GROWTH COMPANY AND DIVIDEND POLICY: EMPIRICAL STUDY ON STATE OWNED ENTERPRISES by: 1. AA Gunawan *) 2. Sulaeman Rahman Nidar **)

GROWTH COMPANY AND DIVIDEND POLICY: EMPIRICAL STUDY ON STATE OWNED ENTERPRISES by: 1. AA Gunawan *) 2. Sulaeman Rahman Nidar **) ABSTRACT This study reviews the dividend policy on SOEs in Indonesia based on transaction cost theory, agency costs theory, the trade-off theory and pecking order theory. This study is an explanatory study to determine the effect of independent variables on the dependent variable. As the dependent variable is the dividend payout ratio. Meanwhile, the independent variable is the variable that is measured by the growth of the company's capital expenditure ratio proxy, state ownership, firm size, profitability, cash flow, and the ratio of dividends last year as a control variable. The study uses panel data with a sample of 46 state-owned companies in the form of a limited liability company engaged non-financial sector with the financial period 2005-2009. The sample selection was purposive sampling that samples deposited SOE dividends during the study period. Hypothesis testing using a fixed-effect regression analysis models. As for overcoming heterokedastisitas and autocorrelation using the method of generalized least squares (GLS). The results found that the company's growth variables and firm characteristics variables simultaneously significant effect on dividend policy. To model the dividend policy, partial, variable capital expenditures, capital structure, firm size, and cash-flow negative and significant alias dividend payout ratio, while profitability and state ownership variables having an positive and significant dividend payout ratio. The study also found that non-listed state-owned companies have an average dividend payout ratio lower than the listed SOEs. Keyword: capital spending, sales growth, firm characteristics, fixed effects models, generalized least squares, recursive models, dividend policy, SOE 1. Background Research In this reserach consider that there are some interesting phenomena in the management of stateowned enterprises in Indonesia are as follows, the state has a significant role in the growth of the national economy, where the state-owned companies are generally engaged in strategic industry sectors and acquire a monopoly of authority, such as the cement industry, telecommunications, mining, transportation and construction. Based on data from the Ministry of State Enterprises, in 2010 the total book value (book value) of assets owned by all SOEs approximately 2,500 trillion rupiahs, or about 39% of the total Indonesian GDP in 2010 amounted to 6,422 trillion rupiahs. While the capital expenditure (CAPEX) SOEs reached 184 trillion rupiahs, larger than the central government capital expenditure in the state budget amounted Year Budget 2010, 80.28 trillion rupiahs. However, investment or capital expenditure (CAPEX) this state is still relatively small compared to the operational expenditure (OPEX). Comparison between expenditures for operational expenditure (OPEX) and capital expenditure outlay for (CAPEX) SOEs from 2005 to 2009, there were expenditures for capital expenditure (CAPEX) for five years of state-owned companies is much smaller than the expenditure to operational expenditure (OPEX). 1

This means that the activities of state-owned companies are still focused on routine operations, which does not encourage growth. This can happen due to several causes, among others, investment opportunities SOEs still little. But as a monopolistic company, the investment opportunities in the state should be much more than a non-soes. Other causes include funds to finance investment activities not enough. In addition there is the possibility of SOE managers are reluctant to invest due to the constraints of bureaucracy or intervention in the activities of these investments. Government revenue from SOE dividend payments this number continues to increase. In 2005, the government targets a dividend payment of 8.9 trillion rupiahs, while the realization reached 12.84 trillion rupiahs. In 2009 the target dividend payment amounted to 28.61 trillion rupiahs, while the realization was 26.05 trillion rupiahs. SOE dividend ratio of the net profit is also likely to increase. In 2005 the dividend payout ratio of 30.26% of his, in 2006 increased to 39%, in 2007 slightly decreased to 37%, in 2008 remained 37%, in 2009 dropped to 30%. On the average dividend payout ratio over the entire SOE five years is 34.65%. This ratio is much higher when compared with the average dividend payout ratio (DPAY) non-soe firms belonging to the LQ-45 at the Indonesia Stock Exchange in the amount of 24.65%. Research on the relationship between the growth of the company as a proxy for the investment policy and dividend policy financing, dividend theory approach and the theory of capital structure. Study on dividend initiated by Miller and Modigliani's research (1961) which concluded that the dividend payment does not affect the value of the company. Assuming the presence of perfect market conditions (perfect capital markets), dam Modigliani Miller (MM) argued that firm value is only determined by the company's ability to generate earnings, not on how to separate the company's earnings and dividends to retained earnings. So the dividend payment now or later is irrelevant because it produces the same value of the company. Opinions MM dividend irrelevance theory known. Another suggestion to the contrary that the dividend will increase the value of the company. This opinion was expressed by Gordon (1959.1963). Gordon stated that in a situation of uncertainty, dividend payments are preferred by investors compared with capital gains, as dividends are considered more uncertain. Investors assume that dividends received today have less risk than capital gains earned in the future. So that the current dividend payout is believed to reduce investor uncertainty. Conversely, if the dividend is reduced or not paid, the level of uncertainty for investors will increase and lead to an increase in the desired returns (required rate of return on equity). Opinions Gordon is known for its bird-in-the hand theory. However, according to Modigliani & Miller's opinion is wrong with the argument that the risk of companies assessed by the cash-flow operations, instead of how the company divides its profits. Modigliani & Miller judge the reason-the theory of bird in the hand is a fallacy. According to Easterbrook (1984), dividends may reduce agency problems, due to the payment of dividends, internal funds (free-cash flow) for the financing of investments to be reduced which will encourage companies to seek financing from outside (external financing). This in turn raises monitoring by capital market activities (capital markets) or the creditors of the company that will reduce agency costs (agency cost) from the free-cash flow. While Jensen (1986) argues that the debt financing (debt) can be used as a substitute dividends to reduce the agency cost. Other theories are transaction cost theory of dividends. This theory is based on the argument that the payment of dividends, the company will seek funding from outside (external financing) to finance its investment needs, which will lead to transaction costs (transaction costs) for the 2

