Relationship between ownership concentration and dividend policy evidence from listed companies in HOSE

RELATIONSHIP BETWEEN OWNERSHIP CONCENTRATION AND DIVIDEND POLICY: EVIDENCE FROM LISTED COMPANIES IN HOSE In Partial Fulfillment of the Requirements of the Degree of MASTER OF BUSINESS

RELATIONSHIP BETWEEN OWNERSHIP CONCENTRATION AND

DIVIDEND POLICY: EVIDENCE FROM LISTED COMPANIES IN HOSE

In Partial Fulfillment of the Requirements of the Degree of

MASTER OF BUSINESS ADMINISTRATION

DIVIDEND POLICY: EVIDENCE FROM LISTED COMPANIES IN HOSE

In Partial Fulfillment of the Requirements of the Degree of

MASTER OF BUSINESS ADMINISTRATION

In finance

By Ms: Tran Thu Phuong ID: MBA04029 International University - Vietnam National University HCMC

To complete this thesis, I have been benefited from the following people First of all, I would like to express my special thanks to my advisor – PhD Nguyen Kim Thu – who recommended the research idea to me and for all of her enthusiastic instruction and encouragement during the time I do this thesis Besides, I would like to dedicate my gratitude to IU lecturers who communicated abstruse knowledge to me during MBA program, and the members of the Examination Committee for taking time and giving valuable comments to improve this study

I also want to give my sincere thanks to my friends and my classmates who are my inspiration and always encouraged me during the past time

Last but not least, I want to express my loving thanks to my family who always stand beside me My greatest gratitude is to my mother, Trinh My MBA program in IU - VNU would have not been completed without her love and encouragement

HCM City, August 2013

TRAN THU PHUONG

Plagiarism Statements

I would like to declare that, apart from the acknowledged references, this thesis either does not use language, ideas, or other original material from anyone; or has not been previously submitted to any other educational and research programs or institutions I fully understand that any writings in this thesis contradicted to the above statement will automatically lead to the rejection from the MBA program at the International University – Vietnam National University Ho Chi Minh City

Chapter 1 – Introduction 1

1.1 Background of the Study 1

1.2 Problem Statement 2

1.3 Rationale of the Research 4

1.4 Importance of the Research 6

1.5 Research Question 6

1.6 Research Objective 7

1.6.1 General objective 7

1.6.2 Specific objective 7

1.7 Scope and Limitation of the Research 7

1.8 Research Approach 8

1.9 Thesis Structure 9

Chapter 2 – Literature Review 10

2.1 Related Theories 10

2.1.1 Dividend theories 10

2.1.2 Effect of Ownership concentration on dividend policy 23

2.2 Previous empirical studies 25

2.2.1 Abroad 25

2.2.2 Vietnam 30

Chapter 3 – Research Methodology 33

3.1 Data Collection 33

3.2 Research Model 34

3.2.1 Variable definition and empirical studies 34

3.2.2 Model and hypothesis 38

Chapter 4 – Data analysis 43

4.1 Descriptive Statistics 43

4.2 Analysis of the results 45

4.2.1 Pearson Correlation 45

4.2.2 Pooled OLS and Random Effect Model (REM) 48

4.2.3 Model (1) and Model (2) with Pooled OLS analysis 52

4.2.4 Model (3) and Model (4) with REM 52

5.1 Discussion 57

5.1.1 Ownership concentration and dividend policy 57

5.1.2 Firm’s characteristics and dividend policy 58

5.1.3 Ownership concentration types and dividend policy 59

5.2 Conclusion 62

5.3 Recommendation 62

List of Tables

Table 2.1 Correlation relationship results between dividend policy and independent

variables in Vietnam firms 32

Table 3.1 Definitions of variables 38

Table 3.2 Empirical result of the effect of government and foreign investors on dividend policy 41

Table 4.1 Ownership concentration types of the sample firms 44

Table 4.2 Descriptive statistic of the sample firms 45

Table 4.3 Partial covariance analysis 46

Table 4.4 Pearson Correlation 47

Table 4.5 Pooled OLS 49

Table 4.6 Random effect model 50

Table 4.7 Breusch-Pagan test 51

Table 4.8 Breusch-Godfrey Serial Correlation LM test 52

Table 4.9 Tobit analysis 56

List of Figures

Figure 1.1 Cash dividend/par value of listed companies in HOSE in 2011 and 2012 5 Figure 2.1 Definition of Variable 29

List of Abbreviation

The main purpose of this research is to study the relationship between ownership concentration and dividend policy in term of cash dividend of listed companies in HOSE

The sample consisted of 154 listed firms in HOSE in two years 2011 and

2012 The sample was divided into 5 groups of ownership concentration in order to achieve the different relation of ownership concentration types with dividend policy Pooled OLS, random effect model and Tobit model were used to test the hypotheses

