Relationship between abnormal returns on the dividend policy of listed companies in Tehran Stock Exchange

:504-513 www.amiemt-journal.com Relationship between abnormal returns on the dividend policy of listed companies in Tehran Stock Exchange Ariyan Nasirzadeh, Dr. Hasan Hemati, Mahboobeh Khanahmadi 1,3 Department of Accounting, Science and Research Branch, Islamic Azad University, saveh, Iran 2 Department of Accounting and faculty member at University of Ghazvin univercity Abstract This study evaluated the impact of Dividend Policy on abnormal returns on listed companies at Tehran Stock Exchange. In this study, quality of earnings and abnormal returns in the time domain between the years 2005 to 2010 companies listed in Tehran Stock Exchange will be tested. The present study was quasi-experimental and ex post facto and Based on actual market data and financial statements of listed companies in Tehran Stock Exchange was and using information from the financial statements of 132 listed companies in Tehran Stock Exchange and using multiple linear regression analysis, using the software Spss16 results The results of this study indicate that the results of the test show that the first sub-hypothesis. The size of the abnormal returns between Dividend Policy is impressive. The first sub-hypothesis testing shows that the growth rate of firms with abnormal returns between Dividend Policy influences. The main hypothesis of this study indicate that in the first The first sub-hypothesis test, the earnings per share and return on equity for small firms than larger firms there is no significant relationship, Consequently, the hypothesis was not confirmed. The second subhypothesis of abnormal returns between dividends and stock for small companies than larger companies, there is no significant relationship, and the hypothesis is confirmed. The third sub-hypothesis of abnormal earnings per share and return on equity for companies with high growth compared to firms with low growth, there is a significant correlation between the results confirmed the hypothesis. The fourth sub-hypothesis of non-ordinary dividends and stock returns for firms with high growth compared to companies with lower growth as a result of the hypothesis is a meaningful relationship has been verified. Key words: Dividend Policy, Abnormal return, Size, Growth 1. Introduction Today, due to the expansion of economic activities, development of financial markets and stimulate investment in capital markets, particularly the stock exchange by natural and legal persons, the most important tool for making decisions and the expected gain and optimum use of resources access to accurate and timely information and analysis that is accurate and realistic Including the information that the users of financial information in their decisions are considered, relevant information about the company 's return on equity. The company and any other information somehow related to the company's performance and financial condition may be associated with the impact on returns. If accounting information is useful for determining the efficiency of the change in accounting data must be due to changes in stock returns. Therefore, shareholders and investors, the capital market expects that the accuracy of the information received by the firms listed on the stock exchange confirmed the to achieve the benefits of buying and selling stocks are kept. Shareholders and investors to evaluate companies using a variety of factors, among which are, earnings per share, EPS and DPS is the dividend that the changes can provide information on the status and performance of the present and future, Dividends and earnings per share to be used as agents for the control and management performance evaluation by their shareholders. The relationship between dividends and earnings per share reflects the company's dividend policy. Dividend policy could attract new shareholders to the company. 504

Dividend decision is one of the most important decisions of the company. One way is to earn dividend yields. Dividend policy can be stacked to create a balance between corporate profits and cash from one side and the other side of the new equity defined (Khodadadi, 2008). The data on stock prices and consequently have an effect on stock returns and you can contain is information on consumer behavior, particularly the impact of actual and potential shareholders increase or decrease the price and the efficiency of Abnormal (efficiency more than the Market Efficiency) cause. In this study, the introduction of the abnormal stock returns and dividend policy given the importance and necessity of research and then the main problem is the discussed. The explanation about the aims and hypotheses of research and application results are presented and finally, the season will end with a definition of key terms. This chapter will explain the general framework of the research question. 