external financing (Battarchaya, 1979). The implication according to transaction cost theory that the company should maintain the level of retained earnings (retained earnings) to finance investment needs. The new dividend payment can be made if the company has sufficient retained earnings to finance investment. Several studies have been conducted to examine the effect of firm growth (growth) on dividend policy, among others, by Rozef (1982), Smith and Watts (1992), Graver & Graver (1993), La porta et.al (1999), by using the theory of cost transaction (transaction cost theory) and the theory of agency costs (agency cost theory). Rozeff find the model minimizes the cost of the dividend is a dividend payment models that minimize the transaction costs of external financing (transaction costs of external financing) and agency costs (agency cost). In Rozeff study, it was found that the growth of the company as a proxy of transaction costs, negatively affect dividend policy. Smith & Watts and Graver & Graver test the effect of investment opportunities (IOS) on the financing policy and dividend policy. Smith & Watts concluded that the larger companies have growth opportunities (growth opportunity), will have a level of leverage, cash flow, and lower dividend ratio. Research Gaver & Gaver, among others, concluded that debt financing has a negative and significant relationship with growth opportunity. Another discovery is a company that is growing (growth firms) have a ratio of debt / equity is lower than non-growth firms. For dividend policy, research results concluded that the company's growth in the group of firms paying lower dividends than companies in the non-growth firm. In addition it was found that the dividend yield and significantly negatively associated with growth opportunity. Several previous studies also found several factors which affect the characteristics of the company dividend policy, among others, ownership structure, leverage ratio, firm size, the level of earnings, cash flow, and dividend ratio last year, Shleifer and Vishny Research (1986.1987) in Wang et.al (2011) concluded that institutional owners (institutional owners) is an effective means of control of company policy including dividend policy. This influence will increase if the state is a majority owner. Further research findings Gul (1999), Gugler (2003), Wang et.al (2011) found that state ownership has a positive effect on dividend policy. Research Rozeff (1982), Crutchley and Hansen (1989), Mollah (2001) found that companies that have a high leverage ratio turned out to pay lower dividends. Holder et al. (1998), Al-Malkawi (2007) and Al-Kuwari (2009) found a positive relationship between the size of the company dividend policy. While research Lintner (1956), Jensen et al. (1992), and Fama and French (2000) found that corporate earnings have a positive and significant relationship with the dividend policy. Lintner also find that dividend payments last year (past dividend) positive effect on the payment of dividends in. Then Holder (1988), Amidu and Eco Chic (2006), and Anil and Kapoor (2008) found a positive effect of cash flow on dividend policy. Based on the background research and the identification of the above problems, the research problem can be formulated as follows: 1) How does the company's growth (growth of firm), capital structure (capital structure), state ownership (state ownership), firm size (firm size), profitability (profitability), cash flow (cash flow), and dividends last year (past dividend) on dividend policy (dividend policy), either simultaneously or partially. 2) How does the difference between the dividend policy of non-listed SOEs with state-owned listed. 3

2. Framework Most of the state-owned enterprises in Indonesia still a private company. Of the 128 state-owned SOEs shaped, only 18 companies that have public company (listed SOEs). The government has adopted a policy to privatize state-owned enterprises by encouraging already eligible to sell their shares to the public. Privatization aims performance increase of corporate value, increase benefits for public and countries, as well as broaden share ownership by the community. Several studies on privatization, among others, Megginson & Netizens (2001) and Sun & Wilson Tong (2002) concluded that the performance of the company after privatization showed better performance than the performance before privatization, among others, measured by increased profitability and capital expenditure (capex) and declining debt ratio (leverage), while the dividend payout ratio is likely to increase. The growth of the company is increasing the size of the asset or company. Growing companies, will be reflected in the growth rate of the company's sales or revenue increases. Growing companies will also need a source of financing for investment costs. The financing, could come from internal funds (retained earnings), using debt (debt) or by using external equity. According to the theory of transaction costs (transaction cost theory), the company that is growing, is likely to hold its earnings to finance investment. If the dividend is paid, then the funds will be reduced so take a concerted internal financing from outside (external financing), transaction costs, so the dividend payment made where sufficient internal funds to finance its investment. This research that supports this theory is the study Rozeff (1982), Lloyd et al, (1985), Moh'd, et al (1995), and Holder (1998). According to the agency cost theory of Jensen & Meckling (1976), the separation of ownership and managerial agency