The results show that ownership concentration has negative and significant relationship with dividend policy in general Different types of ownership concentration have different impact on dividend policy

Keywords: Ownership concentration, dividend policy

Chapter 1 – Introduction

In order to make clear why this study was conducted, chapter 1 brings an overview of the research Background of the Study generally raises the issue about dividend policy and ownership concentration as the start point of the Problem Statement

in which the arguments about this relationship are controversial Vietnam market reality also is described in this section as the support reason for the researcher to choose this topic The Scope and Limitation of this research are also provided

1.1 Background of the Study

Dividend policy has been one of the topical issues in financial management because dividends are a major cash outlay for many corporations It is the decision on how much profits will be paid out as dividend to shareholders and how much will be kept as retained earnings to reinvest in the company

According to Miller and Modigliani (1961) dividend irrelevance proposition, dividend policy does not matter in a tax-free world The argument was that if the firm pays dividend to their shareholders, the firm doesn’t have enough retained earnings to reinvest, they would borrow funds, and the loss in value of existing share would be exactly equal to the dividend paid out as a result of borrowing instead of using internal fund (Davies, Boczko & Chen, 2008) Thus shareholders are indifferent between dividend and capital gain However, this argument has been challenged at present If dividends are irrelevant, why companies still pay dividends and why investors are aware

of dividends In reality, there are many theories and research support to the dividend relevant argument such as Bird-in-hand theory by Gordon and Walter (1963), Agency

theory by Jensen and Meckling (1976), Tax preference hypothesis by Brennan (1971), Transaction cost theory These theories showed that, a change in dividend policy will influence the stock value, by that, affecting the market value of the firm These theories will be summarized in Literature Review part

Consequently, if paying dividends is important for companies, then we need to identify factors influencing the dividends policy Following theories of dividend policy,

we can see that dividend policy has a significant effect on shareholders’ benefit, thus they should control the dividend policy to act on their best interest Their pressure on managers may make them follow their willing to pay or not to pay dividend and how much of the profit of the company should be distributed as well as how much of it should be invested in the form of accumulated profit in the company However, different investors will have a different perspective about dividend Some ask for high dividend, some want the firm to keep retain-earning to continue invest, while others just think that there is no difference between dividend and capital gain Therefore, this thesis aims to identify the relationship between dividend policy and the ownership concentration factor with evidence from HOSE stock exchange

1.2 Problem Statement

The ownership of a firm involves many investors, and the ownership structure is divided to two main structures: concentrated and dispersed According to T X T Nguyen (2010) and Ullah, Fida and Khan (2012), the number of large-block owners and the total percentage of the company's shares that they own define ownership concentration (large-block shareholders are investors who typically own at least five

defined as has large number of shareholders with smallholdings (www.openlearningworld.com)

According to the agency cost theory, the more dispersed the ownership structure

is, the more severe the agency problem is, therefore, the need for monitoring managers also increases Monitoring managers’ behavior is more difficult when the ownership structure is widely dispersed If shareholders involve in the monitoring, they have to incur the full cost of monitoring, while they just gain the benefit according to their relative share of ownership, which is very small The monitoring activity would be more effective if there were some agents who monitored managers on shareholders’ behalf If dividends can act as a monitoring mechanism by reducing cash available for managers’ perquisite consumption, a positive relationship between ownership dispersion and dividend payout ratio is expected (Chen & Dhiensiri, 2009)

However, according to Jensen and Meckling (1976), Anwar and Tabassum (2011), large owners have stronger incentive and better opportunities to exercise control over managers than small shareholders In that case, they may force managers to have a high dividend payout ratio, so as to reduce cash available to managers, and to transfer wealth from bondholders to shareholders, especially when firms are in financial distress

The controversial between two arguments is still unsolvable, thus the relationship between ownership concentration and dividend policy is unpredicted

1.3 Rationale of the Research

According to The Decision No.127/1998/QDD-TTg dated July 11th 1998, the Securities market of Vietnam was established in 2000 Ho Chi Minh stock exchange was implemented the first transaction on July 28th 2000

Listed firms in Vietnam usually pay dividend 2 times per year:

- The first time (usually in March and April): Most of firms, at this time (after the end of financial year), already had the result of last year operating and audit report, they will announce to distribute the profit and dividend that approved by shareholders through general meeting of shareholders

- The second time (usually in July and August): Firms had result of half-year operating usually pay dividend in advance based on target and results achieved in half a year

Follow D.L Nguyen (2008) in his research about dividend policy of listed companies in Vietnam stock market from 2000 to first quarter 2008, at the time when companies announce dividend payment, price of share tend to increase He explained that there were plenty of different reasons for the increase of stock price such as the good condition of Vietnam macro economic in those years, investor’s “herd” psychological, the rumors about achieving profitable in the stock purchase, the information about the sale of the foreign investors to name just a few He concluded that dividend policy is still a factor affecting stock price in Vietnam market, although the level of impact and how the impact is different for different stock and different market