2. Statement of Problem Information needs of external users of financial reporting purposes stems. These financial statements are the primary means of transmitting information to the parties that the final product will be an accounting and financial reporting process. Given these benefits, including a leading indicator of economic activity is measured. Therefore, the accounting system should provide the best indicators to provide a complete and efficient information to be profitable (Pourheydari, 2007). Complete view of the market, one of the most important policies, especially policies and transparency in corporate dividend policy is that market analysts consider it one of the important information. Cash dividends per share by being a tangible object for holding one of the sources of liquidity, it is of utmost importance. The main goal of any business is profit and maximizes shareholder wealth. The purpose of this policy is to determine the role that it has policies to maximize shareholder wealth ( Karami, 2008). In studies conducted at institutions dividend policy, dividend policy that is handled by both institutions to various factors such as regulatory requirements, cost considerations, inflation, the growth rate of the company s ability to finance from the outside, the interests of management (board of directors) and... Depends on factors that money managers have a long-term dividend policy in your organization, consider them (The same source ). Investors typically expect to receive dividends from the Company's knowledge, but certain factors such as liquidity situation, the development of company policies and... The Company s ability to declare and pay dividends may be limited to (Cindy, 2012). Thread dividend ( cash ) and the policies that led to the decision-making process is a very important issue annual, especially for the listed company that is highly regarded as shareholders, annual general meetings, so that the main thread (the worm, 2008). Dividend policy can indicate whether or not the true and fair financial performance results. In fact, when managers tend to pay a benefit for loss of business profits, not to ensure future dividends and profits decline is unlikely. Profits and dividends, not only signaled to pay benefits to the future, the sign earning a degree is something of lesser quality. Than other regular payment of dividend increases, investors will realize the quality of reported earnings (Alavi Tabari, 2008). But the main problem, according to the sensitivity of the interactions between dividend policy to maximize shareholder wealth and investment opportunities And considering that the director tends to report higher earnings quality is Roots causes influencing dividend policy on stock return of listed companies in Tehran Stock Exchange is The main question addressed in this study are as follows: 1) The dividend policy has a significant impact on stock return? 3. The importance and necessity of research In the world of investments the most important part of the decision making process, during which the investment is to maximize the interests of investors and their wealth, they need to make the most efficient decisions. In this regard, the most important decision-making process is information. Process of decisionmaking information can be effective. Because different people will make different decisions. The exchange of information is also influenced investment decisions. Theorists securities markets, financial reporting, investors believe that the most important source of information. So one of the purposes of accounting and preparation of financial statements is to provide information to facilitate decision making. 505

Most investors are looking to maximize the return on your investment. For this reason, the diagnosis of the stock when purchased, to be sold when stored and when they are able to meet the goal of maximizing the benefits of their investments will (khademmoaakher, 2000). The issue of the relationship between abnormal returns on the dividend policy of the company were up by scientific studies and experiment altestson the data in Tehran Stock Exchange occurred over the years on the decision-making to buy and sell effective solutions shares may be offered to investors in the Tehran Stock Exchange The participants in the study to help decide how and financial managers to forecast earnings In fiscal year comparison of actual and anticipated profits realized by the end of fiscal year Finally the modifications necessary. To answer the research question, the following hypotheses were formulated: The first main hypothesis: relationship between dividend policy and stock return there. The first sub- hypothesis: the earnings per share and return on equity for small firms than larger companies, there is a significant relationship. The second sub-hypothesis: between dividends and stock abnormal returns for small firms compared to larger companies, there is a significant relationship. The third sub-hypothesis: the earnings per share and return on equity for companies with high growth versus low -growth firms, there is a significant relationship. The fourth sub-hypothesis: between dividends and stock return for firms with high growth versus low -growth firms, there is a significant relationship. Research area Provides a framework to investigate the scope of research and tests performed during a specific territory and has validated more. Data analysis included descriptive analysis and correlation tests of this study. Descriptive analysis of central and dispersion parameters estimation, basic information about the data and variables, is obtained. Correlation tests specifically related to regression models estimate the relationship between the type and intensity are investigated. All statistical analyzes using statistical software Spss 7 and Excel 2010 is done. 4. The scope of research The study period due to availability of data on the study variables and constraints of cost and time, 6 years old and the beginning of 2005 until the end of 2010. 