Debt tax shield and firm value empirical evidence from listed companies in vietnam

1 MM proposition I: Firm’s capital structure is not effect to the value of the firm 2 MM proposition II: The cost of equity rises with leverage.. On the other hand, at the view of having

UNIVERSITY OF ECONOMICS ERASMUS UNVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES

VIETNAM – THE NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

DEBT TAX SHIELD AND FIRM VALUE: EMPIRICAL EVIDENCE FROM LISTED

BY

NGUYEN THI HONG HOA

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, OCTOBER 2017

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

HO CHI MINH CITY THE HAGUE

VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

DEBT TAX SHIELD AND FIRM VALUE: EMPIRICAL EVIDENCE FROM LISTED

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

Finally, thanks are also due to my classmates for providing me with unfailing support and continuous encouragement throughout my years of study and through the process of researching and writing this thesis This accomplishment would not have been possible without them Thank you

Nguyen Thi Hong Hoa

Ho Chi Minh City, October 2017

ABSTRACT

In the present study, panel data in fiscal year from 2008 to 2015 has been collected to reveal the interaction between debt tax shield and firm value The main purpose is to examine the value of debt tax shield and its effect on firm value toward taxation The reverse approach is employed in which the future profitability is regressed on firm value and debt using non-linear least square The advantage of reverse method is to shift measurement bias in future operating income to the regression residual and to enhance the usefulness of market factors to control for risk and expected growth This way also includes nontax information in the market value variable As a result, debt tax shield has negative effect on firm value The predicted value for debt tax shield approximately gets 37 percent of debt or gets 9.5 percent of firm value

TABLE OF CONTENT

Acknowledgement .i

Abstract ii

Table of content iii

List of tables v

List of figures .vi

1 Introduction 1

1.1 Research problem 1

1.2 Research objective 2

1.3 Scope of study 2

1.4 Thesis structure 3

2 Literature review 4

2.1 Theoretical review 4

2.1.1 Modigliani and Miller and capital structure theory (MM Model) 4

2.1.2 Trade-off theory 5

2.1.3 Theory of Agency costs 9

2.2 Empirical review 11

2.3 Hypothesis development 18

3 Research methodology 21

3.1 Conceptual framework 21

3.2 Estimation method 22

3.3 Variables and measures 29

3.4 Data Collection 36

4 Empirical result and discussions 37

4.1 The statistic descriptions of variables 37

4.2 Empirical result 41

4.2.1 Linear estimation 41

4.2.2 Nonlinear estimation 46

5 Conclusion 55

Reference vii

Appendix .xi

LIST OF TABLES

Table 3.1 Variable description 36

Table 4.1 Descriptive statistics 37

Table 4.2 Correlation 39

Table 4.3 Correlation (divided by total assets) 40

Table 4.4 Summary statistics from linear regression explaining the Value of firm (un-deflated intercept) 42

Table 4.5 Summary statistics from linear regression explaining the Value of firm with deflated intercept 43

Table 4.6 Valuation of debt tax shield ( ) from reverse regression, No Control for Capitalization Rates 44

Table 4.7 Valuation of debt tax shield ( ) from quantile regression according to industry effect 45

Table 4.8 Valuation of the debt tax shield ( ) from nonlinear Regression 49

Table 4.9 Summary statistics from Nonlinear regression with interest expense 50

Table 4.10 State ownership and firm performance from nonlinear Regression 51

LIST OF FIGURES

Figure 2.1 The optimal capital structure and the value of the firm 8 Figure 3.1 Conceptual framework 22 Figure 4.1 Distribution of future operating income 38

CHAPTER 1: INTRODUCTION

1.1 Problem statement

In corporate finance’s perspective, one of the most important decisions of a particular firm is to determine the optimal level of its capital structure or financial leverage However, the issue of firm’s capital structure has been controversially argued among researchers (Akhtar & Oliver, 2009) In addition, financial leverage has become more important since there are a large number of corporations using debt as a main instrument

to raise its capital

The relationship between taxation and capital structure has been empirically examined from a large number of developed countries such as the U.S and European countries with many institutional similarities It is necessary for a research about enterprise taxation influences to operating income in Asian countries Vietnam context would be selected for analysis because Vietnam is an Asian developing country with a low income and fresh stock exchange compared to other economies in Asia

At the aim of maximizing benefit and minimizing risk, a firm will choose the suitable capital structure to balance the costs and the benefits Therefore, the notion of deciding the ratio between debt and equity is always concerned at high level It is believed that the tax policy affect the firm’s financing Indeed tax is an essential component in firm’s activities and affects firm’s debt policy basing on deduction from interest expense It seems like that the only channel for firms to obtain funds is through bank borrowing in Vietnam Discovering how big magnitude tax affects firm profitability to find out the relationship between firm value, debt and corporate tax By this investigation, it is hopeful that there is appropriate guidance for effective application of debt Thus, above research context creates two research questions:

(1) Does debt give impacts on performance of Vietnamese firms?

(2) How big does the magnitude of net debt tax shield affect firm value?

