the report of information is investigated, it is classified in the framework of behavioral studies. This study evaluated the existence or non-existence of phenomena, without applying value judgments and it is discussed in the positive accounting researches field and in the framework of descriptive theories. In this study, we investigated the usefulness of presented data in report of the leader firms on decisions taken by following companies. Thus, decision making model approach is considered and it has directive-inductive nature (Easton et al., 2011). These approaches with efficient market theory and special conditions of capital market efficiency in Iran (capital market researches) are theoretical framework of the present study and they are explained later. According to efficient market theory, it is expected that in capital markets with good efficiency, the information have final effect on prices and abusive use of information is not possible. In such markets, the pricing process is the result of balanced movement of real demand and supply for securities. In capital markets with incomplete efficiency, due to the shortcoming of required data to take investors decision making, the mass movements affected the market prices considerably. In such markets, we observe participation of the groups with mass movements to take investment decisions. Some of interveners guide market and they are considered as market leaders. Such approaches show that the present study is indirectly dependent upon most of the theoretical basics and it is theoretically rich namely those following the generalization of mass behavior of investors (proposed in efficient market theory for the markets without any efficiency) to the similar behaviors for the companies (followers firms) in which industry leaders are the criterion of their decisions Page

3 Review of literature Healy and Wahlen (1999) presented the following definition of earnings management. Earnings management" occurs when managers use personal judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of a company or influence contractual outcomes their conclusion has special benefits (Esmaili, 2010). Bior et al defined manipulation of earnings by management to achieve a part of prejudices of expected earnings (e.g. analysts forecast, previous estimations of management or reduction of earnings dispersion) as earnings management. On the other hand, earnings management is considered intentional interference in external financial reporting process with the purpose of achieving benefit (Fama Eugene et al., 2001). Always, the pressure on managers to meet the expectations of investors (market) and their inclination to show high earnings (low earnings) stimulated them for earnings manipulation or earnings management. If the reported earnings from leader companies (progressive) is predicted and such earnings doesn t meet the market expectations, the pressure on following firms regarding the earnings manipulation to meet the market expectations is reduced (Healy and Wahlen, 1999). Bretten et al., (2012) in their study found that there is a positive association between accrual reporting of performance of following firms with the reported performance of leader firms. This relation showed that if leader firm can not meet the market expectations, the following firms have low motivation to manipulate earnings via the changes in accruals. The evidences showed that when a company announces its earnings, the stock price of other active companies in that industry moves in the same direction. The researchers evaluated that whether the stock price of the companies didn t announce their earnings (or announced late) showed suitable response to the earnings announced by other active firms in industry. The results of their study showed that the stock price of the required companies showed low response to the initial earnings announced by other companies. These researchers found that stock prices are mostly affected by earnings announced late by the firms themselves (Thomas and Zhang 2008). Managers with the aim of maximizing the wealth of stockholders should have balance between the different interests of stockholders in order not to lose profitable opportunities of investment and on the other hand meet the stockholders expectations. Thus, policies of divided being applied by companies managers are of great importance (Etemadi et al., 2011). Easton et al., (2011) based on the studies believed that both investors and analysts (market) responded to the late announced earnings prices (JabarzadeKangarloui et al., 2010). Fama et al., (2001) in their study found that managers pay or increase dividends when they feel their earnings is mostly not affected by manipulation of accounting method and reduction of dividend in future is impossible (Jabarzadekangarluyi et al., 2012). Aggrawal and Samwick (1999) achieve good contracts of strategic competition in a framework in which it is possible that managers via their earnings and earnings of Competitor Company in the similar industry compensate their losses (Gale, 1996) Page