4.1. Variables In this study, the dependent and independent variables were grouped into two categories, and the relationship between them is examined through regression testing. In the present study, the first hypothesis of abnormal stock returns as the dependent variable and the independent variable is the dividend. The second hypothesis of abnormal stock returns as the dependent variable and the independent variables are the quality of earnings. 4.2. Way to estimate the dividend policy Percent dividend payout ratio (dividends paid) indicates the amount of dividend per share. EPS: The number of shares is calculated by dividing net income (Parsaeian, 2005). Dps: the dividend. 506

DPS / EPS Earnings per share / dividend per share = Dividend Measure of abnormal stock returns Return is the difference between the market rate of return for the market-adjusted abnormal returns model (simple model of the market) are used (Izadinia, 2002). ABR it = Ri t Rm t R it at time tthe annual rate of return on stock i R mt :time t Annual market rate at time t Annual stock returns: Stock returns for firms: Pit: price of stock i at the time t, Pio: price of stock i at the initial time t, Dit: dividends paid by firm i at time t 4.3. The annual rate of return on the market: In this paper, based on the total stock market returns will be calculated. I mt : stock index at time t, Imo: stock index since the beginning of t Territory subject This study evaluated the relationship between dividend policy on stock return of listed companies in Tehran Stock Exchange will be reviewed. This research includes studies of companies in Tehran Stock Exchange during 2005-2010 of 132 companies. 4.4. The first sub-hypothesis testing methods: The first derivative of a regression model to test the hypothesis that the abnormal stock return returns as the dependent variable on equity and earnings per share as the dependent variable and a function of firm size as a confounding variable is used. On α1 relationship between earnings per share and return on equity shows the effect of firm size on the relationship between α3 earnings per share and return on equity unusual show. ABR it= α 0 +α 1 EPS i,t +α 2 DSIZE i,t +α 3 EPS i,t* DSIZE i,t + ε i,t 4.5. The second sub-hypothesis testing methods The second derivative of a regression model to test the hypothesis that abnormal stock returns as the dependent variable and firm size as a function of the dividend is a confounding variable is used. On α1 relationship between dividends and stock return and α3 shows the effect of firm size on the relationship between dividends and stock return shows. ABR it= α 0 +α 1 DPS i,t +α 2 DSIZE i,t +α 3 DPS i,t* DSIZE i,t + ε i,t 4.6. The third sub-hypothesis testing methods The third side of a regression model to test the hypothesis that abnormal stock returns as the dependent variable and a function of earnings per share growth of the company as a confounding variable is used. On α1 507

relationship between earnings per share and return on equity shows and α3 impact on the relationship between firm growth, earnings per share and return on equity unusual show. ABR it= α 0 +α 1 EPS i,t +α 2 DGROWP i,t +α 3 EPS i,t* DGROWP i,t + ε i,t 4.7. The fourth sub-hypothesis testing methods The fourth sub-hypothesis of a regression model in which the stock return as the dependent variable and the dividend and the company's growth as a function of the intervening variables are being used. On α1 relationship between dividends and stock return shows the company's growth and α3 impact on the relationship between dividends and stock return shows: ABR it= α 0 +α 1 DPS i,t +α 2 DGROWP i,t +α 3 DGROWP i,t* DPS i,t + ε i,t 4.8. Normality of the variable (s ) associated Research hypotheses in a regression model, the dependent variable is abnormal returns variable. Model test research hypotheses based on regression equations and basic assumptions of normality of the dependent variable regression which normally leads to regression residuals In this research the data normality test of Kolmogorov - Smirnov test was used. The statistical hypothesis test is as follows: H0: the data distribution is normal. H1: the data distribution is not normal. The results of this test are shown in Table 1. Table 1: Testing normality of the dependent variable Kolmogorov-Smirnov test. dependent variable TestStatistics df pvalue Abnormal returns 0.037 792 0.145 According to the results, the significance level of the test, Kolmogorov - Smirnov test for the dependent variable, rather than the experimental error (0/05 = α) and the hypothesis H 0is accepted. In other words, the data on the dependent variable follows a normal distribution and the distribution is close to normal as one of the basic assumptions of regression on this variable will be accepted 5.1. The results of