1.2 Research objective

This first purpose aims to value the magnitude of debt tax shield, besides that there is another tendency to test the effect of tax to debt ratio in the scope of this research of Vietnamese enterprise in the stock market Many researches build firm value as function

of debt and unrecognized measures of future operating income, yet this study is based on

an approach by regressing future operating income on firm value, debt and controlling for firm-level capitalization rates (Kemsley & Nissim, 2002) According to Kemsley and Nissim (2002) relying on reversed approach of future operating income, any unexpected result of profitability is collected to the regression residual without effecting on debt; simultaneously, the market value as independent variable hold nontax information from debt In addition, considering the market value as market-based variable is useful to control for the risk and expected growth by Kemsley and Nissim (2002) We use interest expense to investigate the magnitude of debt In case enterprises receive benefit from corporate tax of debt, it is expected that there will have useful measures from revealing this relationship It is essential to find out the limitation sourcing from debt to restrict this limitation of debt Therefore, the main objectives of this study are:

(1) Examining the impact of debt on firm performance

(2) Value the magnitude of net debt tax shield Giving some implications for Vietnamese firms to improve their performance

(3) Revealing the role of state ownership on firm performance

1.3 Scope of study

This study examines the effects of taxation on firm performance in the context of Vietnamese companies The firm data is collected from 262 companies in Ho Chi Minh Stock Exchange in fiscal year 2008 to 2015 on the following required variables: total assets, net operating assets, interest expense, debt, future operating income, total market value The firm performance in this research only focuses on financial performance The

panel dataset is collected from Orbit database and Ho Chi Minh Stock Exchange database and data from Vietstock

This study attempts to follow quantitative analysis by applying nonlinear least square regression model on the panel data of Vietnamese firms, which are listed in Ho Chi Minh Stock Exchange (HOSE) The panel data would be employed to review the operation of firm performance when putting tax across years At the aspect of econometric model, this study utilizes the nonlinear least square regression model to examine the value of debt tax shield relative to firm value

1.4 Thesis structure

The remaining of this study includes four chapters First, chapter 2 discusses the

theoretical and empirical literature related to taxation and its relationship with firm performance This section primarily introduces the definitions of key concepts in this study, main theories about taxation and debt and lists out main empirical findings of prominent studies on taxation relationship This background would be the basis to form

the conceptual framework utilized in this study Second, chapter 3 reveals the research

methodology including conceptual framework, estimation method and variable description to establish the econometric models based on the conceptual framework

Third, chapter 4 presents data and the descriptive statistics, regression results and discussions on the main findings of the study Finally, chapter 5 expresses the

conclusions, policy implementations based on the main findings In addition, this part also discusses the research limitations and future development of the topic

CHAPTER 2: LITERATURE REVIEW

This chapter comprises the theoretical review and empirical review First, literature review is the summary of common theories of capital structure analysis such as Modigliani and Miller theory, trade-off theory, and theory of agency costs Then, on the ground of on those above theories, empirical review give an overview of the contribution

of previous studies on the development of capital structure analysis

2.1 Theoretical literature review

Overall, there are many theories about firm’s capital structure This study mentions three theories: Modigliani and Miller theory, trade-off theory, and theory of agency costs Furthermore, key determinants of leverage would also be analyzed and summarized to fortify the methodology and variables used in this paper

2.1.1 Modigliani and Miller and capital structure theory (MM Model)

MM model is developed by Modigliani and Miller (1958) and is also named the modern

capital structure theory It is based on these suppositions: (1) asymmetric information, (2)

no transaction costs, (3) there is no firm income tax, (4) without personal income tax, (5) lending interest rate and borrowing interest rate are in consistent, (6) ability to have funding of each individuals or firms is equal, (7) no bankruptcy costs and no financial distress costs, (8) the whole return is shared for owners, there is no share of return for investment

In reality, the market is not similar to the assumptions; yet, two discoveries have the significant impact from the consequence of Modigliani and Miller research At first, hypothesis of without taxation, this is chief discovery and the introduction for the convenience of debt because of tax relief from interest expense Next, Modigliani and Miller theory with the finding of variability of cash flow induce the risk This helps to restrict the risk from mentioning cash flow in case of insolvency

In core, MM theory is stated in two main propositions:

(1) MM proposition I: Firm’s capital structure is not effect to the value of the firm (2) MM proposition II: The cost of equity rises with leverage

Each proposition is considered in turn into the hypothesis of tax and no tax In the view

of no tax, the value of levered firm and value of unlevered firm are equivalent as the formula: V U V L, where V U is value of firm without debt is, V L is value of firm using debt Facing the different leverage level, the value of firms does not change Thus, the change of capital structure does not make the gain to shareholders It means that the capital structure is irrelevant

On the other hand, at the view of having tax, firm value equals the value of unlevered firm plus debt tax shields:

Analyzing the impact of the changes of capital structure on the average cost of capital, the MM proposition II concludes that the weighted average cost of capital will decrease when the cost of equity increases with leverage Nevertheless, MM theory ignores the impact of some other costs such as financial distress costs These financial costs reduce the benefit of tax shield of firms increasing debt ratio Facing this situation the financial distress costs are higher than the tax shield; consequently, the optimal capital structure is broken down (the point that the value of firm is maximized in condition of the weighted average cost of capital minimized) At the point that the capital structure is higher than the optimal level, the weighted average cost of capital is rising and the value of firm is decreasing The result is that gain from tax shield does not have enough magnitude to compensate the financial distress cost Therefore, firm may be fragile and face the status

of insolvency

2.1.2 Trade-off theory

In the view of trade-off theory, capital structure is established on the relationship between debt and equity There is a concept of “tax-bankruptcy trade-off” that a firm relies on tax related gains from leverage to cover bankruptcy costs In addition, the trade-off theory proposes the idea that a company makes decision of capital structure by balancing the costs and benefits The advantage of debt finance is the debt tax shield while the disadvantage is the agency costs and costs of financial distress, such as bankruptcy costs Bankruptcy costs including direct costs (legal and administrative expenses) and indirect cost (cost incurred by a financial distressed firm and cost associated with process

to bankrupt or period of financial distress) When profitability decreases, expected bankruptcy cost increase There is a positive relationship between bankruptcy cost and volatile earnings, this leads to less leverage toward smaller less-diversified firms

In term of taxation, interest tax shield pushes the level of leverage higher In contrast, personal tax cost mitigates the rate of leverage The negative relationship between the marginal corporate tax saving and marginal personal tax cost affect the leverage level of a firm Fama and French (2000) express that tests of the DeAngelo and Masulis (1980) model clearly display optimal leverage based on non-debt tax shields, that is, less leverage toward higher level of non-debt tax shield Besides that, the model also discovers that higher expected payoff from the interest tax shields is paralleled with the more profitable firm; but is related to less volatile earnings