By the data obtained from HOSE stock exchange, we have the overall look on

(Source: Calculated by researcher)

Figure 1.1 Cash dividend/par value of listed companies in HOSE in 2011 and 2012

More than 50 companies out of 273 listed companies in HOSE in 2011 did not have clearly information about their dividend rate This number increased even more in

2012, from more than 50 companies to nearly 80 companies out of 273 listed companies

in HOSE

Besides those companies, we also see that Vietnam companies have quite diversified dividend policies We have companies that pay dividend from 0% to 70% on capital However, the ratio from 0% to 20% is preferred by most companies There were approximately 60% of companies in HOSE choose to pay dividend from 0% to 20% on their capital, and only about 10% in total 273 listed companies in HOSE pay dividend from 21% to 30% There were very least companies pay dividend above 30%

Why does our market have such kind of dividend policies diversification? Is there any effect from the ownership concentration types on it? To find out, this thesis:

Cash dividend/par value

Cash dividend/par value of listed companies in HOSE

Year 2011 Year 2012

“THE RELATIONSHIP BETWEEN OWNERSHIP CONCENTRATION AND

DIVIDEND POLICY: EVIDENCE FROM LISTED COMPANIES IN HOSE in two

years 2011 and 2012” is conducted

1.4 Importance of the Research

The relationship between dividend policy and firm’s ownership structure has been researched by several studies within the agency theory framework for different countries In Vietnam, it is important to examined this relationship because of the variety

of owners in Vietnam companies, for example: government-controlling, foreign-holding, equal share holding by government and foreign investor, institutions, family owners, or mixed of institution, family, foreign and government,… These types of shareholders may have different preferences for dividend payouts

The importance of this thesis stems from the role of dividend policy to attract more capitals and investment and protect the right of minority shareholders Besides, the lack of empirical studies to prove the relationship between ownership concentration and dividend policy in emerging markets and particularly in Vietnam provides one of the motivations for this thesis

1.5 Research Question

This study seeks to answer the following questions:

- Is there any relationship between ownership concentration and dividend policy within the companies which are listed in HOSE? If yes then it is negative or positive effect

- Does different type of ownership concentration have different impact on dividend policy?

1.6 Research Objective

1.6.1 General objective

The general objective of this thesis is to clarify the relationship between ownership structure and dividend policy It will employ an ownership frame work and compare the dividend policy of companies with different ownership concentration types within the companies which are listed in HOSE

- Using theories and realities of Vietnam market to explain these relationships

1.7 Scope and Limitation of the Research

This research focuses on Joint Stock Companies listed in HOSE in 2011 and

2012

Data are collected from annual reports, financial reports, company websites or websites of other securities companies Those data may not show the true value due to the lack of transparency of some company reports and the inconsistence in information disclosed by companies and information on websites

This research is only based on data and the fact that there was a lack of survey to get the perspective of investors on dividend policy and number may not show every aspect of the issue

Only two recent years are researched, therefore, the result will not show the trend

of dividend policy through time

One more limitation is the limited understanding of researcher about statistic technique and statistic software

1.8 Research Approach

The Quantitative Method is used in this study

Theories and empirical studies are used to orient and determine variables in the beginning

EVIEW 6.0 is used to run model and get data analysis

Along with theories and empirical studies, the real situation of Vietnam is also taken into consideration when analyzing and explaining data

1.9 Thesis Structure

The thesis is expected to cover 5 chapters as follows:

Chapter 1 – Introduction: Describes the overview of this study, why the author chose this topic, the purpose and the benefit of the study The scope and limitation of the research also are shown in this section

Chapter 2 – Literature Review: In this chapter, a literature overview of dividend policy and ownership concentration will be provided The section starts by summarizing different theories supporting / ruling out the effect of dividend policy – the controversial arguments After this, some empirical studies will be given as support evidence of the theories

Chapter 3 – Research Methodology: Explains how this study is conducted, method of collecting data, and the variables that are used to study, the tools and method that are used to analyze data In this part, hypothesis also is given

Chapter 4 – Data analysis: After in put data in EVIEW 6.0, the results of Pooled OLS, REM and Tobit regression will be given Statistical analysis will be used to analyze these outcomes

Chapter 5 – Discussion and Conclusion: After have the result of regression, empirical studies, theories and practical of Vietnam market are used to explain the relationships between dividend and other factors, especially concentrate on the effect of ownership concentration and ownership concentration types on dividend policy

Chapter 2 – Literature Review

To understand the controversial in the relationship of ownership concentration and dividend policy, the literature on dividend policy, the effect of ownership concentration on dividend and some empirical studies over the world and Vietnam are reviewed