the first test, the main theory One of the main research hypothesis predicted that the relationship between dividend policy and stock return there. This hypothesis has four sub-hypothesis is that the relationship with a particular emphasis on certain features have favored companies. To test this hypothesis, the regression models presented by hope and Joshua (2006) in which the function of abnormal returns from dividends, earnings per share, the special features of corporations and researching control variable is used. 5.2. The results of the first sub-hypotheses The results of the regression model in Table 3-4 is 1-1sub-hypothesis testing. Model to test this hypothesis, the research hope and Joshua (2006) has been adopted and is as follows. ABR it= α 0 + α 1 EPS i,t + α 2 DSIZE i,t + α 3 EPS*DSIZE i,t+ α 4 BM I,t+ ε i,t 508

ABR: abnormal returns (total returns between stock market returns of capital) EPS: Earnings per share real DSIZE: a dual-mode variable is that the sample size is smaller than the median size of all firms, the value is 1, and zero otherwise (as part of the natural logarithm of assets obtained by(bm: ratio of book value of equity to market value of equity as a control variable. Table 2: Results of statistical analysis to test the hypothesis that sub1/1 R 2 Adjusted D-W statistic F Significant F 0.772 2.205 657.697 variable EPS Dsize EPS * Dsize BM The size of the coefficient β (standardized) 0.116-0.052-0.006 0.78 statistic t 5.079-2.497-0.247 5.717 (P-value) 0.013 0.805 Coefficient regression model 0.772 is stating that this model has been 77.2 %of the sample firms through changes in abnormal returns are explained by the independent variables and controls. Also, the results show that Watson statistic camera between 1.5 to 2.5 and hence, there is a strong correlation between the errors of the regression model and its lack of correlation between the errors, as one of the basic assumptions of the regression model, will be accepted. Regression, analysis of variance (ANOVA), which is based on the F statistic that is a decision for the fitted model in the last column of Table in the first sub-hypothesis testing, has come. Statistical hypotheses about F-statistic analysis is as follows. H0: βi = 0 regression model is not significant H1: βi 1 regression model is significant. F-statistic for the model error level of less than significant level exam (0.05 = α), and hence the assumption of H0 is rejected and the estimated regression cally significant relationships between variables, linear regression is. 1 The estimated coefficient for the variable in the relationship between earnings per share EPS represents the abnormal returns, sort of 0.116, with a significance level of, which is less than 0.05 ( test error ) is. These findings suggest a direct relationship between the variables is significant. Dsize variable is negative and significant coefficient obtained indicates that the abnormal returns of firms with smaller abnormal returns than other firms are.also, the coefficient for the variable is not significant in terms of EPS * Dsize. This finding suggests that the small -sized companies, the earnings per share, there is no significant relationship with abnormal returns.differential BM (as a control variable) is positive and significant, indicating that the ratio of book value to market value of firms' stock returns, there is a direct relationship.the results showed that the small-sized companies, the earnings per share, there is no significant relationship with abnormal returns This finding is consistent with claims in the first sub-hypothesis and the hypothesis is rejected at the 95% confidence level. 5.3. The results of the second sub-hypotheses The results of the regression model in Table are 1/2sub-hypothesis testing. Model to test this hypothesis is as follows. 1 All statistical tests were performed at % 95confidence level. 509

ABR it= α 0 +α 1 EPS i,t +α 2 DSIZE i,t +α 3 EPS*DSIZE i,t+ α 4 BM I,t+ ε i,t DPS: dividend per share Table 3: Results of statistical analysis to test the hypothesis that sub1/2 Significant F statistic F D-W R 2 Adjusted 632.645 2.188 0.765 ( P-value) statistic t The size of the coefficient β (standardized) variable 0.002 3.117 0.086 DPS 0.049-1.973-0.051 Dsize 0.692-0.396-0.146 DPS * Dsize 5.086 0.886 BM Coefficient regression model 0.765 is stating that this model has been 76.5 % of the sample firms through changes in stock return explained by the independent variables and controls.also, the results show that Watson statistic camera between 1.5 to 2.5 and hence, there is a strong correlation between the errors of the regression model and its lack of correlation between the errors, as one of the basic assumptions of the regression model, will be accepted. F -statistic for the model error level of less than significant level exam (0.05 = α) is thus estimated to be statistically significant regression relationships between variables, is linear. DPS estimated coefficient for the variable relationship between dividends paid per share represents the abnormal returns, sort of 0.086, with a significance level of 0.002 is. These findings suggest a direct relationship between the variables is significant.dsize variable is negative and significant coefficient obtained indicates that the abnormal returns of firms with smaller abnormal returns than other firms are.also, the coefficient for the variable is not significant in terms of DPS * Dsize. This finding suggests that the small - sized companies, the dividend per share is not significantly associated with abnormal returns The results showed that the small-sized companies, the dividend per share is not significantly associated with abnormal returns. These findings are not consistent with the claims raised in the second sub-hypothesis and the hypothesis is rejected at the 95 %confidence level. 