Adjustment (financing) cost - Financing cost include the transaction cost associated with new equity and cost originated from management’s superior information about the firm’s outlook and the value of its risky securities A firm will set target leverage if financing cost does not affect other factors In contrast, Myers (1984) bases on the assumption of asymmetric information problem and other financing cost that affect optimal leverage in trade-off model A firm would choose the level of leverage and dividend payout under their no adjustment-cost optimal values As a result, asymmetric information and financing cost enhance the role of trade-off theory about target leverage and dividend payout The status of less leverage and lower dividend payout is for firm with lower expected profits, higher expected investment and heavier volatility of net cash

flow The financing cost is not the main factor that affects the target ratio in the trade-off model

Figure 2.1 The optimal capital structure and the value of the firm

Source: Ross et al (2012), page540, chapter 16, figure 16.6: The static theory of capital structure: the optimal capital structure and the value of the firm

Figure 2.1 displays the trade off between tax savings from debt and financial distress costs V L denotes market value of firm There are three situations in that Figure The first

is the horizontal line extending from V U, this case of the firm value without capital structure (MM proposition I without tax) The second is presented by the upward slopping straight line (MM proposition I with tax ) The third is the case where market value rise to a maximum and fall beyond that point (illustrating in this discussion) If debt exceeds this point, financial distress cost increases V L* is the maximum value of a firm

and D * is the optimal amount of borrowing Comparing the third and the first case, the benefit from leverage is net of distress cost Examining the third case with the second case, there is loss in value from the appearance of financial distress (Ross et al., 2012) Firm’s assets mainly impact to costs of financial distress If most of assets in a firm are tangible assets, these assets can be easily sold without loss in value There is the

existence of incentive for this firm to reach high debt level In contrast, firms that have greater intangible assets such as employee talent or growth opportunities, there is no incentive to borrow The reason is that intangible assets could not be sold In other words, the tendency of borrowing more is favorable in firms with a lower risk of experiencing financial distress compared with firms with greater risk of financial distress Specifically, Harris and Raviv (1990) declare that “firms with higher liquidation value, e.g., those with tangible assets, will have more debt, will have higher yield debt, will be more likely

to default, but will have higher market value than similar firms with lower liquidation value” It means that the debt level is positively related to the liquidation in value Cost of distress is lower in the firm with higher liquidation Andrade and Kaplan (1998) show that hand costs of financial distress increase when leverage rises They also find out the magnitude of financial distress costs that get about 10 to 20 percent of firm value

Briefly, firm could maximize firm value by forming financial leverage to the point that the gain from tax shield on debt is offset by financial distress costs The trade-off theory accents the important of the target leverage Firms have target leverage and balance the leverage against target over time

2.1.3 Theory of agency costs

Relating to agency stories of a firm, we mention about the different criteria between the managers and owners In reality, the interest of managers and security holders are not always on the same direct, managers have a tendency to have bonus with the excess of cash earnings over profitable investment In every enterprise, the capital owners (shareholders) have relationship with managers Employees, especially managers, are hired by the owners in order that the managers have responsibility to manage the firm against receiving salary, other income There is the existence of the relationship between owners and managers (Jensen and Meckling, 1976) The firm manager is always interest

in maximizing their wealth In contrast, the owner concerns in the value of company From that, there is the derivative of big conflict between firm’s managers and firm’s owners The manager sometimes does not complete the management role in the company

In other words, manager does not care to the benefit of the owners (Eisenhardt, 1989)

The agency costs are sourced from the contradiction of interest between managers, shareholders and debtholders of the firm

Jensen and Meckling (1976) also mentions to two kinds of conflicts The conflict between owners and managers is the first The second is the conflict between managers and creditors The first appears due to the different viewpoint between management and ownership in modern market In this case, the owners could not control the managers and the managers have more opportunity to enhance their benefit, simultaneously this leads to the decrease in benefit of shareholders Richardson (1996) claims that the bigger contradiction is positively relates to free cash flow

The second conflict is in the case of company using debt At the aim of maximizing firm value and upgrading firm, the owners need money to invest in new project Creditors lend owners of a company on behalf of managers to maximize profit of both Owners also agree with managers to manage projects in the high level of risk due to the greater profits The owners would receive all profits in the case of successful firm operating but the creditors only receive the definite profit at the accrued interest In contrast, if the firm operates ineffectively, the creditor will loose from receiving the delayed principle and lost the whole lent amount In this case, lenders need gather members to monitor the project disbursement Costs derived from this situation are the agency costs that the owners have to pay for using loan Agency costs could be controlled by increasing the management role, internal audit, regulations and the right of management board

Jensen and Meckling (1976) reveal that debt is one of the main primary factors to reduce the conflict between managers and owners Using more debts disciplines the managers due to fixed interest payment and limits the free cash flow that useful for the managers Jensen (1986) shows that the managers could not use cash in their own purpose because of the obligation of paying interest payment The firm’s agency costs of free cash flow could be controlled by using financial leverage, dividends and managerial ownership to restrict unreasonable spending of managers (Jensen, 1986) Debt and dividends help control agency problem created by free cash flow by decreasing cash flow

to repay debt to avoid bankruptcy and by forcing manager to payout This model predicts

that firms commit a higher ratio of pre-interest earnings to debt payments and dividends toward more profitable assets

In line with agency cost theory, optimal capital structure is the case that cost of agency related debt balances its benefit Hypothesis of all things being equal, high debt level is primary choice in firm with lower risky investments Firms with high level of growth opportunities haves less debt by reason of the unbearable status of paying debt obstruct their reasonable investment in the future If companies often have substantial free cash flow at high level, it should limit the use of high debt to control management role and to close managers’ faulty And a high debt level is also accepted by creditors of firms with great ability of liquidation