4 Lo and MacKinaly (1990) in their study showed that response of big business firms to the common information led to the response of small business companies (Shabahang, 2002). Hirst et al., (1995) in their studies showed that when discretion level is reduced or validity of management report is improved, investors are less inclined to have useful information in evaluation of their existing motivations in management reporting (the probable manipulation of management on the information of financial statements) (Mehrani and Talane, 1998). Healy and Wahlen (1999); Dechow et al., (2010) in their studies showed how managers use their authorities to meet their expectations. They also found managers to report positive News about earnings have definite motivations in relation to market response (Hirst et al, 1995; Mashaykhi and Safari, 2006). However, in Iran, there is no specific study in this regard; some of the related studies are referred later: In their study, they investigated the relationship between earnings management of the consolidated companies with stock returns of the main company listed on TSE. They attempted to find the answer to the question whether the investors, analysts and other users of accounting information perceived earnings management of the combined companies and consider in their decisions? The results of their study showed that there is a significant and inverse relation between earnings management of groups (mentioned in consolidated financial statements) with stock returns of the main company (Bretten et al., 2012). In their study entitled theory of meeting the demands of stockholder and identification of effective factors on dividend policy in TSE investigated the important factors of dividend policy in TSE. The results of their study showed that the managers of active companies in TSE didn t attempt for satisfaction of stockholders and didn t have stable dividend policy. In addition, life cycle of the company and systematic risk were not important factors of dividend policy and some factors as uncertainty of cash flows, fixed assets, cash flow and company size and investment opportunities are the most important effective factors on dividend (Dechow and Schrand, 2010). Research hypotheses As it was said, the following hypotheses are formulated and tested: First hypothesis: There is a positive association between earnings management (emphasis on discretional accruals) in leader and follower firms in various industries. Second hypothesis: There is a positive association between the changes of announced earnings by leader and follower firms in various industries. Third hypothesis: There is a positive association between the changes of dividend to earnings per share in leader and follower firms in various industries. Fourth hypothesis: There is a positive association between the changes of retained earnings in leader and follower firms in various industries. Materials and Methods 2600 Page

5 Based on the study title on one hand, and the necessity of achieving the adequate data on the other hand, sampling criterion among various industries was common regulations based on other studies conducted and some of them are explained later. The regulations and criteria of selecting sample companies based on systematic elimination method - The end of fiscal year of the companies is 29 Esfand of each year. - The sample member companies in the required period had no fiscal period change. - They are not including the active companies in banking, investment fields and credit institutions. - They are in the active industries that in terms of the number of active companies in industry have adequacy to generalize the results. - During the required years, there is no considerable trade stopping. - The required data of sample companies for required time period is available. - Distinguishing leader and follower companies is possible. Based on the above items, seven industries including main metals, car and parts, cement, materials and chemical products, metal inorganic, pharmacology and tile and ceramic were selected and 124 companies were selected as sample and were analyzed during Methods This study is correlation and is conducted based on the information in terms of the type and number is done as data accumulative analysis (accumulative data). In this study, the historical data of companies is used. Thus, the study is ex post facto design. To test the study hypotheses, multi-variate linear regression method is used. To explain and summarize the collected data, descriptive statistics (mean, median and SD) and for analysis of study hypotheses, inference statistics (regression analysis, Durbin-Watson) were applied. Other tests of validation of study models were done based on common statistical models. In this study, to do the calculations of variables, Excel software was used and for hypotheses analysis, Eviwse software was applied. Research models and variables and their calculation method To test the first to fourth hypotheses of the study, the following four models were used to evaluate the significant relationship between the earnings management of follower and leader firms: DA α 0 +α 1 DA l,t +α 2 SIZE +α 3 LEV +α 4 PPE + α 5 B/M + ε 0 (1) % Δ EAT α 0 +α 1 %Δ EAT l,t +α 2 SIZE +α 3 LEV +α 4 PPE + α 5 B/M +ε (2) % Δ DPSR α 0 +α 1 %Δ DPSR l,t +α 2 SIZE +α 3 LEV +α 4 LR +ε (3) %Δ RE α 0 +α 1 %Δ RE l,t +α 2 SIZE +α 3 EBT +α 4 PPE +ε (4) 2601 Page