5.4. The results of the three sub- hypotheses The results of the regression model in Table are 1/4 sub-hypothesis testing. Model to test this hypothesis is as follows. ABR it= α 0 +α 1 DPS i,t +α 2 DSIZE i,t +α 3 DPS*DSIZE i,t+ α 4 BM I,t+ ε i,t DGROW: a dummy variable that the company's growth is above the median of all sample firms, the value is 1, and zero otherwise. 510

Table 4- Results of statistical analysis to test the hypothesis that sub 3/1 R 2 Adjusted 0.771 variable EPS DGROW EPS * DGROW BM D-W 2.2 The size of the coefficient β (standardized) 0.108 0.032 0.013 0.86 statistic F 952.984 statistic t 5.953 1.332 2.52 4.061 Significant F (P-value) 0.183 0.013 Coefficient regression model 0.771 is stating that this model is able to 77.1 of the sample firms through changes in stock return explained by the independent variables and controls.also, the results show that Watson statistic camera between 1.5 to 2.5 and hence, there is a strong correlation between the errors of the regression model and its lack of correlation between the errors, as one of the basic assumptions of the regression model, will be accepted. F -statistic for the model error level of less than significant level exam (0.05 = α) is thus estimated to be statistically significant regression relationships between variables, is linear. The estimated coefficient for the variable in the relationship between earnings per share EPS represents the abnormal returns, sort of 0.183, with a significance level of is. These findings suggest a direct relationship between the variables is significant. The coefficient for the variable EPS * DGROW sort of 0.013, with a significance level of 0.013 are the findings show that firms with high growth direct relationship between abnormal returns and earnings per share are significant. This finding is consistent with the claims raised in the third sub-hypothesis and the hypothesis is rejected at the 95% confidence level. 5.5. The results of the fourth sub- hypothesis testing Results of fitted regression models 1-4 in Table 4-6 are sub-hypothesis testing. Model to test this hypothesis is as follows. ABR it= α 0 + α 1 DPS i,t + α 2 DGROWP i,t + α 3 DPS*DGROW i,t + α 4 BM i,t + ε i,t Table 5- Results of statistical analysis to test the hypothesis that sub 1/4 R 2 Adjusted D-W statistic F Significant F 0.764 variable DPS DGROW DPS * DGROW BM 2.184 The size of the coefficient β (standardized) 0.082 0.055 0.017 0.873 627.325 statistic t 3.671 2.062 2.623 4.085 (P-value) 0.04 0.033 Coefficient regression model 0.764 is stating that this model has been 76.4 Percentage change in abnormal returns of sample firms is explained by the independent variables and controls.also, the results show that Watson statistic camera between 1.5 to 2.5 and hence, there is a strong correlation between the errors of the regression model and its lack of correlation between the errors, as one of the basic assumptions of the regression model, will be accepted. F -statistic for the model error level of less than significant level exam (0.05 = α) is thus estimated to be statistically significant regression relationships between variables, is linear. The estimated coefficient for the variable in the relationship between earnings per share and return on common equity DPS shows, sort of 0.082, with a significance level of is. These findings suggest a direct 511

relationship between the variables is significant. The coefficient for the variable is instrumental DPS * DGROW 0.017 and SGR between 0.033 is the finding that high-growth companies, the dividend per share and that there was a direct relationship with abnormal returns. This finding is consistent with the claims of the fourth sub-hypothesis and the hypothesis is rejected at the 95 %confidence level.this finding suggests that investors and capital market participants' decisions to buy and sell shares of companies are paying attention to the special features and these features into their decision models. 6.1. Summarize and interpret the results of the first main hypothesis Results of testing hypotheses about the sub- prime hypothesis, showing that the variable dividend and earnings per share, there is a significant correlation with abnormal returns. Given the evidence that the claims raised in the first main hypothesis is that the relationship between dividend policy and stock abnormal returns are accepted. Also, the findings suggest that firm-specific characteristics on the relationship between dividends and earnings per share and return on common stocks are effective. 