2.2 Empirical literature review

In the view of trade-off theory, capital structure is established on the relationship between debt and equity There is a concept of “tax-bankruptcy trade-off” that a firm rely on tax related gains from leverage to cover bankruptcy costs In addition, the trade-off theory proposes the idea that a company makes decision of capital structure by balancing the costs and benefits The advantage of debt finance is the debt tax shield while the disadvantage is the agency costs and costs of financial distress, such as bankruptcy costs The Modigliani and Miller (1963) investigate that tax benefit contribute to boost the firm value in proportion to amount of debt used and diminish the cost of debt capital That is the MM debt tax shield hypothesis The pecking order theory implies that firms have priority in choosing of financing sources and follow the order of retain earnings, debt and new equity It is because asymmetric information has influences on the option of internal

or external financing and the choice of debt or equity issuing Myers (1977) infers that the trade off between the tax benefits of debt and cost of extra financial constraints and bankruptcy leads the rise in firm’s leverage Myers and Majluf (1984) find that managers declare risky security in case of overpricing However, this asymmetric information absorbed to the investors lead to the under-value of risky securities Therefore, managers leave profitable investment for new risky securities As a result, Myers gives conclusion that the order choice of a firm for the investment is retailed earnings, safe debt, risky debt

and equity at last Pecking order theory means the internal financing being more favorable

Both two theories are useful ways of recognizing the usage of debt in a firm The trade-off theory introduce the guidance for capital structure in long run The pecking order theory is suitable for firm policy in shorter period, tendency toward investment by external funds

Recently, there are more concerns about the role of debt tax shield Many researches reveal that tax and debt tax shield significantly impact to firm’s financial leverage such as MacKie-Mason (1990), Graham (1996, 1999), Gordon and Lee (2001), Xing (2015), Hovakimian et al (2001) Especially, the concrete investigation is that estimated value of debt tax shield captures approximately 10 percent of firm value (Graham, 2000), (Kemsley and Nissim, 2002) and (Vanbinsbergen, Graham & Yang, 2015) Thus, debt tax shield becomes the extreme significant device for financial criteria (Schepens, 2015) Debt tax shield is highly appraised by the impact of it to debt and the magnitude of debt tax shield also depends on tax policy of each country

Investigating how taxes affect to firm’s specifications is always the subject that appeals many researchers The prominent significant study belongs to Modigliani & Miller (1958) Next, a stream of researches are Graham (1996a, 2000, 2003), Myers (1984, 1998), Fama (2011), Princen (2012), MacKinlay (2015), DeAngelo and Masulis (1980), MacKie-Mason (1990) All of these researches reach the significant level of tax

to firm specifications Perceiving the role of corporate taxation and effective implementation are the great trial in economics (Fama, 2011)

Gordon and Lee (2001) analyse debt policy in response to tax incentives by US statistics of income balance sheet for 46 years from 1950 to 1995 on the whole to compare the debt policies of firms of different sizes, specifically, small firms face different tax rates than larger firms Comparing to smaller firms, larger firms have a tendency to finance 8 percent of their assets with debt This study contributes to the background of taxes leading large effect on debt

Another research uses data in the time period 2001-2007 for the sample of Belgian and French companies by difference approach to study the impact of taxation on capital structure with the significant evidence at the rate of average 2 percent to 7 percent that tax system promotes company to apply more debt than the existence of the equal tax treatment of debt and equity (Princen, 2012)

MacKinlay (2015) finds that in the high interest tax environments, tax rate pushes big effects on firms’ debt policy Tax effects not only firms’ leverage but the elements converged into debt as well Thus, statutory provisions on tax would have little impact on firms’ levevage policy

The process to analyse the debt tax shield mostly includes three essential steps First, the use of debt is encouraged from the debt tax shield in the MM prediction MacKie-Mason (1990), Trezevant (1992), and Graham (1996, 1999) have researches that concentrate on incremental financing decisions They conclude that high marginal tax rates encourage the application of debt Princen (2012) discovers that tax rate appeals firm in using debt larger than the existence of the equal tax treatment of debt and equity Next, the firm value financial statement data is chosen to be the main factor to test the debt tax shield (Graham, 2000) He predicts the mean corporate tax benefit of debt for a large sample of Compustate firms This is not a direct way to calculate the debt tax shield, yet Graham confirm that firms get a tax advantage from applying debt

The last inquiry of valuating the debt tax shield need to be mentioned is the direct market signal about debt tax shield Masulis (1980) induces the hypothesis of corporate debt tax shield effect and discovers the wealth redistribution effect as a result of using capital structure Besides that, Fama and French (1998) apply cross sections to analyse firm value on interest expense and on various controls for profitability They discover a strong negative relation between debt and firm value and conclude: "imperfect controls for profitability probably drive the negative relations between debt and value and prevent the regressions from saying anything about the tax benefits of debt" (p 839)

However, the above steps make an existence of confused status This status is the case that value of a firm’s operations is undetected, firm value could be correlated with debt together nontax dimensions These make the survey confounded

Particularly, at the view of pecking order theory, the debt-intensive firms have lower earnings growth conceivability than low debt-intensive firms According to this theory, the notion of using debt in profitable firms less than the unprofitable firms because of the decrease in the value of operations This decrease is caused by the financial distress costs

of debt, so debt is oustanding relating to the expected future operating profitability

The approach of analyzing firm value on debt has disadvantage The predictted value

of this approach could be distorted by unexpected bias of the operation value Besides that, this method could be bias due to the exclusion of market-based factors used to restrain for risks and expected growth

In a research of Liu, Nissim, and Thomas (2002) studying the valuation of equity by using multiples, Kemsley and Nissim (2002) shows that Liu, Nissim, and Thomas (2002) also discover the best measure of explaining current market value This measure is counted by the mean of consensus earnings for the next five years In addition, Liu, Nissim, and Thomas (2002) also displays that using earnings multiples for valuation is more accurate than using free cash flow multiples because the appearance for lower free cash flow in case of investing by growth trends

In order to implement the research, Kemsley and Nissim (2002) introduce an alternative approach rather than specifying firm value as a function of debt and imperfect measures of future operating profitability The author uses Compustat data to measure the market value and use the future operating income (FOI) as the average operating income for the five year that follow the current year in way of reversing the relationship They specify future operating profitability as a function of firm value, debt, and controls for firm level capitalization rates This method has two advantages First, the future operating profitability in the right hand side of the equation enables to exclude measurement error

to the regression residual that it does not correlate the debt coefficient; and helps to supervise nontax information out of debt that effect to future operating income Second,

getting market value of firm out from the left hand side of regression pushes chance to apply market based factors to supervise risk and expected growth

Additional analysis, it is indeed to mention the role of state in the performance of state owned firms; especially for country maintaining government involvement in business corporations Li et al (2009) found that ownership concentration and state ownership are negatively associated to the evidence of bankruptcy There is advantage of state ownership as the higher leverage, lower taxation, larger markets shares (Faccio, 2006) , but the author also showed the disadvantage of this kind of ownership as the high levels

of corruption, less clear system in the SOEs Borisova et al (2012) discovered that government ownership generally leads to a higher cost of debt, consistent with investment distortion fostered by state influence Since the government guarantee implicitly, default risk of the economic recession or firm distress are solved by the dominant effect of government The higher cost of interest of lenders will push financial distress higher The advantage and disadvantage influence to performance of state ownership enterprise is known as double-edged sword Huang et al (2011) revealed the positive relationship between state ownership and debt ratio The explanation of this positive interact belongs to three things First, on the background of the low probability

of bankruptcy, they have substantial advantages with respect to enter to debt market Secondly, to preserve control and dominance, representatives of state ownership have priority for high level of debt instead of shares Thirdly, firm with high level of state ownership has segregation between voting and cash flow rights, agency problem become more complicated between owners and managers The real owners of state owned shares are the citizens; in contrast the decision in firm is depended on government departments whose income is not determined by their control through firm performance There is no motivation to urge to the effective operation of a firm As consequence, state owned firms prefer high debt to diminish agency costs of equity Another advantage of state control is mentioned, while private firms must bear high cost on capital sources, state owned firms receive a preferential treatment of state owned banks with low cost of loan The prominence of SOEs is that banks support favorable loan for them because they might be

able to build up political connections with politicians to obtain lucrative contracts (Butler

et al., 2009) Lin and Tan (1999) also showed that SOE receive fund from state owned banks regardless of risks but they do not have to fully cover expenses from sales and income, the government often bears the losses of unprofitable SOEs This is also the claim for the prominence of debt usage in firms with high state ownership than others in the research of Poyry and Maury (2010) on 95 listed Russian firms from 2000 to 2004 From above preferential treatment, state ownership enterprises tend to use more debt than other ownership firm This means that the magnitude of debt tax shield in SOEs will

be larger than other kind of ownership The more debt usage is the more financial distress

is High leverage leads to an increase in hand costs of financial distress (Andrade and Kaplan, 1998) Ding et al (2007) concluded that theory agency originated from different concerns of many levels of agents conflict with each other because the government is both the regulator and the manager In adjusting the policies to deal with financial distress, private firms are faster than state own firms Almeida and Philippon (2007) used credit spreads to value distress costs and concluded that there is the apparent reluctance

of firms to have higher their leverage, in spite of the benefit from debt tax shield

Before starting the study in details, we all together review the context about Vietnam economy as followings In 1986, Vietnam reforms the economy as known as “DOIMOI” policy that creates a high jump in VietNam economy by change from a centrally planned economy to a market oriented economy During liberalization process, a State-Owned Enterprises (SOEs) reform program is implemented in 1992 at the purpose of activating the private ownership policy Equitization is the main core of reform program Specifically, the regulation shows that the ownership over 50 percent shares of government is known as state-owned company Sectors as airlines, electricity and telecommunication are still restricted by status of government control The next step in the liberalization process stock market has been established The first stock exhange named Ho Chi Minh Stock Exchange (HOSE) has offically operated on 20 July 2000 and the next stock exchange named the Hanoi Stock Exchange (HNX) has openned on 8 March 2005 The operation of stock exchange establishes the active tendency for the

developing of the economy and creates a high jump for firms in capital mobilization, especially for medium and small enterprises From that, small and medium enterprise plays an important role in the economy

The application of private ownership stimulates the market more competitive and effective If the enterprise operates effectively, the owner has profit In contrast, the uneffective operation of firms leads owner being loss Maximize the value of firm and minimizes the cost of capital are the main purpose of a firm There are bodies of researches to find out the best way for purpose of maximizing profit The notion of using debt is one of approaches in financial decision In Vietnam, some authors study the firm performance and internal factors such as Nguyen and Nguyen (2012), Tran and Neelakantan (2006), Tran and Tran (2008) and Nguyen et al (2012) Tran and Neelakantan (2006) study the capital structure in small and medium size enterprise in the period 1998-2001 This research concludes that short term liabilities are priority in firm financial decision Tran and Tran (2008) research on capital structure and firm performance listed on Ho Chi Minh stock exchange with 50 non-financial listed firms The result is that the firm performance has the negative relationship with leverage when D/E ratio was more than 1.812 Thus firm performance has the positive relationship with leverage Firm should increase the debt ratio to raise the firm value Another research on the effect of capital structure on 92 Vietnam’s seafood processing enterprises from 2005

to 2010, Nguyen and Nguyen (2012) result that firm value and financial leverage have a non-linear relationship that are consistent with the study of Chou and Lee (2010) Specifically, the Vietnam’s Seafood processing firm would be nearly optimal against the debt ratio less than 59.27 percent Nguyen et al (2012) find that there is negative relationship between leverage and profitability It is recognized that there is limitation of the study about Vietnam capital structure and elements relating to financial decisions Indentifying the benefit of company’s capital structure and suitable implementation have become more and more important in financial economics Specially, the benefit from the deductibility of interest expense in taxing, in other words debt tax shield is always a remarkable problem Interest tax shield pushes the level of leverage higher In

contrast, personal tax cost mitigates the rate of leverage The negative relationship between the marginal corporate tax saving and marginal personal tax cost affect the leverage level of a firm Furthermore, Princen (2012) discovers that tax system promotes company to apply more debt than the existence of the equal tax treatment of debt and equity Considering the effect of corporate taxes, this is one of the great alternatives to form capital structure This point of view is sourced by the benefit of the tax deductibility known as debt tax shield The net tax saving from an additional dong in interest is offset

by the increase in expected financial distress costs that create an optimal capital structure

A firm will have a limit to the use of debt financing to generate tax shields in case they

do not apply a great amount of debt, but pay substantial taxes In brief, the tax benefit from leverage plays an important role to firms that are in a tax paying position Loss firms will receive a little benefit from the interest tax shield Similarly, firms that use substantial tax shields will gain less benefit from leverage A tax system is considered in the case that a firm is taxed on its end of period business excluding interest expenses Tax system is counted on a constant marginal tax rate Companies are taxed on their profits The costs as the interest paid could be considered as return to the creditors These interest expenses reduce taxable income, they are said to be tax deductible Briefly, an interest tax shield may encourage a company to finance a project through debt because dividend paid

on stock issues are never deductible The greater the tax rate is applied, the bigger the tendency to borrow is Vietnamese corporate tax system have traditional tax system, there

is the tax deductibility of interest expenses but no deductibility of the capital cost of equity The Vietnamese corporate tax rate was 28 percent in 2007 and in 2008, 25 percent from 2009 to 2013, 22 percent in both year 2014 and 2015

2.3 Hypothesis development

Forthcoming from literature, I therefore believe that debt tax shield truly has significant effects on Vietnamese firm value Specifically, the following hypothesis about the role of debt tax shield can be developed

Hypothesis 1: Debt tax shield has impacts on value of Vietnamese firms

As literature review, the existence of negative reaction of net debt tax shield to debt and firm value is revealed by a study of Fama and French (1998) A strong negative relationship between debt and firm value gives no conclusion about the tax benefit of debt

This notion is expressed that because the high level of debt lead to high value of debt tax shield this will give negative impact on the performance of that particular firm As literature review, the existence of negative reaction of net debt tax shield to debt and firm value is revealed by a study of Fama and French (1998) A strong negative relationship between debt and firm value gives no conclusion about the tax benefit of debt In a research of Tran and Tran (2008) on capital structure and firm performance listed on Ho Chi Minh stock exchange with 50 non-financial listed firms, firm performance and leverage has the negative relationship when D/E ratio was more than 1.812

The negative impact of debt to firm value is not finite in the sense that debt is not chosen

in financial strategy of any firms At the optimist aspect as many researches, debt creates the benefit by debt tax shield that is better for firms applying leverage This relationship

is assumed to hold to a certain degree By way of using too high debt, the firm will face the unbearable status to reimburse the interest expenses

Hypothesis 2: Interest is expected to effect on firm performance at the same direction as debt impact on firm performance

Fama and French (1998) applied cross sections and reveal the strong negative relationship between debt and firm value by using interest expense In recent study, Kemsley and Nissim (2002) discover the positive impact of interest on firm value Using the approach of Kemsley and Nissim (2002) based on FF study, I expect to find same direction of interest on firm value as debt reaction to firm value

Hypothesis 3: State ownership is expected to moderate firm performance

Faccio (2006) revealed the advantage of state ownership as the higher leverage, lower taxation and larger market share Boubakri and Saffar (2017) survey a large hand

collected sample of privatized firms from 62 countries about the link between residual state ownership and bank debt There is a positively association between state ownership and bank debt financing In fact, State support firms with soft budget constraint to be financed at preferential rate Thus, there is the tendency of using more debt in state ownership than other ownership firm It exists the high magnitude of debt tax shield in SOEs Moreover, Yu (2013) concluded that special treatment originated from strong political connection with state may help firms improve their performance when state ownership increase level of concentration Phung & Mishra (2016) revealled that advantage of firms with state ownership, especially high state ownership concentration firms, seems more than other type ownership firms sourced from the connection between political policy and high state concentration firms The advantage of high state concentration can help firm improve firm performance I wait for the positive result of government involvement to firms performance compared to no control of state

In order to accelerate reforms of tax administration and to deepen the competitiveness of domestic enterprises, a tax system reform strategy for 2010- 2020 period is ratified by Prime Minister Constraint interest expense is one of concerns that Vietnamese Finance Ministry is considering Finance Ministry also reveals that dominated interest expense will diminish ineffective operating of firm with high debt level Besides that, controlling interest expense is corresponding with Uniform rule Many developed country such as Germany, Australia, Holland has the debt to equity ratio does not exceed 3 times (3:1), the remainders as France, United States, Canada have smaller ratios about 1.5 to 2 times The Organization for Economic Co-operation and Development (OECD) also recommends that this rate be capped at three times Therefore, firm value and capital structure benchmark of debt tax shield are priority to be investigated Debt tax shield empirical studies in Vietnam become important and should be researched to identify effects

CHAPTER 3: RESEARCH METHODOLOGY

Research methodology provides an analytical framework for this study through the discussion of several important points First, the conceptual framework shows us the description of how industry effect, market value of operation, and capitalization rate give impacts on future operating income of firm Second, the appropriate estimation method is chosen to meet the objective of this study Finally, from the understanding of empirical model, the collected financial factors of Vietnamese firms are described in the variable description

is determined by subcomponents: market value of operations in parallel with the capitalization rate for average operating income (by vector of variable X ), industry

effect and direct relation between FOI and variables in X (by additive form as y X, ) Wherein, vector X include the following components:

(1) U is the industry median beta of operations and is proxy of risk U is omitted in

the direct form of direct relation between FOI and variable X because it is spanned by the industry dummy variable

(2) V L D

NOA





is defined as the market to book ratio of operation, and is calculated by

dividing market value of operation by net operating assets ( NOA )

(3) size is proxy of risk and growth, and is calculated by taking logarithm of NOA

(4) operating liabilities ( OL ) is known as all nondebt liabilities

This thesis concentrate on MM postulate I (with corporate tax): the value of levered firm

Note:  is the estimated value of the net debt tax shield that shows the degree of debt tax shield effecting to firm value

Figure 3.1 Conceptual framework 3.2 Estimation method

Based on the MM theory with tax, financial distress costs from debt do not exist The valuation model is affected by tax (Modigliani & Miller, 1958) as following:

Market value of operations

Industry effect ( )

is linear function ofector including:

 The industry median beta of operations ( )

 Market to book ratio of operation

 Size is logarithm of

 Operating liabilities ( )

State control ( )

( )

U

E FOI V



where,  is the capitalization rate for E FOI( ) which increases in risk E FOI( ) is defined as all expect future operating income and is perceived as influencing expected growth in income Therefore, the fixed E FOI( ) is excluded from proxy for expect future growth, due to this capitalization rate is used as proxy for growth “ ” has two effects as

the increase in risk but the decrease in expected growth)

There are some different affect to V U and T due to nontax cost of debt, personal

taxes and non-debt tax shields, depreciation Weiss (1990) and Andrade and Kaplan (1998) reveal that direct costs of bankruptcy slightly effect firm value From this view of point, V U is not only considered value of unlevered firm but also is perceived as the value

of operation Substituting equation (3.2) into equation (3.1), we have

Forward regression as study of Kemsley and Nissim (2002):

where, 3 is the estimated value of debt tax shield

The fact that debt is correlated with the value of operation (E FOI( ) and ‘ ’) together with several nontax dimensions, growth, financial distress and size dimension

Therefore, 3 would be bias Accurately, market value as dependent variable prevents the application of the market to book ratio as a control for  that Fama and French (1992) find the variant of market to book ratio as common proxies for risk Besides that, Penman (1996) also reveals that market to book ratio of operation is similar to expectation about future operating earnings relating to current book value being the factor for estimating growth in operating earnings

profit from one Vietnam Dong of debt, 3

 This equation implies the inverse

relationship between tax benefit from debt and expected operating income to adjust the market value of the firm Besides its own limitation, this equation restricts some drawbacks of the forward approach Firstly, the bias and inconsistent estimates of the value of the net debt tax is sourced from the measurement error in empirical proxies for

E FOI The regression residual includes any random factors of above errors Secondly, market value includes all signals relating to E FOI( )such as information about growth prospects, financial distress costs, management role, volatility, sourcing from relation between profitability and debt from capital structure This leads to the main function of market value from handling ‘ ’

Hereby, the cross-sectional regressions are applied for following estimation

Linear empirical specifications:

The combination of equation (3.5) and (3.6), we have a linear empirical specification Kemsley and Nissim (2002) base on the method of Fama and French (FF method), all

regression variables are deflated by total asset (excluding the intercept is not deflated by total asset):

1

L V

analogousness of equation (3.8) and (3.9) is the entire equation deflated by total assets

The discrepancy of equation (3.9) to (3.8) is the reverse FOI for V L By applying the

reverse approach, any accidental error in FOI is moved to the regression residuals;

simultaneously, V L on the right hand side of the equation relating to any derivable tax information from debt composes to future operating earnings that could be managed Specifically, size effect composed to market value shall distort the value of beta in the equation (3.8)

non-The special point being mentioned is that the deflated or un-deflated intercept is available for the prediction If the equation (3.9) is with the un-deflated intercept

(equivalent to TA

TA ), the entire equation would be similar to be adding the deflator total assets (Miller & Modigliani, 1966) Nevertheless, total assets are different from other explanatory variables, it includes the past operating income and net investment; this creates the correlation with the measurement error in FOI that leads to distortions of

estimation Specifically, FOI is the future operating income including of imperfect

criteria constructing expected earnings with error These errors could origin from the growth trends and expected future financial distress costs These measurement errors in

FOI not only relate to most of explanatory variables but also distort all regression

coefficients Exactly, in the nature of efficiency market, the market value of firm is not influenced by accounting errors, and the book value of debt frequently represent the

market value of debt The distortion in measurement of FOI rarely affect the

measurement of debt due to the structure of debt is determined across firms The equation (3.9) specified in the deflated intercept, the uncertainty of accounting measurement in

FOI should be orthogonal to explanatory variables In empirical study, Kemsley and

Nissim (2002) declared that the un-deflated case pushes the value of debt tax shield higher than the result of study in 2002 because of derivative unreasonable levels Hence, the deflated intercept specification should be priority

At the aspect of role of the capitalization rate for future operating income,  is not actually a cross sectional constant We specify capitalization rate for future operating income into following factors such as: industry median beta of operation (U) with the

role to control risk; the market to book ratio of operation V L D

specification and firm specific specification will affect the status of firm in different way, thus financial decision of each firm depend on these specifications Showing the effect of industry sectors brings out the magnitude of capitalization rate margin due to enhancing the within-industry variation of variables (Kemsley & Nissim, 2002), as shown as followings:

(explaining these variables into additive form as y X, ), Kemsley and Nissim (2002) show below equation:

1 1

By replacing all variables for X into (3.9), there is the empirical specification hereunder (remark, U is omitted in the structure of y X, since it is moved to the industry sector beta variable) as a nonlinear regression (Kemsley & Nissim, 2002):

Perspective about the deductibility of corporate interest stimulate using debt is discovered

by Modigliani and Miller (1963) DeAngelo and Masulis (1980) contribute the great discovery about optimal capital structure built on the exhausting interest tax shield One more concrete signal about the correlation of interest and firm value is showned in the research of Fama and French (1998) by using method of interest expense and various determinants for profitability

Kemsley and Nissim (2002) rely on this equation (3.10) and apply method of FF to describe the role of interest expense to forecast the value of the net debt tax shield

A nonlinear least square is also applied to forecast a nonlinear in parameters.The research

of Pao (2008) discusses that the probability of catching sophisticated influence in better fit and forecast in panel data are the advantage of the nonlinear model operating compared to the regression model

3.3 Variables and measures

3.3.1 Dependent variable

This research investigates the relationship between the firm value and magnitude tax effects, and the dependent variable would be considered as firm’s operating income There are many measures of firm performance from literature as implied in empirical review

This study applies the research of Kemsley and Nissim (2002) that follow FF and Miller and Modigliani (1966) by using earnings represents for FOI Operating income is modified the earnings before interest, tax and abnormal items for firm i in year t The future operating income (FOI) is mentioned as the average operating income for the two years that follow the current year in way of reversing the relationship The future operating profitability is specified as a function of firm value, debt, and controls for firm level, capitalization rates Baker and Wurgler (2002) reveal that historical market value is strongly related to capital structure that effect to operating income

FOI is counted by the average operating income for the two years after the current year Measure of relying on two years of future income attain the growth trends in operating earnings better than seperate operating income in each year This approach shows all expected future financial distress costs larger than only one year for choice as well The method of using two-year operating income excludes firms facing bankruptcy ahead Yet applying two-year operating income have some bias should be solved by

robustness test (Kemsley & Nissim, 2002) Operating Income consists of income and after tax interest expense (net income + interest expense*(1T C)) such as MM definition

of operating income equal to EBIT*(1T C), T C is tax rate We estimate value of debt tax shield by panel regression, which T C being as a cross sectional constant Liu, Nissim, and Thomas (2002) also mention that assessment with earnings multiples reach more specific estimate of firm value than valuation with free cash flow multiples because application of free cash flow could be absorbed by investment Thus, the applying measure of FF and Miller and Modigliani is the effective way

3.3.2 Controlling variables

There are components that may also influence firm performance according to previous theoretical and empirical literature These components would be considered as control variables and be included to the empirical model Below are all the controlling variables employed in this study:

(1) Industry sector beta (DIND)

The sensitivity of share price to movement in the market price is measured by the definition of Beta coefficient It always exists the risk inherent in the whole financial system, beta coefficient is discovered to measure this systematic risk In capital asset pricing model, Beta coefficient plays an important role to calculate required rate of return

on a stock It is the slope of the security market line

Beta coefficient for each industry is calculated as covariance of a stock's return with market returns of each industry divided by variance of market return Another explanation, beta coefficient equals correlation coefficient multiplied by standard deviation of stock returns divided by standard deviation of market returns (Wikipedia) Generally, it quantifies the magnitude that a share price of a company reacts against the market A beta coefficient of one industry less than one, the share is more stable In contrast, the beta is in excess of one, the share is exaggerating the market's movements Sometimes, a beta of one explains share of that company and the markets are in the same

line Irregularly, there has a negative beta (e.g a case of gold-mining industry), the explanation is that the share price moves in the opposite direction to the broader market Brander and Lewis (1986) discover that industry specific factors interact to financial decisions Firms in the same industry type are interacted by the definite substance of this industry sector Firms in different industry specific standard volatile are in indefinite manner Thus, industry sector is indeed to be regarded for research In addition, Hovakimian et al (2001) investigate that firms leverage is rebalanced by the mean of industry value over time

In this study, the industry sector median beta of operation (DIND) acts as the role to control risk; Sector beta for each industry is counted by the use of monthly stock returns

of each industry for each year DIND slope Ri Rm ,   All the current 262 non-financial listed companies in Ho Chi Minh stock exchange are utilized to construct the risk-return framework and attempt to define the explanatory factors of expected returns for Vietnam market The data relate to monthly stock returns Accounting information is collected through some financial websites such as vietstock.vn, cophieu68.com The perspective of industry type is determined for this research The classification of industry types is as follows: Agriculture, construction and material, construction, food and drink, healthcare, household, information & communication technologies, logistics, machinery, mining and oil, plastics, real estate, retails, rubber production, seafood, utility and wholesales The bank and financial institution are excluded in this survey

(2) Unlevered market beta (U)

The study of Brander and Lewis (1986) reveal that firm specific factor affect the financial decision in different way of each firm Basing on this investigation, studies relating to any nexus to firm value essentially add firm specific sector (market beta of each firm) in research

Similar to industry sector beta, market beta coefficient of each company is counted by the same measure as above mentioned Market beta is counted by the use of monthly

stock returns of each year beta slope Ri Rm ,  , it is applied from some financial websites as vietstock.vn After that, unlevered market beta is counted by the formula that Copeland and Weston (1988) modify market beta for leverage by the equation hereinafter:

Mackie-Mason (1990) and Graham (1996) reveal that status of high marginal tax rate urge firm to issue more debt than status of low marginal tax rate Firm value is independent with using more or less debt Precisely, Miller (1977) displays that there is

no relation between debt and firm value, the tendency of using debt is preferred to get the balance of interest deduction and taxes

(4) Operating liability (OL) and net operating asset (NOA)

Kemsley and Nissim (2002) determine that operating liability is measured by all liabilities subtract debt and net operating asset is defined as the total assets subtract operating liabilities Operating liabilities (OL) is equivalent to nondebt liabilities