6 The variables and their calculation method are shown in Table 1,2 and measurement method of some of the indices are shown in Table 3. Table 1: The list of variables their symbol Symbol Dependent variables DA Variables explanation Earnings management of follower firm (f) in year t % Δ EAT The percentage of changes in earnings announced (earnings after tax) by following firm (f) in year t compared to the previous year % Δ DPSR The percentage of the changes of dividend ratio of follower firm (f) to earnings per share of that firm in year t compared to the previous year % Δ RE The percentage of changes of retained earnings of following firm (f) in year t compared to the previous year Independent main variables (non-minor variables) DA l,t Earnings management of leader firm (l) in year t % Δ EAT l,t The percentage of changes in earnings announced (earnings after tax) by leader firm (l) in year t compared to the previous year % Δ DPSR l,t The percentage of the changes of dividend ratio of leader firm (l) to earnings per share of that firm in year t compared to the previous year % Δ RE l,t The percentage of changes of retained earnings of leader firm (l) in year t Independent auxiliary variables SIZE LEV PPE * B/M * EBT LR compared to the previous year Size of follower firm (f) in year t Financial leverage or book value of total liability ratio to total assets of follower firm (f) in year t Net fixed assets of follower firm (f) in year t Book value to market value ratio of follower firm (f) in year t Net earnings before tax deduction follower firm (f) in year t Liquidity assets ratio to total assets of follower firm (f) in year t For homogeneity of two variables EBT and PPE, the following equations are used: *PPE PPE /A -1 *EBT EBT /A -1 A F,t-1 : Assets of first period of follower firm 2602 Page

7 Table 2: The calculation method of variables The model of calculation of main and sub dependent and independent variables TA - NDA DA TA l,t - NDA l,t DA l,t %ΔEAT %ΔEAT EAT _ EAT EAT l,t -1-1 EAT l,t _ EAT EAT DA DA l,t EAT Δ % EAT l,t Δ % %ΔDPSR DPS EPS DPS EPS DPS EPS DPSR Δ % %ΔDPSR l,t DPS EPS l,t l,t DPS EPS DPS EPS DPSR l,t Δ % %ΔRE %ΔRE l,t RE RE RE -1 REl,t RE RE -1 RE Δ % RE l,t Δ % Ln ABV SIZE SIZE LEV LR Total liability Total Assets liquidity Assets Total Assets The calculation of Earnings management LEV LR Jones (1991) by presenting the following model attempted to control the effects of changing economic conditions of firms to non-discretionary accruals. Based on Jones model (1991), nondiscretionary accruals is calculated as followings: NDA it α 1 (1/A it-1 ) + α 2 ( REV it / A it-1 ) + α 3 (PPE it /A it-1 ) 2603 Page

8 NDA it : Non-discretionary accruals of firm I in year t A it-1 : Total assets of firm I in year t-1 REV it : The change in net income of firm it in year t and t-1 PPE it : Property, machineries and equipments (gross) of firm I in year t α1 و وα2 α3 Estimated parameters of the company and are calculated form the following equation (Aggrawal and Samwick, 1999). TA it / A it-1 α 1 (1/A it-1 )+α 2 (ΔREV it /A it-1 )+α 3 (PPE it /A it-1 ) If non-discretionary accruals are deducted from total accruals, discretionary accruals (denoting the earnings management) are obtained. These accruals are calculated by Jones model as: DATA- NDA In Jones model, the total accrual is calculated as: TA it E CFO TA: Total accruals of firm I in year t E: Operating earnings CFO: Cash flow from operations Table 3: The explanation of the applied variables in calculation models of dependent and independent variables In the above models, the indices are defined as followings: EAT Earnings after tax of follower firm (f) in year t EAT -1 Earnings after tax of follower firm (f) in year t-1 EAT l,t Earnings after tax of leader firm (l) in year t EAT Earnings after tax of leader firm (l) in year t-1 CFO CFO l,t Ln(ABV) Total Liability Total Assets liquidity Assets Total Assets Cash flow from operation of follower firm (f) in year t Cash flow from operation of leader firm (l) in year t National logarithm of assets book value of follower firm (ABV) The total liabilities to total assets ratio of follower firm (f) in year t The liquidity assets to total assets of follower firm (f) in year t or liquidity ratio The separation method of follower and leader firms 2604 Page