6.2. Summarize and interpret the results of the test sub- hypothesis1.1 The findings of the first sub-hypothesis test showed that the relationship between firm size effect on earnings per share and return on equity is not unusual. Results show that the earnings per share and return on common equity, direct connection is established, however, that such a relationship in small firms, the difference is not significant. The findings of the research and the theoretical claims and also in theory, research Manus and colleagues (2012) is inconsistent. The authors find evidence that company size plays an important role in deciding the capital market is presented. Appears to be inconsistent with empirical research results can be due to the small size of the firms in the capital market, he said. Seems that smaller sized firms, abnormal returns are less likely to have a stock market capitalization of the companies in the group is faced with a less popular makes,these firms have abnormal returns or abnormal returns are limited. 6.3. summarize and interpret the results of the test sub- hypothesis 1-2 Findings from the second sub-hypothesis test showed that the effect of firm size on the relationship between dividends paid per share and return on equity is not unusual. According to the findings, the dividend per share and return on common equity, direct connection is established, however, that such a relationship in small firms, the difference is not significant. The findings of the research and the theoretical claims and also the hypothesis, the results Manus and colleagues (2012) is inconsistent. The second sub-hypothesis to interpret the findings and reasons for its disagreement with the empirical evidence discussed in the first sub-hypothesis is equivalent. In this regard, it appears that small firms, because they are growing and need cash resources are rarely available to make dividend distributions and hence, a significant response was seen in the company's dividend. 6.4. Summarize and interpret the results of the test sub- hypothesis 1.3. The third hypothesis, results showed that the relationship between firm growth, earnings per share and return on common equity has an effect. These findings suggest that firms with high growth are a direct correlation between earnings per share and return on common equity is established. This finding and the results of theoretical research Manus and colleagues (2012) is consistent. The researchers, corporate growth factor for the improvement in the capital markets, and they showed that these factors have the potential investors. It seems that companies with high growth in capital markets are more popular. It seems that it is likely that many of the indicators Performance and financial condition of the company, which is highly regarded investors; Compared to other companies, is up and led to higher demand for the company's shares. 6.5. Summarize and interpret the results of the test sub-hypothesis 1/4 512

Results indicate that the relationship between the dividends per share growth rate of firms with abnormal returns is impressive. This finding and the results of theoretical research Manus and colleagues (2012) is consistent. In this context, theoretical evidence suggests that growing firms, managers are trying to re-work the resources to invest in the company and do not want to distribute the benefits among stakeholders. It seems that while the company has grown, rather than managers to create new investment projects, Try to gain the support of the capital market and therefore, are more likely to distribute cash dividends. References Amidu, Mohammed, and Joshua Abor(2006). Determinants of Dividend PayoutRatios in Ghana, The Journal of RiskFinance, Vol. 7 No. 2 pp. 136-145. Baba Naohiko, 2008, "Increased presence of foreign investors and dividend policy of Japnese".www.elserier.com Beaver W. and Cornell B. and Landsman W. and Stubben R. S. (2008) The Impact of Analysts: Forecast Errors and Forecast Revisions on Stock Prices, Journal of Business Finance & Accounting, 35(5), pp.709 740. Belk.,1976, "Distribution of Incomes of Corporations Among DividendsRetained Earnings and Taxes, " American Economic Review, Vol.46, pp. 97-113. belkaoui and picur,2004, The smoothing income numbers some Emprical evidence on systematic difference between core and periphery industrial sectors journal of business finance&accounting. Cheng, T.Y., & Firth, M. (2000). An Empirical Analysis of the Bias and Rationality of Profit Forecasts Publised in New Issue Prospectuses. Journal of Business Finance & Accounting. 27 (3) & (4) April/May 2000. pp : 423 446. Fama, E., French, K. (2008). Dissecting anomalies. Journal of Finance, 63, 1653-1678. Fama, E.F and French, K.R,(1992),"Cross-sectional of expected stock returns",journal of Finance; 36 : 427-465. Francis, J.R. and X. Martin, (2010). "Acquisition profitability and timely loss recognition". Journal of Accounting and Economics 49, 161-178. Jensen,Michael,C. and William Meckling,1976. "Theory of the Firm:ManagerialBehavior.AgencyCocts, and Ownership Structure".Journal of Financial Economics 3,pp.305-360 Kothari, S., Leone, A., Wasley, C., (2005). "Performance matched discretionary accrual measures". Journal of Accounting and Economics 39, 163-197. Li, X., (2009). "Accounting Conservatism and the Cost of Capital" : International Analysis. Working paper, London Business School. 513