2.1 Related Theories

Dividend is the portion of company’s profit after tax, decided by the Board of Directors, distributed to its shareholders

Dividend policy is the policy a company uses to decide how much it will pay out

to shareholders in dividends and how much will be kept as retained earnings to reinvest

in the company

2.1.1 Dividend theories

Over the last fifty years, dividend policy has always been a controversial issue Lots of theories and empirical researches have been conducted and studied to solve this contentious puzzle There are three main contradictory views: (1) dividend policy has no effect on the market value of the firm, (2) an increase in dividend paid out would increase firm’s value, and (3) a rise in dividend payment reduces firm’s value Theories that support to these views are briefly described below:

2.1.1.1 Dividend irrelevance theories – Dividend policy has no effect on the market value of the firm

The most well-known argument supporting this view was published in 1961 by

Miller and Modigliani “Dividend policy, growth and the valuation of shares” (Journal of Business) To prove their argument, an idealized world of perfect markets and rational

investors are assumed that: (1) no tax difference between dividends and capital gains, (2)

no transaction costs on securities trading, (3) investors have the equal and costless access to the same information, (4) no transactions is large enough to affect the price According to MM, the value of a company is only determined by the earning power of its assets and investments They figure out only cash flow created by investment decisions or its earning power and business risks affect firm’s value not by the decisions

on how to distribute that income Therefore, shareholders are indifferent between dividend and capital gain

Numerous studies were conducted and had the same result of no relevant outcomes of dividend and stock prices In 1959, Gordon collected price, dividend and earnings data of companies in four industries (Chemicals, Foods, Steels and Machine Tools) and tested if there was a relationship between stock prices and dividends and income of these firms However, the coefficients were too low to have a real impact Another research by Black and Scholes (1974) extended the capital asset pricing model (CAPM), which was developed by Sharpe (1964) and Lintner (1965) in their research of dividend yield effects based on 25 portfolios of common stocks listed on the New York Stock Exchange [Cited by Kinkki, 2001] The objective of this study is to determine the effect of dividend policy on stock price and they founded that there is no influence of

high yield or low yield pay out policy on stock price Other evidences support for Dividend Irrelevance Hypothesis was presented in studies of Miller and Scholes (1982), Hess (1982), Miller (1986), Bernstein (1996) [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010]

However, it is obvious that those assumptions are unrealistic in the world we are living Thus, if we apply this argument in real world, it may not reflect the true importance of dividend policy As a result, there are reasons to believe that dividend policy has impact on firm’s value Next, we will go through theories support to dividend relevant view

2.1.1.2 Dividend relevance theories – Dividend policy affect the stock price

2.1.1.2.1 High dividend payout increases stock value

Bird-in-hand theory

In 1963, Gordon and Walter developed their theory as a counterpoint to M&M dividend irrelevant theory According to the bird-in-hand theory investors prefer the certainty of dividend payments to the possibility of substantially higher future capital gains They argued that investors are risk averse and they consider current dividend payments to be more trustworthy than the promise of a higher capital gain in the future because it contained risky Hence dividend payments are preferred than future capital gains due to minimizing risk Stated another way is “bird in the hand is worth more than two in the bush” Lintner (1962) and Gordon (1963) have implied that return from the

dividend is less risky than the future growth rate As a higher current dividend reduces uncertainty about future cash flows, a high payout ratio will reduce the cost of capital, and hence increase share value Studies that provide support for this theory include Gordon and Shapiro (1956) Gordon (1959, 1963), Lintner (1962), and Walter (1963) [Cited by Malkawi, Rafferty & Pillai, 2010]

However, M&M showed that the Bird-in-hand theory has a fallacy M&M (1961) argued that the firm’s risk is determined by the riskiness of its operating cash flows, not

by the way it distributes its earnings After M&M, Bhattacharya (1979) also suggested that the reasoning underlying this theory is fallacious He claimed that the riskiness of a firm’s cash flow influences its dividend payments, but not the other way (increases in dividends will not reduce the risk of the firm) Friend and Puckett (1964) found that firms with greater uncertainty of future cash flow (risk) tend to adopt lower payout ratios Rozeff (1982) also found a negative relationship between dividends and firm risk (the higher risk, the lower dividend payment).Recently, Baker, Powell and Veit (2002) surveyed managers of NASDAQ firms including the view of Bird-in-hand theory The question was that: “investors generally prefer cash dividends today to uncertain future price appreciation” Out of 186 responses, there was only 17.2% approving the view of Bird-in-hand theory [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010]

There is limited number of empirical supports for Bird-in-hand, and it has been challenged especially by M&M (1961) with many other strong support studies Nevertheless, the initial reasoning of Gordon (1961), that is, dividend received today is much preferred than highly uncertain capital gains from questionable future investments,

is still sited

Agency theory

In 1976, Jensen and Meckling developed the agency theory which describes the conflict of interests between agents (managers) and the principals (shareholders and other stakeholders) They claim that the different incentives between managers and holders can cause problems