9 To separate two types of firms, the groups of the firms with good indices to average industry index are defined as a model for other companies. However, average industry index is not defined in Iran. In each of special fields, the companies with suitable indices to average measured index are leader firms and other firms are follower firms. In this study, to separate leader and follower firms, the main difference of leader (progressive) firms from following firms (followers) is announce earnings five days earlier than follower firms (Bretten et al., 2012). In addition to financial ratios (sale returns and assets returns) are used as the index of superiority of leader firms to follower firms. Thus, active companies in each industry first that announce earnings at least five days earlier than earnings announcement of other firms and second they have the best pair rations are considered as leader firms and other existing firms in each industry are selected as following firms. Based on the policy and required model to select two groups of companies, during 7-year study, the collected data showed that in most of the industries in the studied period, some firms had close earnings announcement (in other words the period of earnings announcement was less than 5 days), thus in these industries, some companies were considered as leader or follower firms. In the period, the companies with high frequency of earnings announcement (approved data of common assembly) and superior in their industry and their profitability ratio are considered leader firms and other firms are defined as follower firms. In short, there criteria of early earnings announcement, having high net earnings to assets and more net earnings to sale are the ranking criteria of the companies and after ranking the companies based on the scores, the leader and follower firms are divided from each other. Results Table 4: The descriptive statistics data of follower firms (*) Variable Mean median Max Min SD Skewness kurtosis Earnings management The percent of changes in announced earnings prices The percent of changes in dividend The percent of changes in retained earnings Company size Financial leverage Liquidity ratio Net fixed assets Net earnings before tax deduction Book value to stock market value (*)Total observations 99: (693 follower firms during 7 years) Table 5: The descriptive statistics of leader firms (*) 2605 Page

10 Variable Mean median Max Min SD Skewnes s Earnings management kurtosis The percent of changes in announced earnings prices The percent of changes in dividend The percent of changes in retained earnings (*) Total observations 25: (175 leader firms during 7 years) Descriptive statistics were including the number of observations, mean, min and max, SD, skewness and kurtosis. The results showed the approximation of mean and median of data. The low calculated standard deviation showed low dispersion of the studied variables. The study of the skewness and kurtosis of each of the variables and their comparison with normal distribution showed that all the study variables were normally distributed. Based on the results of Limer test (significance level less than 0.05) for study models showed that the applied structure is panel data. The results of Hausman test (significance level less than 0.05) showed using fixed effects method in four models. DA α 0 +α 1 DA l,t +α 2 SIZE +α 3 LEV +α 4 FA + α 5 B/M +ε (1) Variable Coefficient Standard error T- statistics P-value C Earnings management of leader firm Follower firm size Follower firm financial leverage Net fixed assets of follower firm Book value to market value follower firm AR(2) Coefficient of determination of the model 88% Significance statistics of model (F) 8/05* 2606 Page

11 in P-value (Significance level) (0/00<0/05) Durbin-Watson statistics 2/18 Table 6: The results of first hypothesis test The results of Fisher statistics (F statistics) showed significance of first regression model at 95%. As F statistics was 8.05 and significance level (0.05>0.00) is smaller than 0.05, the coefficient of determination showed that the model can explain 88% of dependent variable changes (earnings management of follower firm). AR(2) showed that regression model is estimated by elimination of autocorrelation with two lags. Based on t-statistics as 3.32 and significance level ( smaller than 0.05), it can be said that statistically, the earnings management of leader firm had significant association with earnings management of follower firms in TSE as for one unit increase of earnings management of leader firms, earnings management of follower firms is increased units. Table 7: The results of second hypothesis test % Δ EAT α 0 +α 1 %Δ EAT l,t +α 2 SIZE +α 3 LEV +α 4 PPE + α 5 B/M +ε (2) Variable Coefficient Standard error T- statistics P-value C The percent of changes in announced earnings of leader firm Follower firm size Follower firm financial leverage Net fixed assets of follower firm Book value to market value of follower firm Coefficient of determination of the model 53% in P-value (0/00<0/05) Durbin-Watson statistics 2/04 Significance statistics of model (F) 3/18 * The results of Fisher statistics (F statistics) showed significance of first regression model at 95%. As F statistics was 3.18 and significance level (0.05>0.00) is smaller than 0.05, the coefficient of determination showed that the model can explain 53% of dependent variable changes (changes in announced earnings of follower firm). Based on t-statistics as and significance level (0.018 smaller than 0.05), it can be said that statistically, the changes in announced earnings of leader firm had positively significant association with the changes in announced earnings of follower firms in TSE, as for one unit increase of the changes in announced earnings by leader firm, the changes in announced earnings by follower firm is increased units Page