will cause problems, because their interests are not always the same between. So the ownership structure (ownership structure) will affect the company's financial policies, including corporate dividend policy and capital structure. Furthermore, according to Jensen & Meckling, managerial ownership ratio (managerial ownership) that would lower the greater the agency problems because managers also became the owner of the company, because the interests of the company will be in accordance with the interests of the manager as the owner of the company. In addition to managerial ownership, institutional ownership (institutional ownership) also affects the agency problem. Institutional ownership in a stock typically has a large number of (large-block shareholder) so as to have the effective ability to monitor the behavior of managers in managing the company, Shleifer & Vishny (1986) in Wang et.al (2011). The influence of large-block shareholders in the management of the company even greater if large-block shareholders are government institutions or state. One of the company's decision-making can be controlled is dividend. Shleifer & Vishny (1987) in Wang et.al (2011). Several previous studies have tested the effect of state ownership (state ownership) to the dividend policy. Gugler (2003) examined the relationship between dividend and ownership (ownership structure), with a sample of 214 companies in Australia in the period 1991 to 1999, using the method of ordinary least squares (OLS). Gugler study concluded that the state-owned companies tend to be reluctant to reduce dividends than the family-owned company. Several other studies, among others, Gul (1999), Al-Malkawi (2007), Al-Kuwari (2009), Wang et.al (2011) also reached the same conclusion that the state ownership (state ownership) has a positive effect on dividend policy. 4

To mitigate agency conflicts within a company, government / state as an institutional owner with majority ownership, may influence the decisions of managers in the management of the company, among others, the decision to pay dividends, as well as the decision to borrow from outside (external financing). In addition to investment finance company, goal that outsiders (lenders / capital markets) can monitor the managers that they may take measures in accordance with the interests of the company (Easterbrook, 1984). Also according to Jensen (1986), debt financing (debt) is the substitution of dividends for reducing free cash flow of company. SOE is majority owned by the state. For SOEs that have not gone public (non-listed), state ownership of a hundred percent. With majority ownership of the SOEs, the government can influence the decisions of managers included in the decision to distribute dividends and capital financing. Capital Structure determine the source of non-listed state-owned enterprises, in addition to derived from government capital participation, sourced from debt financing. With most lenders or bank owned by the state, and with the guarantee of the state, then the bank will be easier to make loans (debt) to SOEs for corporate financing. The size of the company describes the value of the total assets of a company. Relationship with the size of the company dividend policy can be explained by the theory of agency costs and transaction cost theory. Based on agency cost theory, large companies tend to have more complex agency problems than small companies. In large companies, the asymmetry of information will increase as the spread of ownership (ownership dispersion), thereby reducing the ability of the owner to monitor the activities of the company and decrease the effectiveness of control by managers. With dividend payments, it will create the need for external financing (external financing), which will increase the company's monitoring activities by creditors (Easterbrook, 1984). Besides large companies tend to have easier access to external finance due to the transaction costs (transaction costs) are smaller than the size of his company. Transaction costs are largely a fixed cost (fixed cost) so that large companies benefit from economies of scale when raising debt financing. With lower transaction costs and increased potential for agency problems, the size of the companies tend to be positively correlated with dividend payments. Other studies have attributed the size of the company by, among others, the research agency cost Lloyd et al. (1985), who found that the large-sized companies, with ownership spread, will have a higher bargaining power thereby increasing agency costs. Therefore, large companies tend to pay higher dividends to reduce the agency costs. This is supported by the opinions Sawicki (2005) which states that the large companies will increase the asymmetry information because of ownership continued to spread, the lower the ability of shareholders to monitor corporate activity, resulting in inefficiencies in the management of the company by a manager. Therefore, large companies would pay a higher dividend that will increase the need for external financing (leverage), then there will be monitoring of the company by creditors. Relationship with the company size of transaction costs (transaction costs) related to how a firm gain access to the capital markets. According to Holder (1988) large firms have better access to the capital markets so it is easier to obtain financing at a lower cost, therefore the big companies are better able to pay higher dividends to shareholders. The results of this study are supported by research Mollah et al. (2002), Travlos et al. (2002). But with easier access in the search for external financing (debt), as an alternative to lower agency costs, will reduce the function of the dividend in order to control the agency problem. Besides large companies more scrutiny from the public, including the media resulting in the 5