According to M&M’s perfect capital market assumptions, there are no conflicts

of interests between managers and shareholders However, in practice where owners and managers of firms are separated, these differences really cause problems The condition

of perfect market in these cases become impossible, managers are always imperfect agents of shareholders This is because, even though maximize shareholder wealth is the primary objective of a business, managers’ interests are not necessarily the same as shareholders’ interests Therefore, managers may not act on the best interest of shareholders As the managers have the effective control of the firm, they have the incentive and the ability to consume benefits that are costly to shareholders For example, when company has surplus cash, managers may over-invest in negative net present value projects Agency cost are an implicit cost from the potential interests conflicting between agent and principal and the payment of dividend to shareholder is seen as the effective control to lowering the agency cost of equity by reducing the discretionary funds available to managers (Rozeff, 1982; Easterbrook, 1984; Jensen, 1986; Alli, Khan & Ramirez, 1993)

In case of bondholders, shareholders are considered as the agent, there might be conflict occur between shareholders’ and bondholders’ interests When a firm pay high dividend to shareholders, it reduce debt payments This action is considered as

shareholders expropriating wealth from bondholders, the result of reduced cash flows are not a healthy sign for a bondholder (Jensen & Meckling, 1976) Thus, bondholders prefer to constraint the dividend payments to secure their interests Conversely, for the same reasons, shareholders prefer high dividend payments (Ang, 1987) [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010]

Following Easterbrook (1984), there are two forms of agency costs:

(1) The cost of monitoring of managers: Because the number of shareholders is large and the dispersion of ownership structure, it would be wealthier if there were some people monitoring managers on shareholders’ behalf The collective actions, however, make the monitoring more difficult and the agency problem more severe The dividend payment can reduce the extent of managers and lower the agency costs (Al-Najjar & Hussainey, 2009)

(2) Risk aversion on the part of managers: Easterbrook claimed that, shareholders are well-diversified investors will only be concerned about non-diversifiable risk or in other way we can said that shareholders tend to want risky projects because they have diversified away their systematic risk Managers, though, often have substantial part of their personal investment tied up in their firm, if the firms do poorly or go bankrupt, managers will lose their jobs Therefore, managers will be concerned about total risk and they will not invest in risky projects By altering firm’s mix of projects and its debt-to-equity ratio, managers can change the risk of the firm The lower the ratio of debt-to-equity, the lower the chance of bankruptcy Once again, there will be conflicts between shareholders and bondholders Shareholders, hence, want high dividend payment to avoid advantage-taking by bondholders just as bondholders want to limit dividend to prevent being taken advantage by shareholders

Easterbrook also showed that dividend could be used to reduce the free cash flow into the hands of managers He argued that paying high dividend will constrain managers to acquire external financing to raise additional funds In this case, firm will

be under the strict control of investment professionals such as bankers, financial analysts

or institutional investors This suggests that by increasing dividend payment, it increase management scrutiny by outsiders and reduce the freedom of managers Therefore, shareholders are able to monitor managers at lower cost and minimize any collective action problems However, increasing dividend payments might force managers to take undesirable actions like increasing firm leverage, which may sometimes increase the riskiness of the firm

Another explanation for why firms pay dividends based on agency theory is given by Jensen (1986) He stated that when firm has excess cash flow, managers tend to using funds in their own wealth instead of shareholders’ best interests Managers want to enlarge their firm size rather than paying dividends since they expect to increase their compensations or excessive salaries they will gain in comparison with the small firms (Gaver & Gaver, 1993) Thus, extracting the surplus cash by paying dividend will reduce the free cash flow under managers’ control, thereby preventing managers to undertake negative NPV projects As a result, paying more dividends will reduce the agency costs between managers and shareholders

However, accepting that paying dividend can solve the agency problem, at the same time, shareholder should be will to accept that firm being more indebted and they have to paying higher personal tax rates on dividend

2.1.1.2.2 Low dividend payout increase stock value

Tax-effect hypothesis

State the opposite perspective with Bird-in-hand hypothesis, tax-effect hypothesis suggests that low dividend payment decrease the cost of capital and increase the stock price

Back to the perfect capital market assumptions of M&M which assumed that there is no difference in tax treatment between dividends and capital gains, this theory really brings a tough challenge to M&M’s irrelevance hypothesis (1961) Tax, in the real world, is a very important issue that needs to be concerned Due to the different in tax treatment between corporate tax and personal tax, its existence may have significant impact on dividend policy and the value of the firm Shareholders may prefer firms to retain their earning to reinvest and then return in form of capital gains with lower tax rate rather than dividends which charge with higher personal tax rate Besides, dividends are taxed immediately while taxes on capital gains are deferred until the stock is actually sold Low dividend payments, hence, attract investors who have favorable tax treatment

on capital gains Therefore, a low dividend payout ratio will lower the cost of equity and increases the stock price