12 Table 8: The results of third hypothesis test % Δ DPSR α 0 +α 1 %Δ DPSR l,t +α 2 SIZE +α 3 LEV +α 4 LR +ε (3) Variable Coefficient Standard error T- statistics P-value C The percent of changes in dividend of leader firm Follower firm size Financial leverage of follower firm Liquidity ratio of follower firm AR(1) Coefficient of determination of model 37% Significant statistics of model (F) 2/92 * in P-value (0/00<0/05) Durbin-Watson statistics 2/07 The results of Fisher statistics (F statistics) showed significance of first regression model at 95%. As F statistics was 2.92 and significance level (0.05>0.00) is smaller than 0.05, the coefficient of determination showed that the model can explain 37% of dependent variable changes (changes in dividend of follower firm). AR (1) showed that regression model by elimination of autocorrelation with one lag was estimated. Based on absolute value of t-statistics as and significance level (0.00 smaller than 0.05), it can be said that statistically, the changes in dividend of leader firm had positively significant association with the changes in dividend by follower firms in TSE, as for one unit increase of the changes in dividend of leader firm, the changes in dividend of follower firm is increased units. Table 9: The results of fourth hypothesis test %Δ RE α 0 +α 1 %Δ RE l,t +α 2 SIZE +α 3 EBT +α 4 PPE +ε (4) Variable Coefficient Standard erro T- statistics P-value C The percent of changes in retained earnings of leader firm Follower firm size Page

13 Net earnings before tax of follower firm Net fixed assets of follower firm The coefficient of determination of model 41% in P-v (0/00<0/05) Significant statistics of model (F) 2 Durbin-Watson statistics 2/15 The results of Fisher statistics (F statistics) showed significance of first regression model at 95%. As F statistics was 2.23 and significance level (0.05>0.00) is smaller than 0.05, the coefficient of determination showed that the model can explain 41% of dependent variable changes (percent of changes of retained earnings of follower firm). The results of significance of coefficients at 95% showed that: Based on absolute value of t-statistics as and significance level (0.031 smaller than 0.05), it can be said that statistically, the changes in retained earnings of leader firm had positively significant association with the changes in retained earnings by follower firms in TSE, as for one unit increase of the changes in retained earnings of leader firm, the changes in retained earnings of follower firm is increased units. Conclusion Based on the purpose of the study (the survey of the following of follower firms of leader in earnings management and dividend policies), the status of earnings management and dividend policies in two groups of leader and follower companies was studied. To do this, the behavioral effects reflected in performance reported by leader firms by the managers on behavior of follower firms managers for reporting their performance were analyzed. Based on the hypothesis of the study and theoretical basics in literature of follower firms active in various industries on earnings management and dividend policies follow the active leader firms in the industries. In other words, the performance of follower firms in earnings announcement was affected by the performance reported by leader firms. Bretten et al., (2011) regarding the manipulation of accruals and earnings management found that there is a positive association between discretional reporting of the performance of follower firms with the performance reported by leader firms. When the leader firm can not meet the market expectations, the follower firms have low incentives to manipulate earnings via the changes in accruals. The result of the present study on earnings management showing the positive association between leader and follower firms is in line with the results of the study of Bretten et al., (2011). The recommendations based on the study results Based on the result of the first hypothesis regarding the positive relation between earnings management (manipulation via changing accruals) in leader firms with their values in follower firms, in other words the support of following of follower firms of leader companies in this regard is recommended to the investors and users of financial information and in decision making for investment in follower firms, they consider nature of behavior of companies (mere 2609 Page

14 following in earnings management and dividend from other firms) with quality analysis of information. Based on the results of second to fourth results, the investors can consider announced earnings, dividend per share and retained earnings of the behavior of leader firms (progressive) active in industry as the basis of response forecast of follower firms. Recommendation for future Studies 1- Using other indices to differentiate between leader and follower firms except the earnings announcement and financial ratios applied in the study with emphasis on the hypotheses of the study. 2- The study and evaluation of modeling or following of follower firms of leader (progressive) firms in disclosure and information transparency. Proposed hypothesis: There is a positive association between information transparency of financial statements presented by leader and follower firms in various industries in TSE. 3- With the aim of comparative evaluation of stock price behavior in leader and follower firms, it is recommended to conduct a study with hypotheses in future. The proposed hypothesis: There is a positive association between price fluctuations in leader and follower firms in various industries in TSE. 4- To justify and interpret the distribution of loans given and comparative evaluation of financial leverage in follower and leader firms, the following hypothesis test is recommended in future studies. Proposed hypothesis: There is a positive association between financial leverage in leader and follower firms in various industries in TSE. Reference Aggrawal, R., Samwick, A., (1999). The other side of the trade-off: the impact of risk on executive 2. compensation. The Journal of Political Economy 107 (1): Barth, M., Beaver,W., landsman. W, (2001). the relevance of the value relevance literature for financial accounting standard setting : another view", journal of accounting and economics, 31: Bretten,B., Jeff, L., Payne, Wayne, Thomas, B. (2012). Earnings Management: Do Firms Play, Follow the Leader, Dechow, P., Ge W., Schrand, C., (2010). Understanding earnings quality: A review of the proxies, their determinants, and their consequences. The Journal of Accounting and Economics. 50(3-2): Easton, P., Gao, G., Gao., P. (2011). Pre-earnings announcement drift. Working Paper, University of Notre Dame. Esmaili, Sh Earnings quality. Accountant journal. No.184. Etemadi, H., Momeni, M., Farajzade, H. (2012). How Earnings management affect the earnings quality of firms? Financial accounting studies. 4(2): Fama Eugene, F., Kenneth, R., French. (2001). Disappearing Dividends: Changing Firm Characteristics or Lower Propensity to Pay?. Journal of Financial Economics, 60: Page

15 Gale, D., (1996). What have we learned from social learning? European Economic Review 40 (3-5): Gul, F., Lundholm, R., (1995). Endogenous timing and the clustering of agents decisions. Journal of Political Economy 103 (5): Healy, P. M., Wahlen, J. M., (1999). A review of the earnings management literature and it s for standard setting. Accounting Horizons,13(4): Hirst, D., Koonce, L., Simko, P. (1995). Investor reactions to Wnancial analysts research reports, Journal of Accounting Research, JabarzadeKangarloui, S., Mohammadzade, H., Bayazidi, A., Muneskhah, Gh. (2010). The study of the relationship between earnings management of consolidated firms with stock return of the main company in TSE. Stock exchange journal. 11(3): Jabarzadekangarluyi, S., Motevasel, M. (2012). The theory of meeting the demands of stockholder and identification of effective factors on dividend policy in TSE by Tobit regression. Scientific and research journal of investment knowledge. 1(3): Jones,J. (1991).Earning management during import relief investigation ", Journal of accounting research. Lo A. (1990). MacKinaly are contrarian profits due to stock market over-reaction. Rev Finance stud 3: Mashaykhi, B., Safari, M. (2006). Cash flow from operation and earnings management in the companies listed in TSE. The accounting and auditing investigations, 44: Mehrani, S., Talane, A. (1998). Dividend in companies. Accountant journal. Year 12, No Mikhail M, B. Walter, Willis, R. (2003). Reactions to Dividend Changes Conditional on Earnings Quality. Journal of Accounting, Auditing and Finance,18 (1): Shabahang, R. (2002). Accounting theory (Vol.1). Audit organization publications. Noravesh, A., Nikbakht, M. (2005). The study of earnings management in the companies listed in TSE. Social sciences and human sciences of Shiraz University. 43: Thomas, J., Zhang, F., (2008). Overreaction to intra-industry information transfers? Journal of Accounting Research 46 (4): Page