monitoring of the public against large corporations, resulting in a commitment to dividends will decline. This leads to firm size is negatively related to dividend payments. The profits of an enterprise is the result of the company's operations that describe the company's revenue is greater than the cost incurred to obtain these revenues in a given period. Income is measured in various levels starting from the operating profit, which reflects the excess of operating revenue minus operating expenses. Earnings before interest and taxes (EBIT) is the total income before deducting interest expense and taxes. If EBIT minus interest expense and taxes represents net income (net income) of the company. The Company may use the net income to finance profitable investment, distributed as dividends to shareholders, or accumulated earnings into retained earnings (retained earning) Source dividend payments from the accumulation of retained earnings (retained earnings) acquired a company. Lintner (1956) pioneered the study of the effect of income on dividend stability of theory of dividend yield. Lintner, among others, suggests that the main factor in dividend payments are income for the year (current earnings) and dividend previous year (past dividends). This conclusion is supported by the research of Baker et.al (1985) and research Pruitt and Gitman (1991). Some studies in developed countries (developed market) found that the level of corporate profits and a significant positive effect on dividend policy (Jensen et.al, 1992), (Han et.al,: 1999), Fama and French, 2002). The results of research in developing countries (emerging markets) find the same thing. Adaoglu (2000) found that earnings (earnings) is a major factor in the decision of dividend payments on companies listed on the Istanbul Stock Exchange. Pandev (2001) found that the majority of companies in Malaysia increase dividend payments when the company profits increased, and reluctant to reduce or not pay dividends when corporate profits fell. Al- Malkawi (2007) found that the rate of profit is a very important factor that affects the amount of dividends paid to companies in Jordan. Al-Kuwari (2009) found the same thing that the rate of profit and a significant positive effect on the company's dividend payout ratio in countries gulfco-operation countries (GCC). In general, dividends are paid with cash (cash-dividend), while profits from the company's cash flow is not always followed. So that the dividend is not only dependent on the profits, but also because of the liquidity position of company. That cash availability affect the company's ability to pay dividends. According to Liu & Hu (2005) cash dividend sourced from free-cash flow indicates the maximum cash payable dividend. If the cash dividend is smaller than the free-cash flow that company has increased residual cash. If the cash dividend is greater than the free-cash flow, the company requires financing by issuing new shares to meet the requirements for payment of cash dividend. According to the agency cost theory, agency conflicts between owners and managers will increase on companies that have large free cash flow, because managers tend to use the cash balance for the benefit of its own that will increase agency costs. Payment of dividends is one of the tools to reduce the agency costs of free cash flow by reducing owned company, that is not used by managers for their own interests that do not suit the interests of the company. Several previous studies found that the cash-flow has an influence on the payment of dividends. Research Agrawal and Jayaraman (1994), Eco Chic Amidu M. (2006), and Anil & Lime (2008) concluded that there is a positive relationship between cash-flow with a dividend payout ratio. Financial leverage ratio reflects the use of debt (leverage) in the capital structure of a company. While the choice of capital structure reflects a company's financing policy. Financing policy is the decision to choose a funding source in order of acquisition of assets, so that the costs incurred 6

to acquire the asset less than the revenue generated from the use of the asset. Companies can choose entirely with equity financing or in part with the use of debt financing (leverage). The higher the proportion of debt in its capital structure, the ratio of the higher financial leverage. According to the theory of transaction costs (transaction cost theory), funding from outside (external financing) either with equity or debt (debt), will lead to transaction costs. Financing with debt raises interest payment obligations and flotation costs such as administrative costs and legal fees in order to finance the debt. So by Rozeff (1982), to maximize shareholder wealth, the company's dependence on external financing that pay low dividends, in order to reduce transaction costs. Several studies support the idea, among others, Mollah et al. (2001) that perform testing in emerging market and find that financial leverage has increased transaction costs. Companies with high leverage ratios will pay lower dividends in order to avoid the cost of transaction. According to transaction cost theory, the level of leverage will be negatively related to the dividend payout ratio. According to the agency cost theory as stated by Easterbrook (1984), the payment of dividends will reduce agency costs, which would cause the company's dividend payout seek outside financing (debt financing). With the financing with debt, then the creditor will monitor the activity of managers that will reduce the cost of monitoring is done by the owner of the company, thereby reducing agency costs (monitoring costs). Then by Jensen (1986), debt can be a substitution of dividends to lower free-cash flow, to prevent the tendency of managers to use the free-cash flow for their own interests. Therefore, based on the theory of agency cost, the company will leverage level is negatively correlated with the level of dividends. Several previous studies support this opinion. Research Jensen et al., (1992), Agrawal and Jayaraman (1994), Al- Malkawi (2005) found that financial leverage negatively affect the dividend payout ratio. In this dissertation research, proxies are used to measure financial leverage or capital structure is ratio of total debt (debt) divided by total equity (DER), which is also used research of Rajan and Zingales (1995), Chen and Roger Strange (2006), and Attaullah Shah & Saifiullah Khan (2007). Payout dividend ratio (DPAY) shows the proportion of net income that is shared in the form of cash dividends, and net income as retained earnings or accumulated retained earnings (retained earnings). This ratio indicates the percentage of corporate profits paid out to shareholders in the form of cash dividends. If the company's retained earnings for the purposes of the company's operations in bulk, meaning profits will be paid out as a dividend becomes smaller. Conversely, if the company prefers to distribute profits as dividends, then it will reduce the portion of retained earnings and reduce internal funding sources. However choose to distribute profits as dividends would increase the welfare of the shareholders, so shareholders will continue to invest their shares to the company. Chen & Dhiensiri (2009) states that normally the dividend payout ratio is defined as dividends divided by net profit after taxes. In his research, Lintner (1956) find that the decision of the dividend payment for the year (the current dividend) is affected by the previous year's dividend payment (past dividends). Several other studies support the notion that the Lintner, Fama and Babiak another study (1968), McCabe (1979), Pruitt and Gitman (1991), and IM Pandey (2003). Based on the framework, the research hypothesis can be formulated as follows: 1) The company's growth (growth of firm), capital structure (capital structure), state ownership (state ownership), firm size (firm size), profitability (profitability), liquidity (cash-flow), and dividends last year (past dividend ) simultaneously significant effect on dividend policy 7

2.1.) Growth companies (growth of firm) partially negative effect on dividend policy 2.2.) State ownership (state ownership) positive effect partially on dividend policy 2.3.) Firm size (firm-size) partially affect the dividend policy 2.4.) Profitability (profitability) partially positive effect on dividend policy 2.5.) Liquidity (cash-flow) partially positive effect on dividend policy 2.6.) Capital structure (leverage ratio) partially negative effect on dividend policy 2.7.) Dividends last year (past dividend) partially positive effect on dividend policy 2.8.) Dividend ratio of non-listed SOEs lower than dividend BUMN Listed 3. Methods and Discussion This type of research is a descriptive study-verification. Descriptive Research is research to gain an overview of some of the variables of the study which includes the company's growth, the characteristics of the company, dividend policy and capital structure. While verification is a research study to test the hypothesis of the study by using statistical analysis to determine the level of significance of the effect of variable growth and dividend policy company to characteristics. The method used is an explanatory method-survey research that aims to interpret the relationship between variables in a way interpretation first conclusions will be obtained through hypothesis testing. There are two types of variables that will be used in this dissertation research is the dependent variable (endogenous variable) and the independent variables (exogenous variables) including control variables and dummy variables. The data used in this dissertation research is primary data in the form of annual financial statements audited SOE-year period 2005-2009 were obtained from the Ministry of State Enterprises in the form of data either hardcopy or softcopy of data. Most secondary data for SOEs listed of Indonesian Capital Market Directory (ICMD). Shape data is a panel data or pooled time-series and cross-section. Selection of the study sample is purposive sample, sample selection based on objectives or criteria as follows: 1) BUMN Limited Liability Company (PT). 2) The state-owned company is not banking sector companies, financing, and insurance (nonfinancial firms) 3) The state-owned company to deposit dividend during the period 2005 to 2009. The target population of this dissertation research is BUMN shaped Persero (PT), amounting to 128 companies, including 18 state-owned companies that have gone public (listed SOEs). The unit of analysis is the state-owned company Persero form, which does not move the financial sector (non-financial firms) totaling 106 companies. Furthermore, selected SOEs to pay dividends during the 2005-2009 period and reduced SOE with incomplete data, so the final sample numbered 46 companies. Furthermore, these data are classified, arranged shaped so that the ratio can be used for data processing using eviews 6.1 applications. Once the data is collected and confirmed that the data can be processed further to subsequently test the model estimation with linear regression. Therefore, this study uses panel data, the Hausman test (Hausman test) to determine better methods, whether using a fixed effect model or random effect model. Hypothesis test is then performed to determine the relationship of the dependent variable and the independent variables. 8

In this discussion will analyze the influence of each independent variable on the dependent variable indicated by coefficients. This is done to see if the results of this study are in accordance with the rules in the theory used as a basis for dissertation research and previous studies referenced in this dissertation research, or whether there is a tendency that the results of this study differ from theory and research The former, which is the typical behavior of the dividend policy and capital structure policy in the management of SOEs Dividend Policy Model 1. Effects of Growth Variables Company From the data processing is seen that the growth of the proxy variable ratio of capital expenditures divided by total assets (capital expenditure to assets ratio) had a regression coefficient of -0.279069, so it can be concluded that there is a negative effect on the company's dividend growth, which means the higher the ratio of capital expenditure, the lower the dividend payout ratio, or vice versa. Regression results influence the growth of the company by proxy capital expenditure to assets ratio is consistent with the research hypothesis predicts that the growth of the company as measured by the proxy of capital expenditure negatively affect dividend policy. This result is consistent with the theory of transaction costs (transaction cost theory). According to transaction cost theory, for companies that are growing tend to hold back profits to finance investment, because if the company pays a dividend, then the internal funding will be reduced, so that the company will seek outside financing sources (external financing). The use of external financing will cause transaction costs. Thus according to the theory of transaction costs, dividend payments will be made if sufficient internal funds to finance investment company. The results of this study support the conclusions of some previous studies that found a negative relationship between growth companies with dividend payout ratio, among other research Rozzef (1982), Lloyd et al, E (1985), and Holder (1998). The results of this study illustrate that the theoretical and based on regression analysis of the data of SOE financial statements, the company's growth as measured by the ratio of capital expenditure proxy, has a negative relationship with dividend. The results of this study may have implications related to SOE dividend policy by the Government which has tended to meet the financing needs of the State Budget (Budget). SOE is one of the pillars of the national economic growth. Additionally SOE currently has two new task together government and private sector obligations to support Program Development Acceleration and Expansion of Indonesian Economic Development (MP3EI) by investment in infrastructure investments whose value is large enough. By away, because it required huge funds to finance these investments. But on the other hand, the role of SOEs during this turns out not optimal as one of the nation's economic growth locomotive. According to data from the Office of State Enterprises, capital expenditure (capex) of SOEs increased from Rp.32, 26 billion in 2004 to 197 trillion in 2010, or an increase of 520.3%. However, capital expenditure (CAPEX) this state is much smaller than its operational expenditure (OPEX). In 2010, OPEX SOEs reached Rp.932, 15 trillion. This means that capital expenditure on SOE is only about 21% compared with its operational expenditure. 9

Compared to the gross domestic product (GDP) of Indonesia in 2010 amounted Rp.6.436 trillion (BPS, 2012), capital expenditure (capex) SOE only about 3%. Based on these data, the government should amend the dividend policy of state-owned enterprises in order to enhance the role and contribution of SOEs to national economic growth and the provision of infrastructure. This can be done for example by changing the SOE dividend function is no longer a source of revenue in the state budget. SOE dividend policy should be made by the corporation with the interests of the company, for example by taking the residual dividend policy, where dividends are paid if the company's investment needs have been met. If the government still requires payment of dividends, the SOE dividend ratio can be lowered to some degree, so that the source of internal funds can still make a significant contribution to investment fund companies. Contribution of SOEs to the state as the owner not only through dividend payments, but can also through corporate tax payments. With the increased investment in corporate profits will rise so that the tax payments will also increase. Besides state-owned enterprises can contribute indirectly such as by the provision of adequate public infrastructure, the provision of goods / services quality and activities of corporate social responsibility or corporate social responsibility, among others, with the help of the small and medium enterprises in order to thrive. Indirect contribution of SOEs can also be done in the form of state budget efficiency, where the provision of public infrastructure can be done by SOE so not using financing from the state budget. While the company's growing influence on dividend policy by using proxy sales growth (sales growth) generates different conclusions. regression coefficient of growth companies with sales growth proxy is -0.004560, so it can be concluded that there is a negative effect of sales growth in the dividend payout ratio, but this effect was not significant. The results of this study do not support the theory of transaction costs (transaction cost theory) underlying the relationship between these variables. These results also do not support the results of previous studies such as the study Rozzef (1982), Lloyd et al, E (1985), and Holder (1998) who found that the growth of the company as measured by the ratio of sales growth proxied by total assets negatively affect the payment of dividends. Relative results of this study, several arguments can be presented, among others, that, the state has a monopoly of authority in some strategic sectors, among others, in the energy sector, mining sector, construction sector, transport sector, agriculture and plantation sector. This causes the SOE monopoly authority has a captive market so that no investment is too large, the value of sales generated will remain high. Another argument that SOE capital expenditure is not too much influence in increasing sales growth, at least in the period of the study. This problem is related to the degree of effectiveness of investments made by SOE managers. In accordance with agency cost theory argument, managers tend to make investments that benefit themselves and do not necessarily correspond with the interests of the company. Investment spending is not driving sales growth indicates that there is over-investment made by the managers of SOEs. 2. Effects of Variable Capital Structure From the results of the regression equation shows that the coefficient of the leverage variable is - 0.046283, which means that the effect of capital structure on dividend policy is negative, where the higher the debt to equity ratio, the lower the dividend payout ratio, or vice versa. Regression 10

results are consistent with the research hypothesis predicts that capital structure or financial leverage has a negative effect on dividend policy. This study supports the theory of transaction costs (transaction cost theory) and the theory of agency costs (agency cost theory). According to transaction cost theory, corporate financing from outside (external financing) with both equity and debt (debt), will lead to transaction costs. Financing with debt raises interest payment obligations and flotation costs such as administrative costs, legal fees in order to finance the debt. To maximize shareholder value, the company's reliance on external financing (external financing) that pay low dividends, in order to reduce the external financing so as to reduce transaction costs. While according to the theory of agency costs (Jensen, 1986) is a substitution of a dividend payable in order to reduce the agency costs of free-cash flow. Dividend payments can reduce agency cost of excess cash flow to the company. Similarly, the debt, because the company will have a debt financing commitment in order to pay the interest on the debt obligations that will reduce the free-cash flow in the management of the manager to prevent the use of the free-cash flow for the benefit of its own managers. In connection with the presence of function substitution, so if the higher dividend payout ratio, the debt ratio would be lower, and vice versa. The results of this study also supports some previous research results, among others, research Cructhley & Hansen (1989), Agrawal and Jayaraman (1994), Mollah (2001), Al-Kuwari (2009) who concluded that the financial leverage ratio negatively affect the dividend payout ratio. In the context of state-owned enterprises, the results of this study indicate that increased debt or leverage ratios lead to lower dividend payout ratio was remitted to the government or otherwise. SOE debt ratio declining tendency, but SOE dividend payout ratio is likely to increase. The regression results are also consistent with the regression results on the effect of firm growth (growth) of the dividend payout ratio. Increased growth companies tend to lower the dividend payout ratio, because the source of funding for most of the state-owned investment using internal funds (retained earnings), thereby reducing the source of funds to pay dividends. 3. Effects of Company Size (Firm Size) From the results of the regression equation shows that the coefficient of firm size variable is - 0.099717, where the influence of the variable size of the company dividend payout ratio (DPAY) is negative, which means that the higher the size of the company, the lower the dividend payout ratio. The results of this regression is different from the research hypothesis predicts a positive effect on the variable SIZE dividends. The results of this study do not support the argument transaction cost theory and agency cost theory. According to the theory of agency costs, large companies tend to have more complex agency problems than small companies that will increase the asymmetry of information, thus reducing the ability of the owner to monitor the company's activities, and decrease the effectiveness of control by managers. Dividend Payment will give rise to the need for external financing (external financing), which will further increase the company's monitoring activities by creditors (Easterbrook, 1984). Meanwhile, according to transaction cost theory argument, large companies tend to have easier access to external finance (debt) because of the relative transaction costs (transaction costs) are smaller than the size of his company. Transaction costs are largely a fixed cost (fixed cost) so that large companies benefit from economies of scale when raising debt financing. With lower transaction costs and increased potential for agency problems, the size of the companies tend to 11

be positively correlated with dividend payments. Holder (1988), Mollah et al. (2002), Travlos et al. (2002). However, there are arguments agency free-cash-flow why firm size is negatively related to the first dividend payment, the company that has a large asset size will have easier access in the search for external financing (debt), where the debt as an alternative to lower agency costs of free cash flow, will further lower the dividend in order to control the functions of the agency problem. Besides large companies also tend to be under the spotlight of the public including the media resulting from the monitoring activities of the public against the big companies, which resulted in a dividend in order to improve the function of monitoring is lowered. In the context of state-owned enterprises, the results of this study indicate that apparently unrelated company size and inversely proportional to the amount of the dividend payment SOE. This can be explained by the argument that, in accordance with the opinion of Borisova et.al (2011), SOE as a state-owned company, has a guarantee from the government that despite having a high dividend payout ratio, so if in need of funds to finance investment, SOEs easier to find financing from outside since it has such a government guarantee. Moreover, based on the results of the other study variables regression above, it turns over SOE investment financing using internal funds. Thus, in determining the value of dividends to be paid into the state treasury by the state, does not consider the size of the company. In addition the amount of the dividend payment decision is determined by the proportion of SOEs is more certain to profit, not by the size of the company's assets. 4) Effect of Variable Profitability From the results of the regression equation shows that the coefficient of profitability by using proxy variables return on equity (ROE) is 0.364211, which means that the effect of profitability on dividend policy is positive where the higher return on equity, the higher the dividend payout ratio (DPAY). Regression results are consistent with the hypothesis in this dissertation research which predicts that the positive effect on the profitability of the company dividend payout ratio (DPAY). The results of this study support the theory of stability of the dividend argument of Lintner (1956), among others, suggests that the main factor role in the payment of dividends is income for the year (current earnings) and dividend previous year (past dividends). The results of this study support the research Baker et.al (1985) and Pruitt and Gitman research (1991), who found that the profit for the year (current earnings) positive effect on dividends. The results also confirm the results of the research study Adaoglu (2000), Pandev (2001), Al-Malkawi (2007) who found that the profit for the year (current earnings) is a very important determinant of dividend policy. 5) Effect of Variable Cash Flow From the results of the regression equation shows that the coefficient of the variable cash flow (CF) with a proxy net cash flow to assets ratio is -0.303404, which means that the effect of cash flow on dividend policy is negative, where the higher the ratio of cash flow, the lower the dividend payout ratio (DPAY) or vice versa. The results of this regression is different from the research hypothesis predicts that cash flow has a positive effect on dividend policy. The results of this study do not support the argument that agency cost theory states that the dividend is a mechanism to reduce free-cash flow of the company so as not to be used for the benefit of managers, so that the ratio of cash flow will be positive effect on dividend policy, where the higher cash flow possessed company, the greater the dividends paid. 12

However, several arguments can be put forward why the cash-flow has a negative influence on the dividend. Cash-flow (CF) in the opposite direction with the graph dividends. At first the average SOE dividends decreased then increased, while the average ratio of cash flow to increase further SOE initially flat. This indicates that although the dividend is a dividend paid cash, but the magnitude of the ratio of SOE dividends are not always followed by the ratio of cash flow of the company. Second, the profits of the SOEs, it does not always reflect the magnitude of the cash flow of the company. This reflects the profits of SOEs tend to be income accrual accounting is not always followed by cash flow. 6) Effect of State Ownership Variables (State Ownership) From the results of the regression equation shows that the coefficient of the variable state ownership (SO) is 10.48203, which means that the effect of state ownership on dividend policy is positive, where the higher the proportion of state ownership, the higher the dividend payout ratio (DPAY) or vice versa. Regression results are consistent with the hypothesis in this dissertation research which predicts that state ownership (state ownership) positive effect on dividend payout ratio (DPAY). The regression results support the agency cost theory (Jensen & Meckling, 1976) regarding the influence of ownership structure on dividend policy and institutional ownership in accordance with the argument of Shleifer & Vishny (1986) in Wang et.al (2011) which states that institutional ownership ( institutional ownership) affects the agency problem because institutional ownership typically have large amounts of stock (large-block shareholder) so as to have the effective ability to monitor the behavior of managers in managing the company. The influence of large-block shareholders in the management of the company even greater if largeblock shareholders are government institutions or state. According to Shleifer & Vishny (1987) in Wang et.al (2011), one of the company's decision that can be controlled include the dividend decision. The results of this study also support several previous studies such as Gugler (2003), who concluded that the corporate state (state owned firms) tend to be reluctant to reduce dividends than family firms (familiy owned firms), as well as research Gul (1999), Al-Malkawi (2007), Al-Kuwari (2009) which concluded that state ownership (state ownership) has a positive effect on dividend policy. 7) Effect of Variable Dividend Past (PDPAY) From the results of the regression equation shows that the coefficient of the variable dividend payout ratio last year (past dividend) is 0.008227, which means that the influence of the past on the policy dividend payout ratio is positive. But the probability of a value of 0.3912, the effect of the ratio of dividends last year (past dividend) the ratio of current year dividend (the current dividend) is not significant. It can be concluded that the dividend policy of the period did not significantly influence the dividend policy of the current period. The results of this study do not support the argument dividend stability theory of Lintner (1956) which states that dividends last year is an important variable in determining the dividend policy of the current year. This indicates that the SOE dividend policy, dividend ratio last year was not an important variable for determining the dividend payout ratio for the year. It also illustrates that the SOE dividend policy more considering the conditions at the time the dividend policy was decided, among others, the need for financing the state budget. SOE dividend policy which does 13

not consider the stability of dividends is also influenced by the status of the majority of SOEs are still a private company (non-listed SOEs) so that the dividend policy change will not affect the value or price of shares of the company. 8.) Differences SOE dividend policy listed and non-listed SOEs From the regression results are known turns between SOE dividend policy listed and non-listed SOEs differ significantly with dummy variable coefficient is -3.547401, which means that the average dividend payout ratio of non-listed SOEs 3.547401 lower than the dividend payout ratio of listed SOEs. These results portray that the dividend payout ratio of non-listed SOEs is smaller than the dividend payout ratio of listed SOEs. The results are consistent with the hypothesis dissertation study predicts that the dividend payout ratio of non-listed SOEs is smaller than the dividend payout ratio of listed SOEs. Several arguments can be explained to support the results of such research include first, listed SOEs usually have a better performance resulting in higher profits than non-listed SOEs. By having a greater profit, then the ratio of dividends paid will also be greater. Second, more listed SOEs have better access to seek external financing, both debt and equity financing. So if internal funds (retained earnings) are not sufficient to pay dividends karean investment, it can seek funding from outside (external financing) with easier and less expensive. Better access to external financing sources is due to several reasons, among others: 1) Listed SOEs has become a public company so that the lender or investor to assess the condition of the company to more easily and accurately. 2) Listed SOEs more financially sound so it is considered to have a better ability to meet its debt obligations to creditors or lenders can provide greater financing. 3) Listed SOEs tend to have larger assets so as to give confidence to investors over the collateral (collateral) against loans. This also leads to easier creditors lend to SOEs. 4) Based on these factors, the creditor or lender will likely apply a lower interest rate for SOEs listed compared to non-listed SOEs. On the average size of SOEs listed company (firm size), the ratio of profit (return on equity), debt ratio (debt to equity ratio) and the ratio of dividends (DPAY) higher compared to the nonlisted SOEs. This means that on average, listed SOEs have better performance than non-listed SOEs views of the four measures. The results of this study confirm the results of research and Netter and Megginson (2001) and Sun & Wilson Tong (2002) who found that the performance of SOEs after privatization are much better indicators measured from the increase in profitability, capital expenditure (capex) and the dividend payout ratio. Test results of the feasibility model (Goodness of fit models) Good model, it must meet several requirements to meet the goodness of fit model. From the 14