Base on the fact that in many countries, dividend as personal income is taxed at higher rate than capital gains Thus, investors who have to bear high tax rate might require higher pre-tax return to hold stocks with higher dividend yield in compensate for the tax disadvantage of dividend income The relationship between pre-tax return and dividend yield is the basis concern of tax-effect hypothesis

To test this relationship, Brennan (1970) developed an after-tax version of capital asset pricing model (CAPM) He found out that a stock’s pre-tax returns should be positively and linearly related to its dividend yield and to its systematic risk It means that stocks have higher pre-tax return also has higher dividend yield to compensate investors for the tax disadvantage of these returns The suggestion is that, with other things remain the same, stock with higher dividend yield will sell at lower price to uphold the same after tax rate of return

Using Brennan’s model (1970), there were many studies support to his argument such as Litzenberger and Ramaswamy (1979), Blume (1980), Poterba and Summers (1984), Kalay and Michaely (2000) This hypothesis, however, still contain some weakness Using the same model as Brennan (1970), Black and Scholes (1974) found no evidence of tax effect They conclude that low or high dividend yield stock has no impact on stock’s return either before or after tax In addition, Miller and Scholes (1982) debated that positive yield-return relation is due to information effects but not tax effects [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010]

Transaction cost theory

Transaction cost is the cost incurred when buying or selling securities As the fact that firm with low transaction cost on equity or debt issuance is more likely to pay high dividend, transaction cost theory is another explanation for why firms pay dividend

Transaction cost may influence different clienteles’ preference toward dividend policy For small investors such as retirees, income-oriented investors, and so on, who consider dividend as a primary source of income for their consumption needs, high and

selling stock might be significant for such investors On the other hand, some wealthy investors who don’t consider dividend as the needs of liquidity, low dividend stocks are preferred to avoid the transaction costs (Bishop et al., 2000) The transaction cost theory incurred in making an economic exchange really criticized M&M’s notion of homemade dividend The existence of such cost in the real world brings a strong argument that dividend policy is not irrelevant

Another consequence of transaction cost when paying dividend is that firm may need to raise external fund such as issue new stock or bond to restore the cash paid out

as dividend to meet the firm requirement in a new investment opportunities The cost of issuing new equity or debt might be significant, hence firms are most likely to retain a high proportion of their earning to invest rather than rely on external source of fund This argument also explains why firm (in developing or developed capital market) use their retained earnings as the major source of finance (Fazzari, Hubbard & Petersen, 1988) To reduce or avoid such significant cost, firm can lower their dividend pay-out ratio or not paying at all As a result, a negative relationship is expected between transaction costs and dividend policy In the researches of Higgins (1972), Cruthchley and Hansen (1989), Alli et al (1993), they also had the same conclusion

However, in practice, we still see that many firms choose to pay dividend and issuing new equity and debt at the same time Therefore, other factors also has effect on dividend policy are suggested

2.1.1.2.3 Other theories of dividend policy

Clientele theory

Clientele theory suggests that different groups of investors will have different preference about the type of shares to invest in based on the fact that investors have to face different tax treatments for dividend and capital gains, and transaction costs will be incurred when they trade securities M&M (1961) pointed out that this preference might

be impacted by the imperfections of market such as tax regime, transactions costs or government regulations Therefore, clientele effect might influence a firm’s dividend policy to attract certain clienteles However, they still maintained their point of view that

in a perfect market condition, each clientele is “as good as another”, thus the value of firm is not affected and dividend policy remains irrelevant

The result of the research by Allen, Bernardo and Welch (2000) had brought a significant proof of clientele theory Their result suggest that clienteles such as low-tax institutional investors tend to be attracted by dividend paying firms because of the institutional charters such as the “prudent man rule”1 that prevent them from investing in non-paying or low-dividend stocks On the aspect of firms, good quality firms also prefer to pay dividend to attract institutional clienteles because institutional shareholders are better informed than individual investors and have more ability to monitor or detect

1 Prudent man rule: the requirement that a trustee, investment manager of pension funds, treasurer of a city

or county, or any fiduciary (a trusted agent) must only invest funds entrusted to him/her as would a person

of prudence, i.e with discretion, care and intelligence Thus solid "blue chip" securities, secured loans, federally guaranteed mortgages, treasury certificates, and other conservative investments providing a reasonable return are within the prudent man rule Some states have statutes which list the types of investments allowable under the rule Unfortunately, the rule is subjective, and some financial managers have put funds into speculative investments to achieve higher rates of return, which has resulted in

firm quality, which boosts equity value to a higher level than that of non (or less) dividend paying firms They conclude that “…these clientele effects are the very reason for the presence of dividends ”

Signaling theory

Signaling theory was developed by Bhattacharya (1979), Miller and Rock (1985) and John and Williams (1985) [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010] This theory is based on the fact that, there always is an existence of asymmetric information between the insiders (such as managers and directors) and the outsiders (such as shareholders and bondholders) Managers usually have more information about the firm’s performance and future prospects than which is outsiders possessed These gaps of information between managers and investors may cause the market makes a wrong estimation about firm’s true intrinsic value If so, share price may not always be

an accurate measurement of the firm’s value, hence, the stocks might be traded undervalue To solve out this problem, managers tend to share their internal information

to the market so they can estimate more accurately the true value of the firm Because of this asymmetric information, the outsiders have to assess the market value of the securities based on the actual cash flows (Baskin & Miranti, 1997) For this reason, dividend seems to be a useful tool for managers to convey their private information to investors

According to signaling theory, by choosing a dividend pay-out ratio, managers send implicit information about firm’s performance Usually, announcements of increasing dividend signal positive information about firm’s profitability, stability and future prospect In contrast, a dividend decrease is considered as a negative signal about

future earning prospect and produces a decrease in share price Therefore, managers should firstly well estimate about firm’s prospects before share this information to the market Secondly, the internal information of the firm should be true If not, a poor performance firms can not maintain its dividend payment in the future It has to cut down the dividend soon or late, and this action, as we mention before, sends a bad signal

to investors This theoretical point was also supported by Lintner (1956), Lipson, Maquieira and Megginson (1998), unless managers believe in the long-run sustainable earning, they don’t tend to raise dividend [Cited by Malkawi, Rafferty & Pillai, 2010] This is also known as the “dividend-smoothing hypothesis” Hence, if these two conditions are fulfilled, the market can rely on the signal and should react favorably to the increasing of dividend and unfavorably otherwise (Ang, 1987, Koch & Shenoy, 1999) [Cited by Malkawi, Rafferty & Pillai, 2010]

Even though signaling theory may sound sensibly, sometime dividend policy can also send a false signal to the market When a firm reduces its dividend payment, as a result, market will see that as a negative signal about firm’s performance and its future prospect However, this firm in fact decides to retain its earning to invest in a new project that would increase its future value The case of FPL Group, the parent company

of Florida Power and Light Company is a typical case about this market mistaken Soter, Brigham and Eavanson (1996), according to this case show that, market actually did respond negatively with the dividend cut down and the share price of FPL dropped [Cited by Malkawi, Rafferty & Pillai, 2010] However, after realizing the reason for the dividend reduction, its stock price recovered This case not only showed the market false about signaling but also the positive relationship between signaling effects and dividend

There are many studies supporting signaling theories such as studies of Petit (1972), Aharony and Swary (1980), Woolridge (1983), Asquith and Mullins (1983, 1986), Kalay and Lowenstein (1985), Healy and Palepu (1988), Michaely, Thaler and Womack (1995) They found that market reacts positively (increase) to dividend initiations and negatively (decrease) to dividend omissions Recently, there are some findings in favor of signaling theory such as Nissim and Ziv (2001), Travlos, Trigeorgis and Vafeas (2001) [Cited by Kinkki, 2001; Malkawi, Rafferty & Pillai, 2010]

However, some studies also showed that dividend changes may not a good assessment of firm’s future Watts (1973), Gonedes (1978) and Benartzi et al (1997) found that dividend changes reflected the past growth of firm’s earning rather than signal the change in future earnings [Cited by Kinkki, 2001]

2.1.2 Effect of Ownership concentration on dividend policy

In corporate policies, the level and the type of ownership concentration is very significance as they can use their rights (for example voting right) to influence the important decisions of the firm such as dividend policy Ownership concentration refers

to the amount of stocks held by large shareholders Large shareholders or shareholders are investors who own equal or greater than 5% of a firm’s stocks (Ullah, Fida and Khan, 2012; T X T Nguyen, 2010)

block-2.1.2.1 Higher the ownership concentration, higher the dividend payment

Jensen and Meckling (1976), Anwar and Tabassum (2011) supported for the viewpoint that large owner have stronger incentive and more opportunities to monitor managers than small shareholders Therefore, in case those large shareholders are not

insiders, to set managers under their control and act on shareholders interest, they may force the firm to pay high dividend Claessens and Djankov (1999) claimed that the high concentration of owners reduce managers’ intentions to invest in low return project, hence more cash flows can be pay-out as dividends Kouki and Guizani (2009) showed that Tunisian firms having high level of ownership concentration pay-out more dividends Ramli (2010) with the evidence of Malaysian listed companies found a positive relationship between ownership concentration and dividend payment because controlling shareholders have greater influence over the dividend payout policy

2.1.2.2 Higher the ownership concentration, lower the dividend payment

On the other hand, there are many studies which had been conducted to define this relationship and had the conclusion that the higher the level of ownership concentration, the lower the dividend pay-out ratio Based on agency cost theory, more concentration in ownership reduces agency conflicts Hence, high dividend payment for the purpose of controlling managers’ action is not necessary In fact, several studies suggest that firms with large concentrated ownership have low dividend pay-out ratio According to Shleifer and Vishny (1986), concentrated ownership creates the incentives for large shareholders to control the firm’s management with the purpose of protecting their investment Ownership concentration controls the free-rider problem2 associated

2 Free-rider problem (Definition from: http://www.investopedia.com/terms/f/free_rider_problem.asp ) (1) In economics, the free rider problem refers to a situation where some individuals in a population either consume more than their fair share of a common resource, or pay less than their fair share of the cost of a common resource

(2) In the context of a brokerage firm, a free rider problem refers to a situation where a client has been allowed to purchase shares without actually paying for them, and then subsequently sells the shares

with ownership dispersion (dispersed ownership refers to large number of small shareholders who own less than 5% of a firm’s stock) In addition, because large shareholders play an important role in firm’s decision, managers will also give them preferences Farinha (2003), Chen et al (2005) showed that firms in US, UK and Hong Kong respectively, ownership concentration (especially concentrated insider ownership) have the negative effect on dividend policy When interests of shareholders and managers are close, less dividend are paid out The result of the research by Maury and Pajuste (2002), Gugler and Yutoglu (2003) showed that firms with the existence of high ownership concentration tend to pay lower dividends Most recently, Harada and Nguyen in their studies about "Ownership concentration and dividend policy in Japan" in 2011 also showed the negative effect of ownership concentration on dividend

Khan (2006) presented an analysis of 330 large quoted firms on UK stock exchange from 1985 to 1997 to figure out the relationship between ownership structure

and dividend policy An econometric model of company dividend was used analyze the relationship:

D i,t = α + βΠ i,t + γS i,t + φLev i,t + θO i,t-1 + u i,t

u i,t = S i,t where η i + υ i,t

In which:

Di,t is the level of gross dividend paid by firm i in period t

S i,t is sales

Π i,t is net profits

Lev i,t is financial leverage

previous financial year

is the composite error term

η i is the unobserved firm specific effects

υ i,t is the idiosyncratic shock

The result showed a negative effect of ownership concentration on dividend policy Renneboog and Trojanowski (2005) also came up with the same conclusion [Cited by Khan, 2006] Further more, Khan claimed that the identity of shareholders also play an important role, a positive relationship between dividend and insurance companies ownership, and negative with individual ownership

Harada and Nguyen (2011) tested the effect of Ownership concentration on dividend policy in Japanese firms They examined listed firms on Tokyo Stock

were used to evaluated dividend policy are dividend pay-out (PAYOUT) and dividend yield (DIVEQTY) PAYOUT is the total dividend payments to operating income and DIVEQTY is measure by dividends to book value of equity Ownership concentration was calculated by summing the squared percentage of shares controlled by the five largest shareholders Control variables are firm size (SIZE), Profitability (ROA), current growth (GROW), growth opportunities (Q), financial leverage (DEBT) Dummy variables consist of year dummies, DLOSS dummy disclose that ROA is negative, and finally KD dummies are business group dummies A logit regression was used to analyze data The result of this study is a negative relationship between ownership concentration and dividend payment Harada and Nguyen also found that firms with dominant shareholders reluctant to increase dividends even when profitability increases and tend to ignore dividend when investment opportunities improve However, when debt is high, they increase dividend

Mirzaei (2012) conducted a research “A survey on the relationship between ownership structure and dividend policy in Tehran stock exchange” The main purpose

in this article is to study the relationship between Ownership Structure and Dividend Policy of Companies Listed at Tehran Stock Exchange The ownership structure as independent variables were divided to 4 categories: INSO: institution ownership, JO: joint ownership, MO: managerial ownership and OC: ownership concentration Also the effect of profit growth rate (G) and dividend policy variables (Di-1) of the previous year was controlled and multiple linear regressions were used

The model was used in this study is the multiple regression patterns which are commonly show as follow:

If the presupposition H0 disapproved, H1 will be accepted This means that, there

is a meaningful relationship between dependent and independent variables being tested

The findings of studying 88 companies during the time period between 2004 and

2009 showed that Joint ownership affects the ratio of dividends of firms accepted in Tehran Stock Exchange positively and Institution ownership affects it negatively However, there are some reasons which show there does not meaningful relationship between Management Ownership and the amount of Ownership Concentration with dividend policy

Thanatawee (2013) worked on the research “Ownership Structure and Dividend Policy: Evidence from Thailand” This paper examines the relationship between ownership structure and dividend policy in Thailand in a sample of 1,927 observations over the period 2002-2010 To examine the relation between ownership structure and dividend policy, the following regression is estimated: